Organic Letters
Letter
nitro-substituted 1e reacted under our reaction conditions and
gave 3q in moderate yield (40%).
resulting in the formation of O(SiMe2Ph)2 which was also
observed as a side product. The water that protonates 7 can
originate either from the imine condensation or from the
solvent, as nondried PhCl was used. To the best of our
knowledge, assisted tandem catalysis consisting of three
coupled catalytic cycles that employ the same precatalyst are
rare.17
In conclusion, the one-pot transformation of arylepoxides to
β-arylamines has been reported. The reactions are catalyzed by
B(C6F5)3 and proceed in easy to handle nondried solvents. Our
protocol has been tested for the scope of epoxides and amines
and is a complementary strategy to the classic Lewis acid
mediated synthesis of amino alcohols from amines and
epoxides. Therefore, it represents an important addition to
the toolbox of synthetic chemists. We propose that the reaction
occurs via an unprecedented Meinwald rearrangement−
reductive amination tandem reaction that is catalyzed by
B(C6F5)3. We suggest that these metal-free transformations
occur via three coupled catalytic cycles that are all catalyzed by
B(C6F5)3, highlighting the versatility of B(C6F5)3. Current
investigations in our laboratory focus on the development of an
asymmetric protocol of this transformation.
When moving from 1a to the trans-isomer 1f no product
formation was observed under our optimized conditions. The
case was similar when styrene oxides 1g, 1h, and 1i were tested.
As already reported by Yamamoto and co-workers,8 the ability
of an epoxide to undergo a B(C6F5)3-catalyzed Meinwald
rearrangement is solvent dependent. Indeed, when stirring 1f in
1,4-dioxane in the presence of 10 mol % borane, followed by
addition of 1.2 equiv 2a in PhCl and 1.2 equiv 4 in PhCl, we
observed the formation of 3a in 64% yield. Also styrene oxides
1g, 1h, and 1i reacted under our modified protocol to afford the
desired β-arylamines 3r (30%), 3s (26%), and 3t (24%).
Unfortunately, the 1,2-dialkyl-substituted epoxide 1j gave no
product 3u under our conditions. However, to the best of our
knowledge, the catalytic results presented here are the first
examples of a Meinwald rearrangement−reductive amination
protocol yielding β-functionalized amines.
Based on our studies and reported mechanism for the
Meinwald rearrangement7g and the B(C6F5)3-catalyzed reduc-
tive amination,13 we propose the overall mechanism depicted in
Scheme 1. This mechanism consists of three coupled catalytic
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
Scheme 1. Proposed Mechanism of the B(C6F5)3-Catalyzed
Tandem Meinwald Rearrangement−Reductive Amination of
1a
■
S
Experimental procedures, spectral data, and other
characterization data (PDF)
AUTHOR INFORMATION
Corresponding Author
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
M.O. thanks the sustainability scheme of Utrecht University for
funding. Henk Kleijn (Organic Chemistry and Catalysis, Debye
Institute for Nanomaterials Science, Universiteit Utrecht) is
acknowledged for technical support.
REFERENCES
■
(1) (a) Gansauer, A.; Bluhm, H.; Pierobon, M. J. Am. Chem. Soc.
̈
cycles that are all B(C6F5)3-catalyzed. The first cycle is the
Meinwald rearrangement where the catalyst activates 1a via
coordination. After ring opening a 1,2-migration occurs,
resulting in the formation of aldehyde 5 and catalyst
regeneration. The second cycle is the imine condensation
where B(C6F5)3 activates 5 for a nucleophilic attack of 2a. After
proton migration and water elimination, imine 6 is obtained
and the catalyst is regenerated. The third cycle is the imine
reduction. Here, first B(C6F5)3 abstracts a hydride from silane 4
1998, 120, 12849−12859. (b) Gansauer, A.; Fan, C.-A.; Piestert, F. J.
̈
Am. Chem. Soc. 2008, 130, 6916−6917. (c) Gansauer, A.; Klatte, M.;
̈
Brandle, G. M.; Friedrich, J. Angew. Chem., Int. Ed. 2012, 51, 8891−
̈
8894.
(2) (a) Justicia, J.; Oller-Lop
Cuerva, J. M.; Bunuel, E.; Car
́
ez, J. L.; Campana, A. G.; Oltra, J. E.;
denas, D. J. J. Am. Chem. Soc. 2005, 127,
̃
́
̃
14911−14921. (b) Gansauer, A.; Worgull, D.; Knebel, K.; Huth, I.;
̈
Schnakenburg, G. Angew. Chem., Int. Ed. 2009, 48, 8882−8885.
(c) Gansauer, A.; Shi, L.; Otte, M. J. Am. Chem. Soc. 2010, 132,
̈
to generate HB(C6F5)3 and PhMe2Si+. PhMe2Si+ is able to
−
11858−11859. (d) Gansauer, A.; Otte, M.; Shi, L. J. Am. Chem. Soc.
̈
2011, 133, 416−417.
−
activate 6 for a hydride transfer from HB(C6F5)3 resulting in
(3) (a) Mulzer, M.; Whiting, B. T.; Coates, G. W. J. Am. Chem. Soc.
2013, 135, 10930−10933. (b) Mulzer, M.; Tiegs, B. J.; Wang, Y.;
Coates, G. W.; O’Doherty, G. A. J. Am. Chem. Soc. 2014, 136, 10814−
10820.
the formation of “silyl amine” 7 and B(C6F5)3. Alternatively, the
formation of 7 could also proceed via a borane−silane complex,
as reported by Piers, Tuononen and co-workers.16 7 is
protonated by water to give 3a and HOSiMe2Ph. Very likely,
HOSiMe2Ph is able to protonate a second equivalent of 7
(4) Rowley, J. M.; Lobkovsky, E. B.; Coates, G. W. J. Am. Chem. Soc.
2007, 129, 4948−4960.
C
Org. Lett. XXXX, XXX, XXX−XXX