Insertion of molecular oxygen into a palladium(II) methyl bond: A radical chain mechanism involving palladium(III) intermediates

Article abstract of DOI:10.1021/ja9061932

The reaction of (bipy)PdMe2 (1) (bipy ) 2,2′-bipyridine) with molecular oxygen results in the formation of the palladium(II) methylperoxide complex (bipy)PdMe(OOMe) (2). The identity of the product 2 has been confirmed by independent synthesis. Results of

Full text of DOI:10.1021/ja9061932

Published on Web 10/14/2009
Insertion of Molecular Oxygen into a Palladium(II) Methyl
Bond: A Radical Chain Mechanism Involving Palladium(III)
Intermediates
Luc Boisvert, Melanie C. Denney, Susan Kloek Hanson, and Karen I. Goldberg*
Department of Chemistry, UniVersity of Washington, Box 351700,
Seattle, Washington, 98195-1700
Received July 23, 2009; E-mail: goldberg@chem.washington.edu
Abstract: The reaction of (bipy)PdMe (1) (bipy ) 2,2′-bipyridine) with molecular oxygen results in the formation
2
of the palladium(II) methylperoxide complex (bipy)PdMe(OOMe) (2). The identity of the product 2 has been
confirmed by independent synthesis. Results of kinetic studies of this unprecedented oxygen insertion reaction
into a palladium alkyl bond support the involvement of a radical chain mechanism. Reproducible rates, attained
in the presence of the radical initiator 2,2′-azobis(2-methylpropionitrile) (AIBN), reveal that the reaction is overall
first-order (one-half-order in both [1] and [AIBN], and zero-order in [O
implies that the reaction proceeds by a mechanism that differs significantly from those for organic autoxidations
and for the recently reported examples of insertion of O into Pd(II) hydride bonds. The mechanism for the
2
]). The unusual rate law (half-order in [1])
2
autoxidation of 1 is more closely related to that found for the autoxidation of main group and early transition
metal alkyl complexes. Notably, the chain propagation is proposed to proceed via a stepwise associative homolytic
substitution at the Pd center of 1 with formation of a pentacoordinate Pd(III) intermediate.
5
9
Introduction
oxygenase” catalysis) remain scarce. One possible mechanism
for the incorporation of molecular oxygen into an organic substrate
The need to develop more efficient, cost-effective, and
environmentally benign processes for the selective oxidation
of organic compounds has led to a growing interest in the
reactivity of organometallic compounds toward molecular
2
involves the insertion of O into a palladium alkyl bond. However,
such reactivity involving oxygen has not been reported for
palladium alkyl complexes and, in fact, there are few documented
examples of O insertions into late transition metal-alkyl bonds.
The known examples are almost exclusively limited to Co(III) alkyl
1,2
oxygen. From both environmental and economic perspectives,
oxygen represents an ideal oxidant. Particularly promising for
the discovery of new processes is the wide variety of reactivity
2
10
complexes of macrocyclic nitrogen ligands and to two recent
11-13
reports involving Pt(II)-Me complexes.
reactions of alkyl complexes of main group and early transition
metals with O to yield metal alkyl peroxides or metal alkoxides
In contrast, the
2
patterns that have been observed in reactions of O with
3
transition metals. However, the current understanding of the
mechanisms by which organometallic complexes react with
molecular oxygen is still limited, hindering the widespread use
of this oxidant in homogeneous catalytic reactions.
Among the late transition metals, palladium has attracted
particular attention for its ability to catalyze numerous homo-
2
14-16
are well known.
The usefulness of this transformation can be
illustrated by its involvement in the aluminum-based “Ziegler
alcohol” processes for the industrial production of fatty alcohols
as well as in a variety of synthetic contexts such as the synthesis
17
2,4
geneous aerobic oxidation reactions. Efficient processes have
been developed with “palladium oxidase” type systems, where
5
(9) Most known systems that allow the palladium-catalyzed incorporation
of oxygen atoms from O
it is often proposed that O
2
into organic substrates are ill-defined, and
itself is not the actual oxidant. For the
molecular oxygen acts as an oxidant for the Pd center (or for a
2
6
cocatalyst), as exemplified by the aerobic oxidation of
oxidation of alkenes, see: ref 8a and (a) Muzart, J. J. Mol. Catal. A:
Chem. 2007, 276, 62. For the oxidation of benzene to phenol, see: (b)
L u¨ cke, B.; Narayana, K. V.; J a¨ hnisch, M. K. AdV. Synth. Catal. 2004,
2
,4-7
8
alcohols
In contrast, palladium-catalyzed transformations where oxygen
atoms from O are incorporated into the substrate (“palladium
and by the Wacker process.
3
46, 1407. See also: (c) Zhang, J.; Khaskin, E.; Anderson, N. P.;
2
Zavalij, P. Y.; Vedernikov, A. N. Chem. Commun. 2008, 3625.
10) (a) Chavez, F. A.; Mascharak, P. K. Acc. Chem. Res. 2000, 33, 539.
(b) Gupta, B. D.; Kanth, V. V.; Singh, V. J. Organomet. Chem. 1998,
570, 1. (c) Bianchini, C.; Zoellner, R. W. AdV. Inorg. Chem. 1997,
44, 263. For catalytic reactions, see: (d) Tokuyasu, T.; Kunikawa, S.;
Masuyama, A.; Nojima, M. Org. Lett. 2002, 4, 3595.
(
(
(
(
(
1) Punniyamurthy, T.; Velusamy, S.; Iqbal, J. Chem. ReV. 2005, 105, 2329.
2) Stahl, S. S. Science 2005, 309, 1824.
3) Bakac, A. Coord. Chem. ReV. 2006, 250, 2046.
4) (a) Gligorich, K. M.; Sigman, M. S. Chem. Commun. 2009, 3854. (b)
Nishimura, T.; Uemura, S. Synlett 2004, 201. (c) Stoltz, B. M. Chem.
Lett. 2004, 33, 362.
(11) Grice, K. A.; Goldberg, K. I. Organometallics 2009, 28, 953.
(12) Taylor, R. A.; Law, D. J.; Sunley, G. J.; White, A. J. P.; Britovsek,
G. J. P. Angew. Chem., Int. Ed. 2009, 48, 5900.
(5) Stahl, S. S. Angew. Chem., Int. Ed. 2004, 43, 3400.
(6) Piera, J.; B a¨ ckvall, J.-E. Angew. Chem., Int. Ed. 2008, 47, 3506.
(7) (a) Sigman, M. S.; Jensen, D. R. Acc. Chem. Res. 2006, 39, 221. (b)
Muzart, J. Tetrahedron 2003, 59, 5789. (c) Sheldon, R. A.; Arends,
I. W. C. E.; ten Brink, G.-J.; Dijksman, A. Acc. Chem. Res. 2002, 35,
(13) Me· radicals from the decomposition of a transient Ni(III)-Me
complex can be intercepted by O to form a short-lived Ni(III)-OOMe
2
species: Sauer, A.; Cohen, H.; Meyerstein, D. Inorg. Chem. 1988, 27,
4578.
7
74.
(14) Reviews: (a) Davies, A. G. J. Chem. Res. 2008, 361. (b) Alexandrov,
Y. A. J. Organomet. Chem. 1973, 55, 1. (c) Sosnovsky, G.; Brown,
J. H. Chem. ReV. 1966, 66, 529.
(
8) (a) Cornell, C. N.; Sigman, M. S. Inorg. Chem. 2007, 46, 1903. (b)
Takacs, J. M.; Jiang, X.-t. Curr. Org. Chem. 2003, 7, 369.
1
5802 9 J. AM. CHEM. SOC. 2009, 131, 15802–15814
10.1021/ja9061932 CCC: $40.75  2009 American Chemical Society

Insertion of O
2
into Pd(II)-Me
A R T I C L E S
of alcohols and alkyl hydroperoxides from main group organo-
nisms have been evaluated in relation to the experimentally
determined rate law and a viable mechanism for this reaction is
proposed. Important implications for the development of catalysts
for the functionalization of alkanes that could proceed via an
oxygen insertion reaction are also discussed.
18
metallics, the epoxidation of alkenes via in situ generated zinc
19
20
alkylperoxides, and the use of alkylboranes as radical initiators.
Autoxidation reactions of this type involving late transition metal
complexes could be particularly valuable due to the capacity of
late transition metals to generate metal alkyls by activation of C-H
21
Results
bonds. The coupling of these two transformations could be a
powerful combination for alkane functionalization.
Reaction of (bipy)PdMe
2
(1) with O
2
. When a solution of
Described herein is the first observation of the insertion of O
into a Pd(II)-Me bond, forming a Pd(II) alkylperoxide complex.
The reaction of (bipy)PdMe (1) with O to form (bipy)Pd-
Me(OOMe) (2) is a rare example of insertion of O into a late
transition metal-alkyl bond, and represents significant progress
toward the development of a “palladium oxygenase” paradigm.
The results of mechanistic studies suggest that the reaction proceeds
by a radical chain process. Several possible radical chain mecha-
2
(bipy)PdMe (1) in degassed C D
heated at 50 °C for 18 h, < 5% decomposition occurred
(measured by integration of signals in H NMR spectra against
an internal standard), although traces of Pd black were visible.
6 6
However, when a solution of 1 in C D was exposed to oxygen
(1-10 atm) at room temperature, the novel methylperoxide
species (bipy)PdMe(OOMe) (2) was formed in yields of up to
ca. 70% as determined by NMR spectroscopy (eq 1). Experi-
ments in which independently prepared (bipy)PdMe(OOMe) (2)
2
6 6
under reduced pressure was
1
2
2
2
2
2
(
15) For representative examples involving different metals, see the
following. Lithium: (a) Panek, E. J.; Whitesides, G. M. J. Am. Chem.
Soc. 1972, 94, 8768. Magnesium: (b) Walling, C.; Cioffari, A. J. Am.
Chem. Soc. 1970, 92, 6609. (c) Garst, J. F.; Smith, C. D.; Farrar, A. C.
J. Am. Chem. Soc. 1972, 94, 7707. Titanium, zirconium, hafnium: (d)
Blackburn, T. F.; Labinger, J. A.; Schwartz, J. Tetrahedron Lett. 1975,
(
see below) was added to solutions of the product formed in
the reaction between 1 and O unambiguously confirmed the
2
identity of the product as the methylperoxide derivative 2. It
1
was also found that the H NMR chemical shifts of 2 are very
similar to those of the methoxide derivative (bipy)PdMe(OMe)
1
6, 3041. (e) Lubben, T. V.; Wolczanski, P. T. J. Am. Chem. Soc.
(3). Notably, addition of 3 to the same product solutions clearly
1
987, 109, 424. (f) Brindley, P. B.; Scotton, M. J. J. Chem. Soc., Perkin
Trans. 2 1981, 419. (g) Ryan, D. A.; Espenson, J. H. J. Am. Chem.
Soc. 1982, 104, 704. (h) Hess, A.; H o¨ rz, M. R.; Liable-Sands, L. M.;
Lindner, D. C.; Rheingold, A. L.; Theopold, K. H. Angew. Chem.,
Int. Ed. 1999, 38, 166. (i) Atkinson, J. M.; Brindley, P. B. J.
Organomet. Chem. 1991, 411, 139. Tungsten: (j) Parkin, G.; Bercaw,
J. E. J. Am. Chem. Soc. 1989, 111, 391. (k) Chen, T.; Zhang, X.-H.;
Wang, C.; Chen, S.; Wu, Z.; Li, L.; Sorasaenee, K. R.; Diminnie,
J. B.; Pan, H.; Guzei, I. A.; Rheingold, A. L.; Wu, Y.-D.; Xue, Z.-L.
Organometallics 2005, 24, 1214. Iron: (l) Arasasingham, R. D.; Balch,
A. L.; Cornman, C. R.; Latos-Grazynski, L. J. Am. Chem. Soc. 1989,
demonstrated that 3 was not formed in the oxygenation reaction.
1
11, 4357. Cadmium: (m) Alexandrov, Y. A.; Lebedev, S. A.;
Independent Synthesis of (bipy)PdMe(OOMe) (2) and
Confirmation of Structure. The palladium(II) methylperoxide
complex 2 can be independently prepared by the reaction of the
palladium(II) methoxide complex 3 or of the hydroxide complex
Kuznetsova, N. V.; Razuvaev, G. A. J. Organomet. Chem. 1979, 177,
9
1
1. Boron: (n) Allies, P. G.; Brindley, P. B. J. Chem. Soc., B 1969,
126. (o) Davies, A. G.; Ingold, K. U.; Roberts, B. P.; Tudor, R.
J. Chem. Soc., B 1971, 698. (p) Korcek, S.; Watts, G. B.; Ingold,
K. U. J. Chem. Soc., Perkin Trans. 2 1972, 242. (q) Rensch, R.;
Friebolin, H. Chem. Ber. 1977, 110, 2189. Various metals: (r) Davies,
A. G.; Roberts, B. P. J. Chem. Soc., B 1968, 1074. (s) Brindley, P. B.;
Hodgson, J. C. J. Organomet. Chem. 1974, 65, 57. (t) M o¨ ller, M.;
Husemann, M.; Boche, G. J. Organomet. Chem. 2001, 624, 47.
16) Some metal alkylperoxide complexes resulting from the insertion of
(bipy)PdMe(OH) (4) with MeOOH (Scheme 1). Complexes 3 and
4
were prepared from the known iodide derivative (bipy)PdMe(I)
23
(5). The reaction of 5 with silver trifluoromethanesulfonate and
sodium methoxide in methanol cleanly generated the methoxide
(
1
complex 3 (Scheme 1). The H NMR data for 3 (in CD
3
OD)
O
2
into M-R bonds have been characterized crystallographically.
Magnesium: (a) Han, R.; Parkin, G. J. Am. Chem. Soc. 1992, 114,
48. (b) Bailey, P. J.; Coxall, R. A.; Dick, C. M.; Fabre, S.; Henderson,
matched that previously reported by van Koten and co-workers,
who prepared 3-d by dissolving (bipy)PdMe(OCH(CF ) in
OD. The new hydroxide complex 4 was obtained by addition
of H O to solutions of 3 in benzene (Scheme 1).
7
3 2
)
3
L. C.; Herber, C.; Liddle, S. T.; Lorono-Gonzalez, D.; Parkin, A.;
Parsons, S. Chem.sEur. J. 2003, 9, 4820. Zinc: (c) Lewinski, J.;
Suwala, K.; Kubisiak, M.; Ochal, Z.; Justyniak, I.; Lipkowski, J.
Angew. Chem., Int. Ed. 2008, 47, 7888. Aluminum: (d) Lewinski, J.;
Zachara, J.; Gos, P.; Grabska, E.; Kopec, T.; Madura, I.; Marciniak,
W.; Prowotorow, I. Chem.sEur. J. 2000, 6, 3215. Gallium: (e) Power,
M. B.; Cleaver, W. M.; Apblett, A. W.; Barron, A. R.; Ziller, J. W.
Polyhedron 1992, 11, 477. Indium: (f) Cleaver, W. M.; Barron, A. R.
J. Am. Chem. Soc. 1989, 111, 8966.
24
CD
3
2
Complex 2 could be efficiently synthesized by a variety of
routes involving the use of methyl hydroperoxide. Since
MeOOH is not commercially available, it was obtained by
methylation of hydrogen peroxide with dimethyl sulfate under
(
17) Noweck, K.; Grafahrend, W. Fatty alcohols. In Ullmann’s Encyclo-
pedia of Industrial Chemistry, 7th ed.; Wiley-VCH: New York, 2007
Scheme 1
(
electronic release).
(
18) (a) Walling, C.; Buckler, S. A. J. Am. Chem. Soc. 1955, 77, 6032. (b)
Lambert, G. J.; Duffley, R. P.; Dalzell, H. C.; Razdan, R. K. J. Org.
Chem. 1982, 47, 3350. (c) Brown, H. C.; Midland, M. Tetrahedron
1
987, 43, 4059. (d) Cadot, C.; Dalko, P. I.; Cossy, J.; Ollivier, C.;
Chuard, R.; Renaud, P. J. Org. Chem. 2002, 67, 7193. (e) Klement,
I.; L u¨ tjens, H.; Knochel, P. Tetrahedron 1997, 53, 9135. (f) Brown,
H. C.; Midland, M. M. Angew. Chem., Int. Ed. 1972, 11, 692.
19) (a) Yamamoto, K.; Yamamoto, N. Chem. Lett. 1989, 1149. For
asymmetric variants, see: (b) Enders, D.; Kramps, L.; Zhu, J.
Tetrahedron: Asymmetry 1998, 9, 3959. (c) Hussain, M. M.; Walsh,
P. J. Acc. Chem. Res. 2008, 41, 883.
(
(
(
20) Ollivier, C.; Renaud, P. Chem. ReV. 2001, 101, 3415.
21) (a) ActiVation and Functionalization of C-H Bonds; Goldberg, K. I.,
Goldman, A. S., Eds. ACS Symposium Series 885; American Chemical
Society: Washington, DC, 2004. (b) Handbook of C-H Transforma-
tions; Dyker, G., Ed.; Wiley-VCH: Weinheim, Germany, 2005.
J. AM. CHEM. SOC. 9 VOL. 131, NO. 43, 2009 15803

A R T I C L E S
Boisvert et al.
Table 1. Selected Values of H and 13C NMR Chemical Shifts for Complexes 1-4a
1
a
, 233 K, 500 MHz. 13C NMR chemical shifts determined by HMQC analysis.
26
2 2
Conditions: CD Cl
25
basic conditions and used as a ∼1: 2 MeOOH/Et
2
O solution.
P to
: immediate formation of Ph PO
chemical shifts are likely due to hydrogen bonding between 2/3
The identity of MeOOH was confirmed by addition of Ph
a sample of MeOOH in CDCl
3
and ROH. Notably, it has been reported that complexes of the type
3
1
3
3 2
(bipy)PdMe(OR) (OR ) OCH(CF ) , OAr) form strong hydrogen
bonds with ROH in solution and in the solid state. In fact, in
27
28
and MeOH was observed by H NMR spectroscopy.
Methylperoxide complex 2 was obtained and characterized
1
some preparations of complexes 2 and 3, analysis by H NMR
in situ (by NMR spectroscopy), together with one molar
2
spectroscopy showed the presence of some ROH (ROH ) H O,
equivalent of ROH (R ) Me or H), when MeOOH in Et
was added to a C or CD Cl solution of (bipy)PdMe(OMe)
3) or of (bipy)PdMe(OH) (4) (Scheme 1). Alternatively, when
2
O
MeOH, MeOOH), even after extensive washing of the solids.
However, addition of independently prepared samples of 2 and 3
to solutions of the product of the reaction of 1 with oxygen in
6
D
6
2
2
(
MeOOH was added to toluene solutions of either methoxide
complex 3 or hydroxide complex 4 followed by cooling of the
reaction mixture to -35 °C, complex 2 precipitated from
solution and was isolable as a yellow solid.
Even though complexes 2 and 3 have very similar structures,
differing only by one oxygen atom, it was possible to clearly show
C D
6 6
conclusively identified 2 as the product.
It was also observed in the course of characterization of the
products that the oxygenated Pd complexes 2, 3, and 4
decomposed in solution to Pd black over minutes to hours at
room temperature and were only slightly soluble in aromatic
solvents at low temperature. Therefore, although reactions
that the product of the reaction between 1 and O
2
was the
between 1 and O
2 6 6
to form 2 were conducted in C D at
methylperoxide complex 2. Initially, the analysis was complicated
temperatures ranging from 298 to 323 K (see below), detailed
1
1
13
by the fact that the H NMR chemical shifts recorded in C
D
6 6
of
analysis of the products by H and C NMR spectroscopies
was better carried out in CD Cl at low temperatures (ca. 233
K). Under these conditions (233 K in CD Cl ), no decomposition
complexes 2 and 3 vary with the presence of protic impurities.
Thus, simple matching of the chemical shift data from the spectra
2
2
2
2
1
of the reaction mixture recorded in C
D
6 6
to that of the indepen-
was observed and the H NMR chemical shifts of complexes 2
dently prepared complexes was ambiguous. These variations in
and 3 were only weakly affected by protic impurities.
1
The H NMR spectra (CD
2 2
Cl , 233 K) of complexes 2, 3, and
4
exhibit an unsymmetrical bipyridine environment and a singlet
(
22) For reports where the nature of the products formed from the reaction
between O and palladium complexes suggests that insertion of O
at 0.5-0.7 ppm for the Pd-Me groups (Table 1). The proton of
the hydroxide group of 4 can be seen as a broad singlet at around
-
proton was shown by NOESY experiments to be in chemical
exchange with traces of water in the sample. As seen in Table 1,
2
2
into a Pd-C bond may have been involved, see: (a) Sol e´ , D.; Solans,
X.; Font-Bardia, M. Dalton Trans. 2007, 4286. (b) Vicente, J.; Abad,
J.-A.; Frankland, A. D.; de Arellano, M. C. R. Chem.sEur. J. 1999,
29
1.4 ppm, as has been described for similar complexes. This
5
, 3066. (c) Muzart, J.; Riahi, A. Organometallics 1992, 11, 3478.
(
d) McCrindle, R.; Ferguson, G.; McAlees, A. J.; Arsenault, G. J.;
Gupta, A.; Jennings, M. C. Organometallics 1995, 14, 2741.
23) (a) Byers, P. K.; Canty, A. J. Organometallics 1990, 9, 210. (b) de
Graaf, W.; Boersma, J.; Smeets, W. J. J.; Spek, A. L.; van Koten, G.
Organometallics 1989, 8, 2907.
1
complexes 2 and 3 can be readily differentiated by both H and
(
13
C NMR spectroscopies in CD
chemical shift difference observed by C NMR spectroscopy for
the -OOCH and -OCH groups of 2 and 3, respectively. More
importantly, when the C solvent was removed from the
completed reaction between 1 and O and replaced with CD Cl
2 2
Cl . Most significant is the large
13
(
24) Kapteijn, G. M.; Dervisi, A.; Verhoef, M. J.; van den Broek, M. A. F. H.;
Grove, D. M.; van Koten, G. J. Organomet. Chem. 1996, 517, 123.
25) See Experimental Section for details. The protocol for the synthesis
of MeOOH was adapted from: (a) Rieche, A.; Hitz, F. Ber. 1929, 62,
3
3
6 6
D
(
2
2
2
,
2
9
458. (b) Vaghjiani, G. L.; Ravishankara, A. R. J. Phys. Chem. 1989,
3, 1948. See also: (c) Davies, D. M.; Deary, M. E. J. Chem. Soc.,
the spectra were indistinguishable from those of 2 obtained by
reactions involving MeOOH.
Perkin Trans. 2 1992, 559. (d) O’Sullivan, D. W.; Lee, M.; Noone,
B. C.; Heikes, B. G. J. Phys. Chem. 1996, 100, 3241.
Complex 2 is a rare example of a Pd(II) alkylperoxide
3
0
(
26) Pure alkyl hydroperoxides are potentially explosive and should be
avoided if possible. They can be violently decomposed by adventitious
catalysts (acids, metals) with formation of, among others, oxygen gas.
Distillation of alkylperoxides is not recommended. See: (a) Sharpless,
K. B.; Verhoeven, T. R. Aldrichimica Acta 1979, 12, 63. (b) Hill,
J. G.; Rossiter, B. E.; Sharpless, K. B. J. Org. Chem. 1983, 48, 3607.
It has been documented that explosions occurred when pure MeOOH
was heated (ref 25a) and that the residue obtained after distillation of
pure samples of MeOOH exploded violently (ref 25b).
complex, and it appears to be the first reported Pd(II)
methylperoxide complex. Notably, two Pt(II) methylperoxide
complexes, formed by the overall insertion of O
bonds, were recently characterized by X-ray crystallography.
Effects of Light and AIBN on the Reaction between
bipy)PdMe (1) and O . The reaction between (bipy)PdMe (1)
and O (1-10 atm) to form (bipy)PdMe(OOMe) (2) in C
2
into Pt(II)-Me
11,12
(
2
2
2
2
6 6
D
27) MeOOH is easily distinguished from MeOH by both H and 13C NMR
1
(
spectroscopies, as shown by HMQC analysis performed on a ∼1:2:2
(28) Kapteijn, G. M.; Dervisi, A.; Grove, D. M.; Kooijman, H.; Lakin,
M. T.; Spek, A. L.; van Koten, G. J. Am. Chem. Soc. 1995, 117, 10939.
(29) Ruiz, J.; Martinez, M. T.; Florenciano, F.; Rodriguez, V.; Lopez, G.; P e´ rez,
J.; Chaloner, P. A.; Hitchcock, P. B. Inorg. Chem. 2003, 42, 3650.
3 3 2 2 2 3
CH OOH/CH OH/Et O mixture (CD Cl , 500 MHz, 233 K): CH OOH
13 1
1
(
3
H NMR: 3.78 ppm, C NMR: 65.3 ppm); CH OH ( H NMR: 3.36
13
ppm, C NMR: 50.4 ppm).
5804 J. AM. CHEM. SOC. 9 VOL. 131, NO. 43, 2009
1

Insertion of O
2
into Pd(II)-Me
A R T I C L E S
dioxane internal standard, remained approximately constant
throughout the reactions and is reported as the average value
3
7
determined over at least two half-lives.
Half-order and first-order kinetic plots with respect to (bipy)-
PdMe (1) for representative reactions at pressures of oxygen of 5
2
and 10 atm are presented in Figure 2. More examples of such
comparisons can be found in the Supporting Information, Figures
S2 and S3. In addition, kinetic plots for three-halves order and
second-order dependences for the data shown in Figure 2b are also
provided in Figure S1. It is clear from examination of these data
that the rate of consumption of 1 exhibits a well-behaved half-
order dependence on [1]. Furthermore, Figure 2 also shows that
this behavior is observed both at high (Figure 2a) and low (Figure
Figure 1. Concentrations of (bipy)PdMe
2
(1) and (bipy)PdMe(OOMe) (2) as
a function of time as determined by H NMR spectroscopy. Reaction conditions:
1] ) 5.2 mM, [AIBN] ) 5.8(1) mM, p(O ) ) 10 atm, C , 323 K.
1
2b) concentrations of AIBN.
[
0
2
6 6
D
Determination of the Order in [AIBN]. Values of the rate
1
obs
constant k , calculated from the slope of the half-order plots
was monitored by H NMR spectroscopy. Initially, the rates
were found to be highly irreproducible, which is often indicative
of the involvement of radical intermediates. Notably, a majority
1
/2
1/2
according to [1]
t
- [1]
0
) -(kobs/2)·t, were obtained for
reactions at different concentrations of AIBN. As shown by the
plots in Figure 3 for reactions at pressures of oxygen of 5 and 10
of the known reactions of insertion of O
been found to proceed by radical chain pathways.
reactions, the effects of radical initiators or inhibitors and light
have been well documented. Similar effects were observed in
2
the reactions between 1 and O . For example, in ambient light
at room temperature, 42% of 1 reacted after ca. 6 h, compared
with 7% for a sample kept in the dark (see Experimental
Section). When reactions between 1 and O
2
into M-R bonds have
10-16
obs
atm, a linear relationship is observed between the value of k and
For these
1
/2 38
1
[
AIBN] . Meanwhile, a linear fit to [AIBN] is much less
satisfying.
The half-order dependence on the concentration of AIBN was
confirmed by treatment of the kinetic data as shown in Figure
3
1a
n
4
.
If the rate of the reaction is proportional to [AIBN] (eq
2
), then the kinetic data can be plotted as indicated in eq 3.
2
were conducted in
the presence of the radical initiator AIBN, higher rates and yields
n
1
rate ) -∆[(bipy)PdMe ]/∆t ∝ [AIBN]
(2)
(3)
(
up to ca. 85% of 2 as determined by H NMR spectroscopy)
2
relative to those for reactions with no added initiator were
observed (see Experimental Section). Significantly, the presence
of AIBN led to reproducible reaction rates, so that kinetic studies
n
[
(bipy)PdMe ] ∝ [AIBN] ·t
2
3
1
Equation 3 shows that for reactions with similar initial
concentrations of (bipy)PdMe (1), plots of [1] versus [AIBN] ·t
could be carried out (Figure 1).
Determination of the Order in [(bipy)PdMe
of the reaction between 1 and O were reproducible when AIBN
n
2
2
]. Since the rates
should be superimposable. Figure 4 shows such an analysis for
two reactions having similar initial concentrations of 1 but
different concentrations of AIBN. The significantly better
overlay found in Figure 4 for plots of [1] versus [AIBN] ·t
than for plots of [1] versus [AIBN] ·t confirms a half-order
dependence on [AIBN] for the reactions of autoxidation of 1.
2
was used as a radical initiator, investigations into the mechanism
of the reaction were pursued via kinetic studies of the reaction
in the presence of AIBN. Kinetic experiments were conducted
1
/2
3
2,33
at 323 K in C
6
D
6
at pressures of oxygen of 5 and 10 atm,
1
and H NMR spectra were collected over at least three half-
3
4
lives. The disappearance of the bipyridine signal of 1 ([1] )
0
(
33) The concentration of dissolved O
to the concentration in C . Assuming that the ideal gas law and
Henry’s law are applicable, a doubling of the pressure of O is expected
to lead to a doubling of the concentration of dissolved O . The
concentration of O in C H at 323 K (using an Ostwald coefficient
2 6 6
in C D is expected to be similar
3
.8-5.2 mM) at 8.8 ppm could be easily monitored (and
6
H
6
35
integrated against an internal standard) and maximum yields
2
of ca. 85% of 2 were observed after complete consumption of
2
3
6
2
6
6
1
.
Notably, no intermediates were detected during the reac-
tions. The concentration of AIBN (2-32 mM), determined by
integration of its CH signal relative to the signal of the 1,4-
3
23K
323K
L
(
of 0.247 (ref a) and a density for C
H
6 6
of d
) 0.8481 g/mL
ref b)) at partial pressures of O of 5 and 10 atm is calculated to be
2
3
of about 47 and 93 mM, respectively. (a) Oxygen and Ozone; Battino,
R., Ed.; Solubility Data Series; Pergamon Press: Oxford, 1981; Vol.
7
, p 250. (b) Lide, D. R. Handbook of Chemistry and Physics, 85th
ed.; CRC Press: Boca Raton, FL, 2004-2005; p 15-28.
(34) The determination of initial rates at low [AIBN] with (bipy)PdMe
proved unsatisfactory as too
few data points could be acquired before reaching >10-20% conver-
sions.
(35) Similar results were obtained when the rates were determined by
following the disappearance of the methyl signal of 1 but overlap with
the signal for AIBN precluded a full analysis.
(30) For other examples, see: (a) Kujime, M.; Hikichi, S.; Akita, M. Chem.
Lett. 2003, 32, 486. (b) Miyaji, T.; Kujime, M.; Hikichi, S.; Moro-
oka, Y.; Akita, M. Inorg. Chem. 2002, 41, 5286. (c) Oshima, N.;
Hamatani, Y.; Fukui, H.; Suzuki, H.; Moro-oka, Y. J. Organomet.
Chem. 1986, 303, C21. (d) Strukul, G.; Ros, R.; Michelin, R. A. Inorg.
Chem. 1982, 21, 495. (e) Mimoun, H.; Charpentier, R.; Mitschler,
A.; Fischer, J.; Weiss, R. J. Am. Chem. Soc. 1980, 102, 1047.
31) The appearance of the curves in Figure 1 suggests that auto-acceleration
or auto-catalysis effects, for which sigmoidal curves would be expected,
are not involved in the present reactions. Effects of this type have been
observed in the autoxidation of alkyl complexes of boron (see ref 15p)
and cadmium (see ref 15m). For recent discussions concerning their
involvement in autoxidation reactions, see: (a) Look, J. L.; Wick, D. D.;
Mayer, J. M.; Goldberg, K. I. Inorg. Chem. 2009, 48, 1356. (b) Hermans,
I.; Jacobs, P. A.; Peeters, J. Chem.sEur. J. 2006, 12, 4229.
2
or
(4,4′-di-tert-butyl-2,2′-bipyridine)PdMe
2
(
(36) In all reactions, complete consumption of 1 was observed and extents
of conversion reported in the different plots refer to the maximum
conversion that could be reliably determined by integration of the
1
signals in the H NMR spectra.
(37) The half-life of AIBN at 50 °C in C
H
6 6
is calculated to be ca. 87.5 h
-
6
-1
(k ) 2.2 × 10
s ). See: Polymer Handbook, 4th ed.; Brandrup, J.,
Immergut, E. H., Grulke, E. A., Eds.; John Wiley & Sons: New York,
1999.
(
32) The use of high pressures (several atmospheres) ensures that the
reaction will not be affected by inefficient mass transport of O
2
through
(38) The negative intercepts of the plots of kobs versus [AIBN]1/2 are most
likely due to the presence of small amounts of adventitious inhibitors.
This behavior is also consistent with the irreproducible kinetics
observed at low concentrations of AIBN or when no AIBN was added.
the narrow gas/liquid interface of the NMR tubes used in the
experiments. For recent discussions, see: ref 31a and Steinhoff, B. A.;
Stahl, S. S. J. Am. Chem. Soc. 2006, 128, 4348.
J. AM. CHEM. SOC. 9 VOL. 131, NO. 43, 2009 15805

A R T I C L E S
Boisvert et al.
2 6 6
Figure 2. Representative half-order and first-order kinetic plots for reactions under (a) 5 and (b) 10 atm of O in C D at 323 K.
Figure 3. Dependence of the value of the observed rate constant on the concentration of AIBN showing a linear relationship between kobs and [AIBN]1/2
for reactions with pressures of O of (a) 5 and (b) 10 atm. Reaction conditions: [1] ) 3.8-5.3 mM, C , 323 K.
2
0
6 6
D
-4
Other similar examples showing much better agreement between
half-order plots than between first-order plots can be found in
the Supporting Information, Figure S4.
O
2
of 5 atm (Figure 3a), a value of k ) 9.8(3) × 10 s-1 is
2
obtained, and for the experiments with pressures of O of 10
-
4
-1
atm (Figure 3b), a value of k ) 9.1(3) × 10
The very similar values of k at pressures of O
indicate that the rate of the reaction is independent of [O
s
is obtained.
Determination of the Order in [O
2
]. From the slope of a plot
2
of 5 and 10 atm
1
/2
33
of kobs versus [AIBN] for the experiments with pressures of
2
].
1
5806 J. AM. CHEM. SOC. 9 VOL. 131, NO. 43, 2009

Insertion of O
2
into Pd(II)-Me
A R T I C L E S
Figure 4. Overlay of plots according to eq 3 showing a better agreement between plots of [1] vs [AIBN]1/2 ·t than between plots of [1] vs [AIBN] ·t.
Empirical Rate Law for the Reaction between (bipy)PdMe
1) and O . Overall, the insertion of O (5-10 atm) into a palladium
methyl bond of (bipy)PdMe (1) in the presence of AIBN in C
2
is proposed, and this mechanism is compared to ones that have
(
2
2
been proposed for insertions of O
bonds.
2
into late metal-hydride
2
6 6
D
at 323 K is a first-order process, half-order with respect to 1 and
half-order with respect to AIBN. There is a zero-order dependence
on the concentration of oxygen. These results are summarized in
the empirically derived rate expression shown in eq 4.
Autoxidation of Organic Compounds and Extension to the
Autoxidation of 1. The autoxidation of organic compounds to
form alkyl hydroperoxides is usually thought to occur by the
43,44
general mechanism shown in Scheme 2.
In this mechanism,
the chain-propagating and product-forming step is a hydrogen
atom abstraction (step 5a).
Kinetic studies on these reactions are usually conducted in
the presence of a radical initiator like AIBN to provide
reproducible reaction rates and avoid variable initiation periods.
1
/2
rate ) -d[(bipy)PdMe ]/dt ) kobs[1]
)
2
1
/2
1/2
0
k[1] [AIBN] [O ] (4)
2
Discussion
2
In addition, sufficient pressures of O are typically used to ensure
The kinetic studies presented herein, together with the
observed effect of radical initiators and light, strongly suggest
the involvement of a radical chain mechanism for the reaction
that the dominant termination process is the homocoupling of
43,44
two alkylperoxyl radicals (step 6a).
Under these conditions,
2 2
of (bipy)PdMe (1) with O to form (bipy)PdMe(OOMe) (2).
(
41) For representative examples of the involvement of mononuclear Pd(I)
Since 1 is a Pd(II) complex, the possibility that Pd(I) and/or
Pd(III) complexes might be involved as chain carriers must be
considered. Notably, while mononuclear complexes of Pd(I) and
complexes, see: (a) Fafard, C. M.; Adhikari, D.; Foxman, B. M.;
Mindiola, D. J.; Ozerov, O. V. J. Am. Chem. Soc. 2007, 129, 10318. (b)
Burns, C. T.; Shen, H.; Jordan, R. F. J. Organomet. Chem. 2003, 683,
240. (c) Alb e´ niz, A. C.; Espinet, P.; Lopez-Fernandez, R.; Sen, A. J. Am.
Chem. Soc. 2002, 124, 11278. (d) Stadtm u¨ ller, H.; Vaupel, A.; Tucker,
C. E.; St u¨ demann, T.; Knochel, P. Chem.sEur. J. 1996, 2, 1204. (e)
Ishiyama, T.; Miyaura, N.; Suzuki, A. Tetrahedron Lett. 1991, 32, 6923.
(f) Stille, J. K.; Lau, K. S. Y. J. Am. Chem. Soc. 1976, 98, 5841. (g)
Osborn, J. A. In Organotransition-Metal Chemistry; Ishii, Y., Tsutsui,
M., Eds.; Plenum Press: New York, 1975; p 65. (h) Kramer, A. V.;
Osborn, J. A. J. Am. Chem. Soc. 1974, 96, 7832. (i) Kramer, A. V.;
Labinger, J. A.; Bradley, J. S.; Osborn, J. A. J. Am. Chem. Soc. 1974,
3
9,40
Pd(III) complexes are rare,
their intermediacy has been
proposed or verified in a variety of situations, most commonly
during radical oxidative addition/reductive elimination processes
4
1,42
and during electrochemical studies.
The reproducible rates of reaction in the presence of AIBN
and the lack of a dependence on the oxygen pressure for the
reactions between (bipy)PdMe (1) and O are reminiscent of
2 2
the observations recorded for autoxidations of organic substrates.
The general radical chain mechanism established for organic
autoxidations is discussed below and compared to the results
9
6, 7145. (j) Stille, J. K.; Hines, L. F.; Fries, R. W.; Wong, P. K.; James,
D. E.; Lau, K. AdV. Chem. Ser. 1974, 132, 90.
(42) For representative examples of the involvement of mononuclear Pd(III)
complexes, see: (a) Kraatz, H.-B.; van der Boom, M. E.; Ben-David,
Y.; Milstein, D. Isr. J. Chem. 2001, 41, 163. (b) Matsumoto, M.; Itoh,
M.; Funahashi, S.; Takagi, H. D. Can. J. Chem. 1999, 77, 1638. (c)
Blake, A. J.; Crofts, R. D.; de Groot, B.; Schr o¨ der, M. J. Chem. Soc.,
Dalton Trans. 1993, 485. (d) Pandey, K. K. Coord. Chem. ReV. 1992,
2
obtained for the experiments involving (bipy)PdMe (1). In
addition, important differences between the general radical chain
mechanisms for autoxidation of organic compounds and that
of main group and early transition metal alkyl complexes are
highlighted and discussed in the context of the autoxidation of
1
21, 1. (e) Reid, G.; Schr o¨ der, M. Chem. Soc. ReV. 1990, 19, 239. (f)
McAuley, A.; Whitcombe, T. W. Inorg. Chem. 1988, 27, 3090. (g)
Lane, G. A.; Geiger, W. E.; Connelly, N. G. J. Am. Chem. Soc. 1987,
109, 402. (h) van Leeuwen, P. W. N. M.; Roobeek, C. F.; Huis, R. J.
Organomet. Chem. 1977, 142, 233. (i) Manolikakes, G.; Knochel, P.
Angew. Chem., Int. Ed. 2009, 48, 205.
1
. A mechanism for the reaction between 1 and O
2
to form
(bipy)PdMe(OOMe) (2) consistent with the experimental results
(
43) (a) Reich, L.; Stivala, S. S. Autoxidation of Hydrocarbons and
Polyolefins-Kinetics and Mechanisms; Marcel Dekker: New York,
1969; Chapters 2 and 3. (b) Huyser, E. S. Free-Radical Chain
Reactions; John Wiley & Sons: New York, 1970; Chapters 3 and 11.
(c) Howard, J. A. In Free Radicals; Kochi, J. K., Ed.; John Wiley &
Sons: New York, 1973; Vol. II, Chapter 12. (d) Walling, C. In ActiVe
Oxygen in Chemistry; Foote, C. S., Valentine, J. S., Greenberg, A.,
Liebman, J. F., Eds.; Blackie Academic & Professional: London, 1995;
Chapter 2.
(
39) Canty, A. J. In Handbook of Organopalladium Chemistry for Organic
Synthesis; Negishi, E. I., Ed.; Wiley: New York, 2002; Vol. 1, p 189.
40) For binuclear Pd(I) complexes, see: (a) Murahashi, T.; Kurosawa, H.
Coord. Chem. ReV. 2002, 231, 207. For binuclear Pd(III) complexes,
see: (b) Powers, D. C.; Ritter, T. Nat. Chem. 2009, 1, 302. (c) Cotton,
F. A.; Koshevoy, I. O.; Lahuerta, P.; Murillo, C. A.; Sanau, M.; Ubeda,
M. A.; Zhao, Q. J. Am. Chem. Soc. 2006, 128, 13674. (d) Ara, I.;
Chaouche, N.; Forni e´ s, J.; Fortuno, C.; Kribii, A.; Tsipis, A. C.
Organometallics 2006, 25, 1084. (e) Bond, A. M.; Canty, A. J.;
Cooper, J. B.; Tedesco, V.; Traill, P. R.; Way, D. M. Inorg. Chim.
Acta 1996, 251, 185.
(
(44) (a) Porter, N. A. Acc. Chem. Res. 1986, 19, 262. (b) Betts, J. Q. ReV.,
Chem. Soc. 1971, 25, 265. (c) Mayo, F. R. Acc. Chem. Res. 1968, 1,
193.
J. AM. CHEM. SOC. 9 VOL. 131, NO. 43, 2009 15807

A R T I C L E S
Boisvert et al.
Scheme 2
clear that the reaction does not proceed by the mechanism
presented in Scheme 3 that involves radical substitutions at the
carbon of the Pd-Me group of 1.
The methyl group abstraction reactions by radicals from
Pd-Me could also be viewed as bimolecular homolytic
H
substitutions at carbon (S 2). Notably, very few examples of
homolytic displacements at the carbon of transition metal alkyl
45
complexes have been reported. Homolytic substitutions at the
carbon of transition metal alkyls that involve simple alkyl groups
(
e.g., M-R with R ) Me, benzyl), for which a concerted
displacement (S 2) is expected, have thus far been limited to
Co(III) complexes of macrocyclic nitrogen ligands. This
reactivity is likely related to the facility with which these types
of complexes generate alkyl radicals by homolysis of Co -R
bonds.
H
4
6
III
1
0,47
Rare examples of radical substitutions at the alkyl
4
8
49
49
group of complexes of other metals like Fe, Rh, and Ir
involve unsaturated groups of the type M-CR -CR)CR (e.g.,
M-(η -allyl), M-(η -cyclopentadienyl)). For these, and for
2
2
1
1
4
6a,c
many Co complexes,
mechanistic studies suggest that the
homolytic displacement most likely does not involve direct
attack at the carbon bound to the metal (R position) but rather
an initial attack at the γ carbon of the allyl system, with an
overall transposition of the allyl moiety. Considered together
with the inconsistent rate law, this literature background
the rate of a reaction proceeding by the mechanism shown in
Scheme 2 will be independent of the pressure of oxygen and
will show a first-order dependence on the concentration of the
substrate and a half-order dependence on the concentration of
4
3b,c
AIBN (eq 5).
concerning S
the conclusion that a mechanism for the reaction of (bipy)PdMe
1) with O that involves radical substitution at a methyl group
of 1 (i.e., methyl group abstraction from 1) is unlikely.
H
2 reactions at metal alkyl species further supports
1
1/2
0
rate ∝ [RH] [AIBN] [O ]
(5)
2
2
(
2
The reaction of insertion of O
2
into a Pd-Me bond of 1 in
the presence of AIBN could potentially proceed by a mechanism
analogous to the one found in Scheme 2 for the autoxidation of
organic compounds. Such a mechanism would involve the
abstraction by (bipy)PdMe(OO ·) of a methyl group from
Autoxidation of Main Group and Early Transition Metal
Alkyl Complexes and Extension to the Autoxidation of 1. The
formation of metal alkylperoxides (or other oxygenated deriva-
tives) by autoxidation reactions has been documented for a wide
variety of main group and early transition metal alkyl
2
(bipy)PdMe (1) (steps 5b in Scheme 3) by homolytic displace-
1
4-16
ment at the carbon of Pd-Me. In this mechanism, Pd(I) radicals
complexes.
Mechanistic studies have been reported on a
4
1
would be involved as chain carriers.
relatively limited number of these reactions and radical chain
1
4a
Analogously to what was presented in Scheme 2 for the
autoxidation of organics, the third step of the initiation would
then consist of a methyl group abstraction from 1 by InOO·
and, under a sufficiently high pressure of oxygen, the termination
step would involve the reaction of two (bipy)PdMe(OO·)
species to generate non-chain propagating species.
The rate law derived for this series of steps (eq 6) would
have the same form as the one found for the autoxidation of
organics (eq 5).
mechanisms are usually involved. In the absence of radical
initiators, the initiation step generally proposed for these
reactions consists of the generation of R · radicals by homolysis
1
4a
of M-R bonds. For reactions conducted in the presence of
a radical initiator like AIBN, a general mechanism can be
described by the series of steps illustrated in Scheme 4. In such
n
(
45) The following two-step transformation has been documented: M +
n+1
n+1
n
Me· f M (Me) followed by M (Me) + Me· f M + Me-Me.
The second step may involve a homolytic displacement at the alkyl
group (or attack at the metal center followed by reductive elimination)
but mechanistic studies are lacking. For examples with Ni complexes,
see: (a) Kurzion-Zilbermann, T.; Masarwa, A.; Maimon, E.; Cohen,
H.; Meyerstein, D. Dalton Trans. 2007, 3959. Radical oxidative
addition and reductive elimination processes of transition metal
complexes can involve steps related to homolytic substitutions at alkyl
groups. For relevant discussions on these types of reactions, see: (b)
Kochi, J. K. Organometallic Mechanisms and Catalysis; Academic
Press: New York, NY, 1978. (c) Chanon, M. Bull. Soc. Chim. Fr.
1982, II–197.
1
1/2
0
rate ∝ [1] [AIBN] [O ]
(6)
2
For the reactions of autoxidation of (bipy)PdMe
2
(1), a half-
order dependence on [AIBN] was indeed found (eq 4). Ad-
ditionally, a zero-order dependence on [O ] was determined (for
reactions at pressures of O of 5 and 10 atm), suggesting that
2
2
3
2
(
(
1) the reaction is not limited by O
2) a sufficient excess concentration of O
-derived radicals. As
described above, a zero-order dependence on [O ] is common
2
mass-transfer effects and
2
is present to ensure
(
46) For a review, see: (a) Johnson, M. D. Acc. Chem. Res. 1983, 16, 343.
For more recent examples, see: (b) Li, S.; de Bruin, B.; Peng, C.-H.;
Fryd, M.; Wayland, B. B. J. Am. Chem. Soc. 2008, 130, 13373. (c)
Gupta, B. D.; Vijaikanth, V. J. Organomet. Chem. 2004, 689, 1102.
termination by the coupling of two O
2
2
for radical chain autoxidations of organic compounds conducted
under a sufficient excess of oxygen. This zero-order oxygen
dependence has also been reported for the autoxidation of several
main group and early transition metal alkyl complexes (see
(
(
1
d) Stolzenberg, A. M.; Cao, Y. J. Am. Chem. Soc. 2001, 123, 9078.
e) Huston, P.; Espenson, J. H.; Bakac, A. J. Am. Chem. Soc. 1992,
14, 9510.
(
(
(
47) (a) Halpern, J. Acc. Chem. Res. 1982, 15, 238. (b) Peng, C.-H.; Li,
1
4,15i,g,n-p
below).
1] is expected from eq 6, a well-behaved half-order dependence
on [1] was found experimentally for the reaction between
bipy)PdMe (1) and O in the presence of AIBN. It is therefore
However, while a first-order dependence on
S.; Wayland, B. B. Inorg. Chem. 2009, 48, 5039.
48) (a) Fabian, B. D.; Labinger, J. A. Organometallics 1983, 2, 659. (b)
Rosenblum, M.; Waterman, P. J. Organomet. Chem. 1981, 206, 197.
49) Crease, A. E.; Gupta, B. D.; Johnson, M. D.; Moorhouse, S. J. Chem.
Soc., Dalton Trans. 1978, 1821.
[
(
2
2
1
5808 J. AM. CHEM. SOC. 9 VOL. 131, NO. 43, 2009

Insertion of O
2
into Pd(II)-Me
A R T I C L E S
Scheme 3
Scheme 4
the incoming radical and the cleavage of the bond with the
H
departing radical can be concerted (S 2). This reactivity is
expected for displacements that occur at centers that do not have
the ability to form hypervalent intermediates, such as the carbon
of simple alkyl groups (as in Scheme 3). Homolytic displace-
ments can also be stepwise associative processes, involving an
intermediate formed by expansion of the valency at the reacting
center. A typical example of this behavior is encountered in
53
the autoxidation of trivalent phosphorus compounds. Because
many transition metals can easily expand their valency, a variety
of mechanisms are expected for homolytic displacements at
transition metal centers, including concerted (as was presented
in Scheme 5) and stepwise associative processes.
The reactivity patterns observed within the chemistry of Pt(II)
complexes are often used as models for analogous Pd(II)
complexes. In this context, it should be noted that there is
substantial precedent for radical substitution reactions at Pt(II)
a mechanism, the chain-propagating and product-forming step
is a bimolecular homolytic substitution at the metal center (step
t
centers. For example, non-chain radical substitutions by BuO·
5
0
5
c).
Under conditions of sufficient oxygen concentration such that
radicals at Pt(II) centers to release alkyl radicals have been
54
documented (Scheme 6a). The displaced R · radicals have been
step 4a is rapid, the rate law derived for this mechanism is given
by eq 7. Accordingly, when radical initiators such as AIBN are
used in these reactions, a first-order dependence on the
concentration of the metal alkyl and a half-order dependence
detected by electron spin resonance spectroscopy and character-
ized by spin-trapping experiments. In addition, radical chain
substitution reactions of alkyl groups by RS · radicals at Pt(II)
54,55
centers have been reported,
and it has been shown that some
1
5o
on the concentration of AIBN are often found. Notably, this
rate law is similar to that found for reactions involving organics
55
of these reactions benefit from the addition of AIBN. It has
also been proposed that the photolysis of Pt(IV)–methyl
complexes leads to the formation of five-coordinate Pt(III)
complexes, and that Me · radicals can be eliminated from these
species to form Pt(II) complexes (Scheme 6b). Furthermore,
the reactions of (diimine)PtMe
(
see eq 5).
1
1/2
0
rate ∝ [MR] [AIBN] [O ]
(7)
56
2
i
2
complexes such as 6 with PrI
The observed reaction of autoxidation of (bipy)PdMe
2
(1)
in the presence of air are proposed to proceed by a radical chain
could potentially proceed by a mechanism analogous to the one
found in Scheme 4 for the autoxidation of main group and early
transition metal alkyl complexes. Such a mechanism would
involve homolytic substitutions at the Pd center of 1 by MeOO ·
radicals (step 5d in Scheme 5). In analogous fashion to what
was presented in Scheme 4 for the autoxidation of main group
and early transition metal alkyl complexes, the third step of the
initiation would then consist of a homolytic substitution of a
Me· group of 1 by InOO · and, under a sufficiently high pressure
of oxygen, the termination step would involve the homocoupling
of two MeOO · radicals. However, the derived rate law for such
a process would have the same form as eq 6 derived above for
Scheme 3, so that a first-order dependence on [1] would be
57
mechanism as shown in Scheme 6c. In this mechanism, the
addition of a peroxyl radical, formed by the interception of an
(
51) Related discussions about the distinction between radical substitution
at the alkyl group and at the metal center for radical chain insertion
reactions into M-R bonds have been previously presented. For
example, based on the well-documented reactivity of Co(III) alkyl
III
complexes, the radical chain insertion of SO
2
into Co -R bonds was
proposed to proceed via homolytic displacements at carbon rather than
via homolytic displacements at the metal center (refs a and b). In
contrast, in ref 15g a mechanism involving radical substitutions at
the metal center was favored for the reaction between a chromium
2
alkyl complex and O . (a) Crease, A. E.; Johnson, M. D. J. Am. Chem.
Soc. 1978, 100, 8013. (b) Gupta, B. D.; Roy, M.; Oberoi, M.; Dixit,
V. J. Organomet. Chem. 1992, 430, 197.
(
(
(
52) (a) Walton, J. C. Acc. Chem. Res. 1998, 31, 99. (b) Schiesser, C. H.;
Wild, L. M. Tetrahedron 1996, 52, 13265. (c) Collman, J. P.; Hegedus,
L. S.; Norton, J. R.; Finke, R. G. Principles and Applications of
Organotransition Metal Chemistry; University Science Books: Mill
Valley, CA, 1987; Chapter 4.
expected. Thus, if the reaction between (bipy)PdMe
were to proceed by concerted homolytic displacements at
the carbon of Pd-Me (Scheme 3) or at the Pd center (Scheme
), the rate of the reaction would be expected to show a half-
order dependence on [AIBN] and a first-order dependence on
(bipy)PdMe ] (eq 6). Experimentally, a half-order dependence
on both [AIBN] and [(bipy)PdMe ] is found (eq 4). Therefore
a different mechanism must be operative.
Mechanism of the Reaction between 1 and O
substitution reactions can proceed by a variety of mecha-
2
(1) and
O
2
5
53) (a) Ogata, Y.; Yamashita, M. J. Chem. Soc., Perkin Trans. 2 1972,
7
30. (b) Hwang, W.-S.; Yoke, J. T. J. Org. Chem. 1980, 45, 2088. (c)
[
2
Schwetlick, K. Pure Appl. Chem. 1983, 55, 1629. For reactions
between BuOO· radicals and trivalent phosphorus compounds, see:
t
2
5
1
(d) Furimsky, E.; Howard, J. A. J. Am. Chem. Soc. 1973, 95, 369.
54) (a) Cardin, D. J.; Lappert, M. F.; Lednor, P. W. J. Chem. Soc., Chem.
Commun. 1973, 350. See also: (b) Kaptein, R.; van Leeuwen,
P. W. N. M.; Huis, R. J. Chem. Soc., Chem. Commun. 1975, 568.
2
. Free radical
45b,50,52
(55) Johnson, A.; Puddephatt, R. J. J. Chem. Soc., Dalton Trans. 1975,
nisms.
For example, the formation of the new bond with
1
15.
(
56) Perkins, D. C. L.; Puddephatt, R. J.; Tipper, C. F. H. J. Organomet.
Chem. 1979, 166, 261 (see also ref 42h).
(
50) Davies, A. G.; Roberts, B. P. Acc. Chem. Res. 1972, 5, 387.
J. AM. CHEM. SOC. 9 VOL. 131, NO. 43, 2009 15809

A R T I C L E S
Boisvert et al.
Scheme 5
III
Scheme 6
Pd(III) intermediate, (bipy)Pd (Me)
(OOMe) · (8) (Scheme
2
4
2,59
7
, steps 5i and 5ii).
The involvement of this intermediate
could be key in differentiating the observed reactivity of 1
from reactions of autoxidation of organic compounds where
the formation of an analogous hypervalent intermediate is
not possible.
It is reasonable to propose that the reaction between 1 and
MeOO· radicals to form 8 would be reversible (Scheme 7, steps
6
0
5
i and -5i), and that loss of either MeOO · or Me· should be
possible from intermediate 8. However, at the high pressures
of oxygen used experimentally, the release of the Me · radicals
(
step 5ii) would be rendered irreversible by their fast trapping
6
1
with O to form MeOO · radicals.
2
The rate law for such a mechanism, if the termination step
were the coupling of two MeOO · radicals as proposed above
6
2
in Scheme 5, is given by eq 6, so that a first-order
dependence of the rate on [1] would still be expected.
However, under the reaction conditions, the concentration
6
3
of 1 is expected to be much larger than that of MeOO · .
Consequently, one might expect that the reaction of MeOO ·
radicals with 1 would yield a significant concentration of
III
64
pentacoordinate Pd(III) species (bipy)Pd (Me) (OOMe) · (8).
2
Under these conditions, it is quite reasonable to suggest that
the coupling of 8 with MeOO · would then become the dominant
6
5
termination step (Scheme 7, step 6e). In fact, a very similar
process (involving the reversible reaction of ROO· with an
inhibitor followed by reaction of the resulting complex with a
second ROO · radical) was originally proposed by Hammond
(
59) Reactions of Pd(II) complexes with in situ generated ROO · radicals
t
2 2
from H O or BuOOH have been proposed: (a) Kamaraj, K.;
Bandyopadhyay, D. Organometallics 1999, 18, 438. (b) Wadhwani,
P.; Bandyopadhyay, D. Organometallics 2000, 19, 4435.
alkyl radical by O
2
, to the Pt(II) center of 6 leads to the
(60) The following reaction between a first-row transition metal complex
(
often with L ) macrocyclic nitrogen ligand; M ) Cr, Mn, Fe, Co,
formation of a five-coordinate Pt(III) intermediate that abstracts
Ni, and also Cu) and Me · or MeOO · radicals in aqueous medium is
well documented: (L)M (H
reactions involving MeOO·, values of K ) 10 -10 have been
determined for Fe (ref d) and Cu (ref e) complexes, while the reaction
i
iodine from PrI. Other examples of oxidative additions involving
n
n+1
2
O) + R· T (L)M (R) + H
2
O. For
2
3
Pt(II) complexes to form Pt(IV) derivatives have also been
proposed to proceed through the intermediacy of five-coordinate
eq
5
8
seems to be essentially irreversible for some Co complexes (refs f
and g). (a) Goldstein, S.; Czapski, G.; van Eldik, R.; Shaham, N.;
Cohen, H.; Meyerstein, D. Inorg. Chem. 2001, 40, 4966. (b) van Eldik,
R.; Meyerstein, D. Acc. Chem. Res. 2000, 33, 207. (c) van Eldik, R.;
Cohen, H.; Meyerstein, D. Inorg. Chem. 1994, 33, 1566. (d) Mansano-
Weiss, C.; Cohen, H.; Meyerstein, D. J. Inorg. Biochem. 2002, 91,
199. (e) Mansano-Weiss, C.; Masarwa, A.; Cohen, H.; Meyerstein,
D. Inorg. Chim. Acta 2005, 358, 2199. (f) Solomon-Rapaport, E.;
Masarwa, A.; Cohen, H.; Valentine, J. S.; Meyerstein, D. Eur. J. Inorg.
Chem. 2002, 2427. (g) Solomon-Rapaport, E.; Masarwa, A.; Cohen,
H.; Meyerstein, D. Inorg. Chim. Acta 2000, 299, 41.
Pt(III) radicals.
These reports indicate that radical substitution reactions at
Pt(II) centers are viable pathways and that five-coordinate Pt(III)
complexes can be reasonably proposed as intermediates. An
extension of such reactivity to palladium leads to the proposed
mechanism shown in Scheme 7 for the autoxidation of 1.
Rather than the concerted displacements presented in Scheme
5
, the displacement of Me · by MeOO · could be a stepwise
associative process with initial formation of a pentacoordinate
(61) The coupling of R · with O to form ROO · is essentially diffusion-
2
limited (for compilations of rate data see refs a and b). A value of k
9
-1 -1
)
4.7 × 10 M
s
has been determined for the reaction of Me ·
(
(
57) (a) Ferguson, G.; Monaghan, P. K.; Parvez, M.; Puddephatt, R. J.
Organometallics 1985, 4, 1669. (b) Monaghan, P. K.; Puddephatt, R. J.
Organometallics 1986, 5, 439. (c) Hill, R. H.; Puddephatt, R. J. J. Am.
Chem. Soc. 1985, 107, 1218.
2 2
with O in H O at 296 K (ref c). (a) Neta, P.; Huie, R. E.; Ross, A. B.
J. Phys. Chem. Ref. Data 1990, 19, 413. (b) Neta, P.; Grodkowski, J.;
Ross, A. B. J. Phys. Chem. Ref. Data 1996, 25, 709. (c) Thomas,
J. K. J. Phys. Chem. 1967, 71, 1919.
58) See: ref 42d and (a) Thomas, S. W., III; Venkatesan, K.; M u¨ ller, P.;
Swager, T. M. J. Am. Chem. Soc. 2006, 128, 16641. (b) Rendina,
L. M.; Puddephatt, R. J. Chem. ReV. 1997, 97, 1735. (c) von Zelewsky,
A.; Suckling, A. P.; Stoeckli-Evans, H. Inorg. Chem. 1993, 32, 4585.
(62) The homocoupling of two alkyl radicals (Me · in cyclohexane at 298
9
-1 -1
K: k ) 1.6 × 10 M
s , ref a) is essentially diffusion-limited. The
rate constant for the homocoupling of two MeOO· radicals in benzene
8
-1 -1
at 295 K is k ) 1.9 × 10 M
s
(ref b). (a) Carlsson, D. J.; Ingold,
(d) Sandrini, D.; Maestri, M.; Balzani, V.; Chassot, L.; von Zelewsky,
K. U. J. Am. Chem. Soc. 1968, 90, 7047. (b) Khursan, S. L.; Safiullin,
R. L.; Martemianov, V. S.; Nikolayev, A. V.; Urozhai, I. A. React.
Kinet. Catal. Lett. 1989, 39, 261. For a compilation of rate data, see
ref 61b.
A. J. Am. Chem. Soc. 1987, 109, 7720. (e) Hall, T. L.; Lappert, M. F.;
Lednor, P. W. J. Chem. Soc., Dalton Trans. 1980, 1448. For relevant
discussions on these types of reactions, see refs 45b, c.
1
5810 J. AM. CHEM. SOC. 9 VOL. 131, NO. 43, 2009

Insertion of O
2
into Pd(II)-Me
A R T I C L E S
Scheme 7
6
7,68
and co-workers to explain the inverse half-order dependence
of the rate on the concentration of the inhibitor in some radical
radical chain mechanisms.
Experimental and computational
investigations on these reactions support a mechanism proceed-
ing, depending on the complex, by either hydrogen atom
2
and recombination to form PdOOH or by a
pathway involving reductive elimination and reaction of Pd(0)
6
6
chain autoxidation reactions.
The rate law describing the rate of disappearance of (bipy)-
PdMe (1) according to Scheme 7 is given by eq 8 (see
6
7
abstraction by O
2
6
8
Supporting Information for a derivation).
2
with O followed by reprotonation.
Significantly however, the insertion of molecular oxygen into
Me2
Me2
rate ) -d[(bipy)PdMe ]/dt )
the Pt(IV)-H bond of Tp PtMe
H (Tp ) hydridotris(3,5-
2
2
1
/2
1/2
dimethylpyrazolyl)borate) (9) has been shown to proceed by a
kScheme7[(bipy)PdMe ] [AIBN]
(8)
(9)
2
31a,69
radical chain mechanism.
It is notable that although the
Pt(IV) complex 9 contains both methyl and hydride ligands,
2
1/2
kScheme7 ) (2ek k k /(k (k + k ))
1
5i 5ii
6e -5i
5ii
(
63) No chemically induced dynamic nuclear polarization (CIDNP) effects
were observed during the reactions between 1 and O . Weak CIDNP
Notably, eq 8 shows a half-order dependence of the rate on the
concentration of both (bipy)PdMe (1) and AIBN and is of the
same form as the empirical rate law determined experimentally
for the reaction of 1 with O (eq 4). The involvement of the five-
coordinate Pd(III) intermediate (bipy)Pd (Me)
2
effects have been observed in some non-chain homolytic displacements
at Pt(II) centers (see ref 54b, see also ref 42h) while no effects were
detected for some radical chain reactions (see ref 55, also ref 57a).
For the observation of strong CIDNP effects in the autoxidation of
boranes, see ref 15q.
2
2
III
2
(OOMe)· (8),
(
64) Apart from the processes described in note 60, rate constants for the
addition of radicals to transition metal complexes have rarely been
formed by attack of MeOO · on 1, thus provides a simple
explanation to the unusual kinetics determined experimentally. This
key intermediate 8 is involved both as a chain carrier to form the
product (bipy)PdMe(OOMe) (2) by release of Me · radicals
i
measured. (a) The rate constant for the addition of Pr· radicals to
(
phen)PtMe (ref 57b) and to trans-IrClCO(PMe ) was estimated to
2
3
2
7
-1 -1
be k ≈ 10 M
s : Labinger, J. A.; Osborn, J. A.; Coville, N. J.
Inorg. Chem. 1980, 19, 3236. (b) In ref 41g, the rate constant for the
addition of the 5-hexenyl radical to Pt(PEt ) was estimated to be k ≈
(Scheme 7, step 5ii), and as a trap for MeOO · radicals to terminate
3
3
6
-1 -1
II 2+
1
H
0 M
s . (c) In ref 45a, for the equilibrium [(L)Ni ] + Me· +
III 2+
the chain process (Scheme 7, step 6e). The deviation from the usual
kinetic behavior found for the autoxidation of organics, where a
first-order dependence of the rate on the concentration of the
2 2
O T [(L)(H O)Ni (Me)] , values of the rate constant for the
forward reaction with three different complexes were determined to
8
-1 -1
be kforward ) 1.6-6.5 × 10 M
s , and the equilibrium constant for
43,44
-1
substrate is typically found,
can thus be easily explained by
five different complexes were found to be Keq ) 450-12000 M
d) Both the forward and reverse reactions involving the Lewis acid-
.
(
the formation of this hypervalent intermediate which is not available
to organic compounds. In that respect, the radical chain mechanism
II
base adduct [(acetylacetonato)
2
Pt · · ·I
2
] for the equilibrium reaction
IV
II
I · + [(acetylacetonato)
2
Pt · · · I
2
] T (acetylacetonato) Pt (I) + I ·
2 2
determined for the autoxidation of (bipy)PdMe
2
(1) seems to be
were found to be diffusion-controlled, with an equilibrium constant
of Keq ) 0.694: Hopgood, D.; Jenkins, R. A. J. Am. Chem. Soc. 1973,
more closely related to the one found for the autoxidation of main
group and early transition metal alkyl complexes for which stepwise
homolytic substitutions at the main group or metal center have
9
5, 4461.
(
(
65) See note 45 for similar reactivity in nonchain radical reactions with
II
transition metal complexes. The two-step transformation LCo
+
14-16
III
III
often been proposed.
Comparison to Known Examples of Insertion of O
MeOO· f LCo (OOMe) followed by LCo (OOMe) + MeOO· f
products has been studied, see refs 60f, g.
2
into
66) Hammond, G. S.; Boozer, C. E.; Hamilton, C. E.; Sen, J. N. J. Am.
Chem. Soc. 1955, 77, 3238.
Late Metal-Hydride Bonds and Implications for Catalysis. In
sharp contrast to the radical chain nature of the reaction of
(67) (a) Denney, M. C.; Smythe, N. A.; Cetto, K. L.; Kemp, R. A.;
Goldberg, K. I. J. Am. Chem. Soc. 2006, 128, 2508. (b) Keith, J. M.;
Muller, R. P.; Kemp, R. A.; Goldberg, K. I.; Goddard, W. A., III;
Oxgaard, J. Inorg. Chem. 2006, 45, 9631. (c) Chowdhury, S.; Rivalta,
I.; Russo, N.; Sicilia, E. Chem. Phys. Lett. 2007, 443, 183.
insertion of O
known examples of insertion of O
Pd(II)-OOH complexes have been shown not to proceed by
2
into a Pd(II)-Me bond of (bipy)PdMe
2
(1), the
into Pd(II)-H bonds to form
2
J. AM. CHEM. SOC. 9 VOL. 131, NO. 43, 2009 15811

A R T I C L E S
Boisvert et al.
only insertion into the Pt(IV)-H bond was observed (eq 10).
The mechanism proposed for the autoxidation of 9 is reminiscent
of the one usually encountered in the autoxidation of organic
compounds, with the key chain carriers formed through H-atom
guide the successful design of transition metal catalysts for
alkane oxidation using molecular oxygen.
Conclusions
3
1a,69
abstraction reactions.
A rare example of insertion of molecular oxygen into a late
transition metal-alkyl bond has been described and the mech-
anism of the reaction investigated through detailed kinetic
2
studies. The autoxidation of (bipy)PdMe (1) results in the
formation of the Pd(II) methylperoxide complex (bipy)Pd-
Me(OOMe) (2). Although a variety of main group and early
transition metal alkyl complexes are known to react with O
form alkylperoxide complexes,
2
to
this type of autoxidation
reaction has not been well established for late transition
1
4-16
1
0-12
metals.
As numerous complexes of late transition metals
2
The mechanism of the autoxidation of (bipy)PdMe (1), which
have been shown to directly activate alkanes to form metal
alkyls, this new mode of reactivity may constitute a promising
lead for a selective alkane functionalization process.
Reproducible rates were achieved in the presence of AIBN,
and the reaction was determined to be one-half-order in both
involves a radical chain process proceeding by stepwise
homolytic substitutions at the metal center, stands in contrast
to the documented mechanisms of autoxidation of late transition
metal hydrides. Given that we are only now beginning to find
well-characterized examples of the insertion of molecular
2
[1] and [AIBN], and zero-order in [O ]. The marked effects of
10-12
oxygen into late transition metal-alkyl bonds,
the general-
light and AIBN on the rate of the reaction suggest the
involvement of a radical chain mechanism. This result is
particularly striking considering that the recently documented
ity of this type of radical chain mechanism remains to be
7
0
determined. However, the recognition that such a mechanism
can be operative has some important implications. It will be
particularly significant to establish whether autoxidation of late
transition metal alkyl complexes can occur at saturated metal
centers, such as in complex 9, that are not able to expand their
valency to accommodate a mechanism such as the one proposed
in Scheme 7. Notably, another rare example of insertion of
molecular oxygen into a late metal-alkyl bond was recently
insertions of O
proceed by radical chain mechanisms.
2
into Pd(II)-H bonds have been shown not to
67,68
On the basis of the
observed kinetic data and on literature precedents, stepwise
homolytic substitution reactions at the metal center of (bipy)Pd-
Me (1) are proposed as key steps in the autoxidation of 1. A
2
salient feature of this scheme is the formation of a five-
coordinate Pd(III) intermediate by expansion of the valency of
the metal center. Overall, the mechanism for the autoxidation
of 1 appears to be more closely related to the general mechanism
proposed for the autoxidation of main group and early transition
metal alkyl complexes than to the general mechanism usually
involved in the autoxidation of organic compounds and recently
proposed for the autoxidation of Pt(IV)-H complexes. Con-
sidering that a wide variety of useful processes based on the
autoxidation of main group metal alkyl complexes have been
1
1
discovered in our laboratory. The insertion of O
2
into a
Pt(II)-Me bond of (PN)PtMe (PN ) 2-(di-tert-butylphosphino-
2
methyl)pyridine) (11) to form (PN)PtMe(OOMe) was found to
proceed readily at room temperature (eq 11).
1
7-20
developed,
development of “palladium oxygenase” catalytic transformations.
Understanding the mechanisms by which O reacts with
such behavior is particularly promising for the
2
transition metal complexes may facilitate the development of a
variety of aerobic oxidation reactions. While associative ho-
molytic substitutions such as those involved in the autoxidation
of 1 have not been systematically studied with late transition
metals, they could potentially be applicable to a wide variety
of transformations. A greater awareness of the requirements for
these types of reactions, such as the necessity to expand the
valency at the metal center, may help in the rational design of
catalysts for the functionalization of alkanes.
Although the mechanism of this reaction has not been fully
studied, preliminary observations suggest the involvement of a
1
1
radical chain mechanism. The fact that both (bipy)PdMe
2
(1)
8
and (PN)PtMe
2
(11) are sixteen-electron d square planar
complexes with the ability to expand their valency to five-
coordinate seventeen-electron species may be an important
feature that allows them to undergo this insertion reaction.
Clearly, further studies of the reaction of molecular oxygen with
both saturated and unsaturated metal alkyl complexes will
deepen our understanding of the mechanisms and requirements
for oxygen insertion into late metal-alkyl bonds and potentially
Experimental Section
Materials and Methods. Unless otherwise noted, all manipula-
tions were carried out using high-vacuum line techniques, under
an atmosphere of dry dinitrogen using standard Schlenk techniques
or in an inert atmosphere glovebox. Glassware was dried for a
minimum of eight hours at >150 °C. Benzene, dichloromethane,
diethyl ether, tetrahydrofuran and pentane were dried by passage
through activated alumina and molecular sieves columns under a
stream of argon. Methanol was freshly distilled from magnesium
(
68) (a) Konnick, M. M.; Gandhi, B. A.; Guzei, I. A.; Stahl, S. S. Angew.
Chem., Int. Ed. 2006, 45, 2904. (b) Popp, B. V.; Stahl, S. S. J. Am.
Chem. Soc. 2007, 129, 4410. (c) Konnick, M. M.; Stahl, S. S. J. Am.
Chem. Soc. 2008, 130, 5753. (d) Chowdhury, S.; Rivalta, I.; Russo,
N.; Sicilia, E. J. Chem. Theory Comput. 2008, 4, 1283. (e) Keith,
J. M.; Goddard, W. A., III J. Am. Chem. Soc. 2009, 131, 1416.
69) Wick, D. D.; Goldberg, K. I. J. Am. Chem. Soc. 1999, 121, 11900.
70) For example, singlet oxygen generated by photosensitization was
implicated in a recently reported example of oxygen insertion into a
Pt(II)-Me bond to form a Pt(II)-OOMe complex. See ref 12.
turnings and iodine. Benzene-d
6
and 1,4-dioxane were vacuum-
(
(
transferred from sodium benzophenone ketyl. Dichloromethane-d
2
and chloroform-d were vacuum-transferred from calcium hydride.
All other reagents were used as received from commercial suppliers.
1
5812 J. AM. CHEM. SOC. 9 VOL. 131, NO. 43, 2009

Insertion of O
bipy)PdMe
2
into Pd(II)-Me
A R T I C L E S
2
3b
(
2
(1) was prepared according to a literature procedure.
in CDCl
by H NMR spectroscopy.
3
: immediate formation of Ph
3
PO and MeOH was observed
1
27
NMR experiments were performed on Bruker AV300, AV301,
DRX499 and AV500 spectrometers using, unless otherwise noted,
Synthesis of (bipy)PdMe(OOMe) (2). Methylperoxide complex
2 could be efficiently generated from either methoxide complex 3
or hydroxide complex 4; the protocol outlined below is illustrated
for a reaction involving 3.
5
mm medium-walled Wilmad 504-PP NMR tubes fitted with a J.
1
Young Teflon valve. H NMR spectra are referenced to residual
protiated solvent signals (CDCl : 7.26 ppm, C : 7.16 ppm,
CD Cl
: 5.32 ppm). 13C NMR spectra are referenced to deuterated
solvent signals (CD Cl : 54.0 ppm). Chemical shifts are reported
3
6 6
D
2
2
Under an atmosphere of dinitrogen, (bipy)PdMe(OMe) 3 (1.3
mg, 4.2 µmol) was weighed into a vial and toluene (0.5 mL) was
2
2
in parts per million (ppm) downfield from tetramethylsilane (0.00
ppm) and coupling constants are reported in hertz (Hz). The
following abbreviations are used: s (singlet), d (doublet), dd (doublet
of doublets), ddd (doublet of doublet of doublets), m (multiplet),
added. A solution of ∼1:2 MeOOH/Et O (1.0 µL; as prepared
2
above) was added by syringe at room temperature and the yellow-
orange solution immediately became yellow. The vial was capped,
shaken for <30 s and stored in the freezer at -35 °C for ca. 11 h.
Pentane (0.5 mL) was added and the sample was allowed to stand
at -35 °C for 2 min. The colorless supernatant was removed at
-35 °C, the yellow solid was washed with pentane (4 × 0.5 mL),
and traces of solvent were quickly removed under vacuum. Complex
2 was unstable in solution at room temperature and was stored as
1
br (broad). Assignments of the H NMR resonances were made
from COSY, NOESY and HMQC spectra of the complexes. The
numbering scheme for the protons on the bipy ligand is illustrated
in Table 1.
Safety Note on the Handling of Alkylperoxides. Pure alkyl-
peroxides are potentially explosiVe and should be aVoided if
1
a solid under an atmosphere of dinitrogen at -35 °C. H NMR
2
6
possible. They can be violently decomposed by adventitious
catalysts (acids, metals) with formation of, among others, oxygen
gas. Distillation of alkylperoxides is not recommended. It has
been documented that explosions occurred when pure MeOOH
2 2
(CD Cl , 500 MHz, 233 K) δ (ppm) 8.96 (d, 1H, J ) 4.4 Hz, H6′),
8.59 (d, 1H, J ) 5.2 Hz, H
7
7
6
), 8.09-7.98 (m, 4H, H
.57 (ddd, 1H, J ) 7.3, 5.2, and 1.2 Hz, H5′), 7.47 (ddd, 1H, J )
.5, 5.6, and 1.3 Hz, H ), 3.64 (s, 3H, Pd–OOCH ), 0.67 (s, 3H,
). HMQC analysis (CD Cl , 500 MHz, 233 K): PdOOCH
3
H NMR: 3.64 ppm, C NMR: 63.8 ppm), PdCH ( H NMR: 0.65
3 4
, H3′, H , H4′),
5
3
2
5a
Pd-CH
3
2
2
3
was heated
pure samples of MeOOH exploded violently.
Synthesis of MeOOH. By modification of literature protocols:
and that the residue obtained after distillation of
1
13
1
25b
(
1
3
2
5
ppm, C NMR: 3.1 ppm). Alternatively, complex 2 was formed
quantitatively, together with one molar equivalent of ROH (R )
6 6 2 2
Me or H), when MeOOH was added to a C D or CD Cl solution
To avoid metal-catalyzed decomposition of the peroxides, all
manipulations were carried out in carefully washed glassware. In
a well-ventilated fume hood, Me
added to a mixture of H O (50 mL) and H
mL) at 0 °C. A solution of KOH (16.8 g, 299 mmol) in H
mL) was then added dropwise with vigorous stirring to the biphasic
reaction mixture over 30-40 min at 0 °C. On addition of the KOH
solution, the formation of the byproduct MeOOMe was immediately
apparent from the liberation of a gas. At the end of the addition,
bubbling was still intense and a single phase was visible. After
stirring a further 40 min at 0 °C (at which point the bubbling was
much less intense), the mixture was warmed to room temperature
of (bipy)PdMe(OMe) (3) or (bipy)PdMe(OH) (4), as observed by
2
SO
4
(15 mL, 159 mmol) was
(30 wt % in H O, 30
O (25
1
H NMR spectroscopy.
2
2
O
2
2
Synthesis of (bipy)PdMe(OMe) (3). Under an atmosphere of
2
dinitrogen, (bipy)PdMe(I) (35.6 mg, 0.0880 mmol) and AgSO
3 3
CF
(
23.7 mg, 0.0924 mmol) were combined in CH OH (4 mL). The
3
reaction mixture was stirred for 1-2 min to allow the formation
of a pale gray precipitate, then NaOMe (9.5 mg, 0.18 mmol) was
added. The mixture was filtered through a PTFE syringe filter, and
the solvent immediately removed under vacuum. The yellow residue
was extracted with CH
2
Cl
2
(2 × 1 mL) and filtered through glass
wool/Celite. Addition of pentane (6 mL) resulted in precipitation
of a pale yellow solid. The solid was washed with pentane (2 × 3
mL) and dried under vacuum. Yield: 20.5 mg (75%). Complex 3
was unstable in solution at room temperature and was stored as a
(
bubbling intensified) and stirred at room temperature until the
liberation of gas was much less intense (ca. 4.5 h). After stirring at
6
0 °C for 45 min, the bubbling had essentially stopped and the
mixture was cooled to room temperature and stirred for ca. 10 h.
1
solid under an atmosphere of dinitrogen at -35 °C. H NMR
The mixture was cooled to 0 °C and neutralized to pH ≈ 7 with
(
2 2
CD Cl , 300 MHz, 298 K) δ (ppm) 8.95 (d, 1H, J ) 4.8 Hz, H6′),
5
2 4
0% H SO . A final volume of ca. 120 mL was obtained. This
8
7
7
.68 (d, 1H, J ) 5.7 Hz, H
.60 (ddd, 1H, J ) 5.7, 4.8, and 2.5 Hz, H5′), 7.43 (ddd, 1H, J )
.5, 5.7, and 3.0 Hz, H ), 3.55 (s, 3H, Pd–OCH ), 0.77 (s, 3H,
Cl , 500 MHz, 233 K) δ (ppm) 8.85 (d,
H, J ) 4.8 Hz, H6′), 8.63 (d, 1H, J ) 5.3 Hz, H ), 8.11-7.94 (m,
H, H , H3′, H , H4′), 7.59 (t, 1H, J ) 5.7 Hz, H5′), 7.45 (t, 1H, J
6.3 Hz, H
NMR data in CD
Koten and co-workers. HMQC analysis of a 1: 1 (bipy)Pd-
Me(OMe) (3)/(bipy)PdMe(OH) (4) mixture (CD Cl , 500 MHz, 233
) ( H NMR: 3.45 ppm, C NMR: 57.4 ppm),
6
), 8.11-7.94 (m, 4H, H
3 4
, H3′, H , H4′),
solution was then continually extracted for 7 h with ca. 25 mL of
1
Et
CDCl
continuous extraction (ca. 25 mL of Et
result. The Et O solution was dried overnight over Na
2
O. Analysis of the Et
, 298 K) revealed a ratio of ∼1:9 MeOOH/Et
O, 12 h) yielded a similar
SO at 4 °C
2
O solution by H NMR spectroscopy
5
3
(
3
2
O. A second
1
Pd-CH
3
). H NMR (CD
2
2
2
1
4
6
2
2
4
3
4
then filtered through glass wool in a wide-bore vial containing a
magnetic stir bar. A thermometer was placed above the opening of
the vial, and the solution was concentrated by heating with an oil
bath with stirring at 40-45 °C until the temperature above the
opening of the vial decreased from 33-37 °C to ca. 30 °C (the
1
)
5
), 3.45 (s, 3H, Pd–OCH
3
), 0.69 (s, 3H, Pd-CH
3
). H
3
OD matched that previously reported by van
24
2
2
1
13
13
K): (bipy)PdMe(OCH
3
1
2
5a
boiling point of pure MeOOH is 38-40 °C at 65 mmHg).
(
(
bipy)PdCH
bipy)PdCH
3
(OMe) ( H NMR: 0.68 ppm, C NMR: 1.4 ppm);
1
Analysis of the residue by H NMR spectroscopy (CDCl
revealed a ratio of ∼1:2 MeOOH/Et O. This mixture was dried
for 1 h over CuSO at 4 °C then filtered through glass wool in a
vial. It was kept in the freezer of the inert atmosphere glovebox at
3
, 298 K)
1
13
3
(OH) ( H NMR: 0.53 ppm, C NMR: -4.2 ppm).
2
Synthesis of (bipy)PdMe(OH) (4). Under an atmosphere of
4
dinitrogen, (bipy)PdMe(OMe) 3 (1.3 mg, 4.2 µmol) was weighed
into a medium-walled NMR tube fitted with a J. Young Teflon
valve. A small amount of hexamethylbenzene (as an internal
standard) and benzene (0.4 mL) was added, and the tube was shaken
to obtain a yellow-orange solution. H O (0.4 µL, 22 µmol) was
added, and the tube was shaken for ca. 2 min until a colorless
1
-
35 °C and used as such for the subsequent reactions. H NMR
O mixture (CDCl , 300 MHz,
OOH), 3.88 (s, 3H, CH OOH),
CH O), 1.20 (t, 6H, J ) 7.0 Hz,
O); lit., (pure MeOOH in CDCl ) δ (ppm) 8.86 (br s,
OOH), 3.87 (s, 3H, CH OOH). Apart from a trace amount
analysis of the ∼1: 2 MeOOH/Et
2
3
2
3
98 K): δ (ppm) 8.72 (s, 1H, CH
.48 (q, 4H, J ) 7.0 Hz, (CH
3
3
2
3
2
)
2
2
5a
(
CH
3
CH
2
)
2
3
supernatant and a beige solid were obtained. The solvent was
removed under vacuum, CD Cl was added by vacuum-transfer,
1H, CH
3
3
2
2
1
1
of water, only MeOOH and Et
spectroscopy; no MeOH or H
storing the sample for several months at -35 °C. The identity of
MeOOH was confirmed by addition of Ph P to a sample of MeOOH
2
O were detectable by H NMR
O could be detected even after
and the sample was kept at -78 °C. Analysis by H NMR
spectroscopy showed complete conversion to 4 with >90% purity.
Complex 4 was unstable in solution at room temperature but
solutions kept at temperatures lower than ca. -35 °C showed no
2
2
3
J. AM. CHEM. SOC. 9 VOL. 131, NO. 43, 2009 15813

A R T I C L E S
Boisvert et al.
1
sign of decomposition. H NMR (CD
ppm) 8.96 (d, 1H, J ) 5.6 Hz, H6′), 8.65 (d, 1H, J ) 5.7 Hz, H
.09-7.96 (m, 4H, H , H3′, H , H4′), 7.60 (ddd, 1H, J ) 6.7, 5.6,
), 0.59
), -1.50 (br s, 1H, Pd-OH). H NMR (CD Cl
00 MHz, 233 K) δ (ppm) 8.87 (d, 1H, J ) 5.6 Hz, H6′), 8.57 (d,
H, J ) 5.7 Hz, H ), 8.08-7.96 (m, 4H, H , H3′, H , H4′), 7.59 (m,
5′), 7.45 (m, H ), 0.53 (s, 3H, Pd-CH ), -1.39 (s, 1H, Pd-OH).
NOESY analysis (CD Cl , 500 MHz, 233 K) showed chemical
exchange between Pd-OH and trace H
O (∼3 ppm) (see Synthesis
of (bipy)PdMe(OMe) for HMQC data of the methyl group of
2
Cl
2
, 300 MHz, 298 K) δ
(2) A stock solution of (bipy)PdMe
2
1 and 1,4-dioxane (as
(
8
6
),
internal standard) in C D was added to two medium-walled NMR
6 6
3
4
tubes fitted with J. Young Teflon valves. To one of the tubes was
1
and 2.2 Hz, H5′), 7.43 (ddd, 1H, J ) 7.4, 5.7, and 1.7 Hz, H
5
added AIBN. Reference H NMR spectra were then recorded at
1
(
5
1
s, 3H, Pd-CH
3
2
2
,
298 K. Both tubes were degassed (two freeze-pump-thaw cycles)
and kept frozen in an ice/methanol bath. The tubes were pressurized
with oxygen (5 atm) using a high-pressure manifold. The samples
were thawed, shaken, and heated at 50 °C in a temperature-
6
3
4
H
5
3
1
2
2
controlled oil bath. The extent of reaction was monitored by H
2
NMR spectroscopy. For the reaction with no added AIBN, 80% of
1 was still present after 40 min, 31% after 160 min. All of 1 was
consumed after 40 min for the reaction with added AIBN.
Kinetic Studies. In a typical experiment, AIBN (0-1.5 mg,
0-9.1 µmol) was weighed into a medium-walled NMR tube fitted
(
bipy)PdMe(OH) 4).
Synthesis of (bipy)PdMe(I) (5). Under an atmosphere of
dinitrogen, (bipy)PdMe 1 (84.0 mg, 0.287 mmol) was dissolved
2
in acetone (20 mL). Upon addition of iodomethane (19 µL, 0.30
mmol), the solution turned a lighter yellow color and the reaction
was stirred for 30 min at room temperature. The volume was then
reduced under vacuum to ca. 5 mL, resulting in the precipitation
of a pale yellow solid. After addition of hexanes (10 mL), the yellow
solid was collected on filter paper, washed with hexanes (2 × 3
with a J. Young Teflon valve. A stock solution of (bipy)PdMe 1
2
(5.3 mM), AIBN (2.2-8.2 mM) and 1,4-dioxane (as internal
standard) (3.5-7.0 mM) in C D (0.3 mL) was added. A reference
6
6
1
H NMR spectrum was then recorded in a NMR probe preheated
to 323 K. The sample was degassed (three freeze-pump-thaw
cycles) and kept frozen in an ice/methanol bath. The tube was
pressurized with oxygen (5-10 atm) using a high-pressure mani-
fold. The sample was thawed, shaken, and placed into a NMR probe
1
mL), and dried under vacuum. Yield: 100 mg (86%). H NMR
6
data for (bipy)PdMe(I) (acetone-d ) matched that previously
reported by Byers and Canty.
2
3a
1
preheated to 323 K. H NMR spectra were acquired using single
SAFETY NOTE on the Handling of Pressurized NMR
Tubes. Extreme caution should be used while handling J. Young
NMR tubes under pressure. While under pressure, NMR tubes were
stored in a protectiVe jacket.
scans separated by at least 60 s in order to ensure accurate
integration. Spectra were collected over at least three half-lives.
The total concentration of AIBN, determined by integration of its
CH
proximately constant throughout the reaction and is reported as the
3
signal relative to the 1,4-dioxane signal, remained ap-
Control Experiment: (bipy)PdMe
of Oxygen. Under an atmosphere of dinitrogen, (bipy)PdMe
.1 mg) was weighed into a medium-walled NMR tube fitted with
a ground glass joint and 1,3,5-tri-tert-butylbenzene was added as
an internal standard. C was added by vacuum transfer and the
2
in C
6
D
6
in the Absence
3
7
1 (ca.
average value determined over at least two half-lives. Rates
(defined as rate ) -d[(bipy)PdMe ]/dt) were determined by
2
0
2
following the disappearance of the bipyridine signal of 1 at 8.8
ppm compared with the 1,4-dioxane signal. Similar results were
obtained when the rates were determined by following the disap-
pearance of the methyl signal of 1. Concentrations of 2 found in
Figure 1 were determined by following the appearance of the OOMe
signal of 2 compared with the 1,4-dioxane signal.
6 6
D
sample was kept frozen in a liquid nitrogen bath. The tube was
flame-sealed under vacuum. The sample was thawed, shaken, and
allowed to react at room temperature. No reaction was observed
1
by H NMR spectroscopy over 24 h. The sample was heated at 50
°
C in a temperature-controlled oil bath for over 18 h. Although
Under these conditions, maximum yields of ca. 85% of 2 were
observed after complete consumption of 1. In all reactions, complete
consumption of 1 was observed and extents of conversion reported
in the different plots refer to the maximum conversion that could
traces of palladium black were visible on the sides of the NMR
1
tube, only 1 was observed by H NMR spectroscopy in >95% of
its initial concentration.
1
Verification of the Effect of Light. A stock solution of
be reliably determined by integration of the signals in the H NMR
1
(
bipy)PdMe
2
1 and 1,4-dioxane (as internal standard) in C
6
D
6
was
spectra. H NMR (signals for 2 in crude reaction mixture, C D ,
6
6
added to two medium-walled NMR tubes fitted with J. Young
Teflon valves. Both tubes were degassed (three freeze-pump-
thaw cycles) and kept frozen in an ice/methanol bath. The tubes
were pressurized with oxygen (5 atm) using a high-pressure
manifold. The samples were thawed, shaken and allowed to react
at room temperature. One tube was kept under room light and the
500 MHz, 298 K) δ (ppm) 9.37 (d, 1H, J ) 3.8 Hz, bipy), 8.36 (d,
1H, J ) 4.7 Hz, bipy), 7.00-6.84 (m, 4H, bipy), 6.61 (m, 1H,
bipy), 6.32 (m, 1H, bipy) 4.26 (s, 3H, Pd–OOCH ), 1.48 (s, 3H,
3
1
Pd-CH ). The H NMR chemical shifts of complexes 2 and 3 in
C D varied with the presence of protic impurities. However, the
3
6
6
addition of independently synthesized methylperoxide complex 2
and methoxide complex 3 to the crude reaction mixture obtained
above showed that the product of the reaction between 1 and oxygen
was complex 2. To fully characterize the product of the reaction
of 1 with oxygen by NMR spectroscopy at low temperature, it was
1
other kept in the dark. The extent of reaction was monitored by H
NMR spectroscopy over three days. For the reaction kept under
light, 80%, 58% and 0% of 1 remained after 3, 6, and 72 h,
respectively. For the reaction kept in the dark, 97%, 93% and 74%
of 1 remained after 3, 6, and 72 h, respectively.
2 2
preferable to remove the volatiles under vacuum and add CD Cl
1
Verification of the Effect of AIBN. (1) In a typical experiment,
by vacuum-transfer. H NMR and HMQC analyses of the resulting
(
bipy)PdMe
NMR tube fitted with a J. Young Teflon valve under an atmosphere
of dinitrogen. C
2
1 (ca. 0.1 mg) was weighed into a medium-walled
mixture at 233 K showed the signals for 2 (see Synthesis of
(bipy)PdMe(OOMe) (2) above), the identity of which could be further
6
6
D and hexamethylbenzene (1-2 crystals) or 1,4-
confirmed by the addition of independently synthesized 2 and 3.
dioxane (ca. 0.2 µL) (as internal standard) were added. The sample
Acknowledgment. This work was supported by the National
Science Foundation (Grants CHE-0137394 and CHE-0719372).
L.B. is the recipient of a postdoctoral fellowship from Fonds
qu e´ b e´ cois de la recherche sur la nature et les technologies.
was degassed (three freeze-pump-thaw cycles) and the tube was
pressurized with O (1-10 atm) using a high-pressure manifold.
2
The sample was thawed, shaken and allowed to react at room
temperature or at 50 °C in a temperature-controlled oil bath and
1
the reaction was monitored by H NMR spectroscopy. Under these
conditions, the rate of consumption of 1 was found to be highly
variable and maximum yields of ca. 70% of 2 were obtained. When
AIBN was present and the reaction was conducted at 50 °C in a
temperature-controlled oil bath, complete conversion of 1 was
achieved and up to ca. 85% of 2 could be obtained.
Supporting Information Available: Additional kinetic plots
and derivation of eq 8. This information is available free of
charge via the Internet at http://pubs.acs.org.
JA9061932
1
5814 J. AM. CHEM. SOC. 9 VOL. 131, NO. 43, 2009