C O MMU N I C A T I O N S
Scheme 2
galvinoxyl (Table 1, and reactions with [PdHCl(PPh
is clearly unrelated to the strength of the T-H bond that is formed,
3 2
) ]). This order
10
but consistent with a kinetic control associated mainly with the
different steric protection of these radical traps. (2) The reaction
rate increases with the electron density on the metal ([PdHCl(PPh
3
)
2
]
+
>
[PdH(PEt )
3 3
] ). (3) Alkyl and aryl complexes are unaffected. If
the reaction with hydrides was associated with H‚ being formed
by reversible bond homolysis (Scheme 2), the radical traps should
react as well with the methyl or phenyl derivatives. Because this is
not the case, it seems unlikely that radicals are formed in these
complexes, either for the Me, for the Ph, or for the hydrido
complexes.
Figure 1. Disappearance of 1-hexene (to give a mixture of 2- and
-hexene), catalyzed by 1, in the absence and presence of radical traps.
3
A mechanism that can explain the observations is the direct attack
of the radical on the coordinated H, leading to polarization of the
electron density and eventual homolysis of the Pd-H bond (eq 4),
as proposed for H-abstraction in organic systems,1 and for H‚
transfer from metal hydrides to the substituted trityl radical (p-
efficient catalyst for olefin isomerization. The transformations
collected in eqs 1-3 were studied.
1
12
6 4 3
tButC H ) C‚. The success of this interaction depends on sufficient
orbital overlap and Pd-H bond polarization. The more polarizable
and diffuse hydride orbitals of neutral complexes should interact
better than the more compact and less polarizable orbitals of a
cationic complex. For the same reason, a less hindered hydride and
a less sterically protected radical trap should react faster. Finally,
the reaction with more hindered alkyls and aryls, needing orbital
reorganization, should be disfavored.
The isomerization of the internal olefin 4 to the conjugated alkene
5
3
in CDCl proceeds with 61% conversion after 1 h at room
temperature in the presence of 0.1 mol % of catalyst 1 (eq 1). This
is close to completion, because the equilibrium mixture was found
to be 4:5 ) 33:67. The presence of galvinoxyl in a similar reaction
mixture (Pd:galvinoxyl ) 1:1) slowed the isomerization down, and
only 34% conversion to 5 was observed after 1 h. When DPPH
was used as additive (Pd:DPPH ) 1:1), only 2% conversion was
observed after 1 h, and no further isomerization occurred in 5 h.
In another test, the isomerization of terminal to internal alkenes
In summary, classical transition metal hydride catalyzed reactions
can be perturbed and arrested by radical traps, and positive test
results can be misleading.
Acknowledgment. We thank the DGI (BQU2001-2015), the
JCyL (VA17/00B), and Iberdrola for financial support.
(eqs 2 and 3) catalyzed by 1 turned out to be instantaneous at room
temperature. The isomerization of 1-hexene in CDCl to give an
3
equilibrium mixture of 2-hexene and 3-hexene was followed at 263
References
K using 0.1 mol % of catalyst. The observed rate constant is kisom
(
1) Sawamoto, M.; Kamigaito, M. J. Macromol. Sci., Pure Appl. Chem. 1997,
-
1
)
0.0266 ( 0.0003 s . The same isomerization was carried out
A34, 1803-1814.
in the presence of radical traps (Pd:trap ) 1:1), and the disappear-
ance of 1-hexene versus time is represented in Figure 1. Galvinoxyl
produces a modest but perceptible decrease of the reaction rate (kisom
(2) Matyjaszewski, K. Macromolecules 1998, 31, 4710-4717.
(
3) Tsuji, J. Palladium Reagents and Catalysts. InnoVations in Organic
Synthesis; Wiley: Chichester, 1995.
4) Tanase, T.; Ohizumi, T.; Kobayashi, K.; Yamamoto, Y. Organometallics
1996, 15, 3404-3411.
(
-
1
)
0.0225 ( 0.0004 s ). With DPPH the initial isomerization rate
-
1
drops dramatically (kisom ) 0.0076 ( 0.0004 s ), and the reaction
is almost halted after 20-30 min; after 1 h only 19% conversion
is observed, versus 90% without additive, or 79% with galvinoxyl.
The experiments reported here demonstrate that radical traps can
react with some palladium hydrides, halting reactions that follow
an insertion mechanism involving Pd-H bonds. In other words,
some of these additives can be efficient palladium hydride traps as
well. The question arises whether they will react also with Pd-
alkyl or Pd-aryl bonds. To check this point, we tested the
(5) Schunn, R. A. Inorg. Chem. 1976, 15, 208-212.
6) NMR tubes were charged with the palladium hydride, radical trap (molar
ratio Pd:radical trap ) 1:1), and CDCl (0.6 mL). The freshly prepared
(
3
samples were monitored every 10 min by 31P and
H NMR.
(7) Alb e´ niz, A. C.; Espinet, P.; Lin, Y.-S. Organometallics 1997, 16, 4030-
032.
8) [PdMeCl(PPh
COD)] (Rulke, R. E.; Ernsting, J. M.; Spek, A. L.; Elsevier, C. J.; van
1
4
(
3 2 3
) ] was prepared by addition of 2 equiv of PPh to [PdMeCl-
(
Leeuwen, P. W. N. M.; Vrieze, K. Inorg. Chem. 1993, 32, 5769-5778).
9) Coulson, D. R. Chem. Commun. 1968, 1530.
(
(
10) (a) Mahoney, L. R.; Mendenhall, G. D.; Ingold, K. U. J. Am. Chem. Soc.
1
973, 95, 8610-8614. (b) Lucarini, M.; Pedulli, G. F.; Cipollone, M. J.
8
9
compounds [PdMeCl(PPh
traps TEMPO and galvinoxyl under the same conditions that had
produced decomposition for [PdHCl(PPh ]. After 7 h, the solutions
3
)
2
] and [PdPhCl(PPh
3
)
2
] with the radical
Org. Chem. 1994, 59, 5063-5070.
(
11) Ingold, K. U. In Free Radicals; Kochi, J. K., Ed.; Wiley: New York,
1973; Vol. I, Chapter 2, p 67.
3 2
)
(
12) (a) Eisenberg, D. C.; Norton, J. R. Isr. J. Chem. 1991, 31, 55-65 and
references therein. (b) Eisenberg, D. C.; Lawrie, C. J. C.; Moody, A.;
Norton, J. R. J. Am. Chem. Soc. 1991, 113, 4888-4895.
of the methyl and phenyl compounds remained unaltered.
The features observed can be summarized as follows: (1) The
reaction rate seems to decrease in the order TEMPO > DPPH >
JA0271126
J. AM. CHEM. SOC.
9
VOL. 124, NO. 38, 2002 11279