there is a preference for the metal catalyst to transfer the
methylene hydrogen atoms from the solvent. Although partial
hydride transfer has been previously reported using ethereal
solvents such as 1,4-dioxane in the presence of palladium
catalysts,9 to the best of our knowledge, hydride transfer has
not been observed with 1,2-dimethoxyethane. Furthermore,
a preference for the abstraction of methylene versus methyl
hydrogens in ethereal solvents under palladium catalysis has
not been described.
Scheme 3. Methanol Trapping Experiment
Although we have determined the source of reducing
hydrogens, the mechanism of hydride transfer and the
formation of methanol remain unclear. Keay has proposed
a ꢀ-hydride elimination pathway for 1,4-dioxane, which is
plausible in the present system. We believe that the stoichi-
ometry of the base is important to the reductive process, as
using less than 2 equiv of Cs2CO3 leads to a drastic reduction
in product yields. Further research is required to fully
elucidate the reductive mechanism.
In summary, we have developed and studied a method
for the synthesis of meta-substituted biaryls through a
reductive ortho-arylation sequence. Through investigation of
the reduction using deuterium-labeled reagents and solvents,
we have determined that palladium-catalyzed decomposition
of DME is responsible for the reduction of the terminal
arylpalladium(II) species.
product. Further control experiments determined that the
nitrobenzoate is not formed in the absence of palladium
catalyst. From these experiments, we can conclude that
methanol is derived from DME under the reaction conditions.
Although our observation of methanol in crude spectra
led us to believe that it is not consumed in the reductive
process, a recent report of deuterium transfer by CD3OD
under palladium catalysis8 prompted us to add it in the
reaction with deuterated ester 5. This substrate was chosen
as the undeuterated ester was found to irreversibly exchange
to the deuterated ester in the presence of CD3OD under the
reaction conditions. When 1 equiv of CD3OD was added to
the reaction of 5, we obtained product with no more than
10% deuterium at the ipso position and believe that it is not
a major contributor to the reductive pathway.
Finally, we synthesized two different isotopomers of DME:
one with deuterium atoms on the methyl carbons (d6-DME)
and the other completely deuterated (d10-DME). Under the
optimized conditions, d6-DME did not afford any observable
ipso deuteration. Conversely, when d10-DME was used, we
observed 80% deuteration of the ipso carbon based upon both
Acknowledgment. We would like to thank the University
of Toronto, the Natural Sciences and Engineering Research
Council of Canada, and Merck Inc. for financial support in
the form of an Industrial Research Chair.
2
1H and H NMR spectra. This result confirmed that the
Supporting Information Available: Experimental details
and characterization data for compounds 3b-3n, 5, d6-DME,
and d10-DME. This material is available free of charge via
solvent was involved in the reduction of the terminal
arylpalladium(II) species. Furthermore, it appears as though
(6) Under the optimized conditions with 25 mol % norbornene, 3d was
OL1019037
obtained in 75% isolated yield. Norbornene was also observed in near
1
quantitative amounts in crude H NMR analyses.
(7) Martins, A.; Lautens, M. Org. Lett. 2008, 10, 5095.
(8) Wisily, A.; Nguyen, Y.; Fillion, E. J. Am. Chem. Soc. 2009, 131,
15606.
(9) Lau, S. Y. W.; Andersen, N. G.; Keay, B. A. Org. Lett. 2001, 3,
181. Up to 1:1 H:D ratios reported. Under our conditions, d8-dioxane only
afforded traces of product and did not incorporate deuterium.
5188
Org. Lett., Vol. 12, No. 22, 2010