Oxidative coupling of amines and ketones by combined vanadium- and organocatalysis

Article abstract of DOI:10.1039/b901282f

The combination of vanadium- and organocatalysis allows for the direct oxidative coupling of cyclic tertiary amines with non-activated ketones without the need for preformed leaving groups.

Full text of DOI:10.1039/b901282f

Showcasing research from Dr. Martin Klussmann’s
laboratory, Max-Planck-Institute for Coal Research,
Mülheim an der Ruhr, Germany.
As featured in:
Chemical Communications
www.rsc.org/chemcomm
Number 22 | 14 June 2009 | Pages 3125–3296
Title: Oxidative coupling of amines and ketones by combined
vanadium- and organocatalysis
The back cover displays the synthesis of the natural product
hygrine by dual catalysis: vanadinite crystals represent oxidative
vanadium catalysis and the chicken stands for catalysis with
proline, a natural amino acid which can be isolated from chicken
feathers.
ISSN 1359-7345
FEATURE ARTICLES
Orion B. Berryman and
Darren W. Johnson
FEATURE ARTICLES
Experimental evidence for
interactions between anions and
electron-deficient aromatic rings
Jason B. Love
A
macrocyclic approach to transition
1359-7345(2009)22;1-1
metal and uranyl Pacman complexes
See Abhishek Sud, Devarajulu Sureshkumar
and Martin Klussmann, Chem. Commun.,
2009, 3169.
www.rsc.org/chemcomm
Registered Charity Number 207890

COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Oxidative coupling of amines and ketones by combined vanadium- and
organocatalysisw
Abhishek Sud,z Devarajulu Sureshkumarz and Martin Klussmann*
Received (in Cambridge, UK) 21st January 2009, Accepted 17th February 2009
First published as an Advance Article on the web 17th March 2009
DOI: 10.1039/b901282f
The combination of vanadium- and organocatalysis allows for
the direct oxidative coupling of cyclic tertiary amines with
non-activated ketones without the need for preformed leaving
groups.
Table 1 Screening of metal catalysts for the oxidative coupling of
amine 1 with acetoneab
Oxidative coupling can be an elegant way to form new
carbon–carbon bonds without the need of special leaving
groups by formally breaking two carbon–hydrogen bonds
and coupling the fragments.1 As a net oxidation of the
substrates occurs, an oxidant is required, typically producing
water as the only byproduct (Scheme 1). These reactions can
be very atom economic and also faster than other routes that
require multiple steps and functionalisations.
Proline/
mol%
Conversion
of 1 (%)
Yieldc
(%)
Entry
Met
t/h
1
2
3
CuCl2
Cu(acac)2
FeCl3ꢀ
10
10
10
12
96
8
Full
Full
Full
o5
43
23
6H2O
We became interested in utilising oxidative coupling reactions
of amines and carbonyl compounds as an alternative to
Mannich reactions, furnishing valuable structural motives
that occur in natural products and pharmaceuticals. The
oxidative functionalisation of amines in the a-position has
been investigated for some time.2 In most of these cases,
iminium ions are thought to be the key intermediate; a
subsequent reaction with good nucleophiles then furnishes
the products.3 Li and co-workers have recently made important
progress in this field by significantly expanding the scope of
these reactions, generally using copper catalysts together with
organic peroxides1b,4 or oxygen.4d In all these examples, the
carbonyl compounds that could be coupled with the amines
exhibited considerable nucleophilicity, a noteworthy exception
are enones that were activated in a Baylis–Hillman-type
fashion.4c The direct coupling with non-activated simple
ketones has not been achieved so far.
4
5
6
7
8
Fe(acac)3
Co(acac)2
VO(acac)2
VO(acac)2
—
10
10
10
—
10
12
24
24
72
24
Full
Full
Full
o5
50
28
41
69
0
6
a
Reaction conditions: amine (0.12 mmol), ketone (0.60 mmol), metal
catalyst (0.012 mmol), L-proline (0.012 mmol), TBHP–decane
b
(0.18 mmol), MeOH (1.0 ml), RT. acac
c
= acetyl acetonate.
Isolated yield.
With copper chloride, only traces of the desired product
were formed, while a moderate yield (43%) was obtained after
four days using the corresponding acetyl acetonate (acac,
entries 1 and 2). Potentially, the chelating acac ligands hinder
a coordination of the metal with proline, which could lead to a
deactivation of both catalysts. Copper(II) is known to form a
stable complex with proline6 and we could show that addition
of proline deactivates oxidative copper catalysis in the
coupling of naphthol to 1,10-bis-2-naphthol.7
To overcome this limitation, we planned to use a secondary
amine as an organic cocatalyst to activate ketones in the form
of a nucleophilic enamine intermediate.5 Subsequent reaction
with iminium ions formed by oxidation of the amine would
then lead to the product. To test our strategy, we tried to
perform the coupling of tetrahydroisoquinoline 1 with acetone
in the presence of a metal catalyst, tert-butyl hydroperoxide
(TBHP) as oxidant and L-proline as an inexpensive and
reliable organocatalyst (Table 1).
Iron and cobalt catalysts gave faster conversions but not
higher yields (entries 3–5). This was achieved with VO(acac)2,
forming 3 with 69% yield after one day (entry 6). No product
was formed without the cocatalyst proline and only trace
amounts without VO(acac)2 (entries 7 and 8), indicating the
success of the dual catalysis concept. The reactions generally
produced several unidentified decomposition products, which
explains the moderate yields of the desired product even in the
cases of high conversion. Other oxidants than TBHP gave
lower yields and no reaction occurred with elemental oxygen.
Optimisation of the reaction conditions led to the establishment
of 10 mol% VO(acac)2 and 10 mol% L-proline, methanol as
the solvent and the use of 1.5 equivalents of TBHP as the
oxidant. Several tetrahydroisoquinolines could be coupled
with a number of acyclic aliphatic ketones (Table 2).
Scheme 1 Carbon–carbon bond formation by oxidative coupling.
Max-Planck-Institut fuer Kohlenforschung, Kaiser-Wilhelm-Platz 1,
45470, Muelheim an der Ruhr, Germany.
E-mail: klusi@mpi-muelheim.mpg.de
w Electronic supplementary information (ESI) available: Experimental
details. See DOI: 10.1039/b901282f
Interestingly, the coupling always occurred at the non-
substituted a-position of the ketone, in contrast to the
z A. S. and D. S. contributed equally to this project.
ꢁc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 3169–3171 | 3169

Table 2 Oxidative coupling of tetrahydroisoquinolines with ketones
by dual vanadium- and organocatalysisab
Yieldc
(%)
Entry
1
Amine
R0
Product
t/d
Scheme 2 Synthesis of hygrine and derivatives by oxidative coupling
with N-methyl pyrrolidine 13 (dr = diastereomeric ratio, determined
1
by H-NMR).
1
Me
1
69
outcome of most organocatalytic reactions.5 Accordingly, we
were not able to obtain products with ketones without an
a-methyl group, possibly due to steric reasons.
2
1
1
1
4
4
5
5
Et
1
1
3
1
1
1
1
65
41
32
51
40
65
36
Next, we turned our attention to other cyclic tertiary
amines. N-Methyl pyrrolidine 13 proved to be another good
substrate, despite giving lower yields than 1. To our delight,
the reaction with acetone proceeded well to yield hygrine 14, a
natural product that is a biochemical precursor for several
other alkaloids.8 The modest yield of only 45% is counter-
balanced by the fact that our methodology provides a direct
one-step synthesis of hygrine, the shortest total synthesis
reported,8,9 using simple and readily available starting materials
(Scheme 2).
3d
n-Pr
i-Bu
Me
Et
4
Lower loadings of VO(acac)2 (5 mol%) could be used
without influence on the yield and isohexane proved to be
the best solvent for this substrate which failed with amine 1,
probably because of its low solubility. We were unable to
obtain more than traces of the coupling products with other
open-chain ketones except acetone. Nevertheless, we were
able to isolate the coupling products with cyclic ketones. In
contrast to the reactions with 1 depicted in Table 2, slow
formation of the products with amine 13 occurred even
without the organocatalyst. This could be due to base-catalysed
enolisation of the ketones with the more basic 13.
5
6
Despite the use of an enantiopure organocatalyst, proline,
nearly all the products isolated were racemic. This can be
rationalised by the facile racemisation of b-amino-ketones.
Hygrine 14 is known to racemise very fast in basic conditions
and even without added base racemisation can occur within
minutes to hours.9 Not surprisingly, we could only obtain
racemic 14. However, enantiopure hygrine would still be
accessible by resolution with tartaric acid.9
7
Me
Non-racemic products were achieved with tetrahydroiso-
quinoline 1, albeit with very low enantioselectivities. For
example, 3 was produced with 7% ee catalysed by L-proline;
screening of various other chiral organocatalysts revealed
Jørgensen-type catalyst 2010 as the most stereoselective with
a product ee of 17% but poor yields (Scheme 3).
8
Et
a
Reaction conditions: amine (0.12 mmol), ketone (0.60 mmol),
VO(acac)2 (0.012 mmol), L-proline (0.012 mmol), TBHP–decane
Further attempts of asymmetric catalysis by using chiral
vanadium complexes all failed, only yielding racemic products.
We found that enantiopure 3, obtained by preparative HPLC,
slowly racemized under the reaction conditions (see ESIw).
b
(0.18 mmol), MeOH (1.0 ml), RT, 1–3 d. PMP = p-methoxyphenyl.
c
d
Isolated yield. Solvent DMSO, 2.0 eq. TBHP.
ꢁc
This journal is The Royal Society of Chemistry 2009
3170 | Chem. Commun., 2009, 3169–3171

the iminium ion. Intermediate C is then hydrolysed to give the
product and reform both catalysts (Scheme 4). Accordingly,
the reaction can be considered an ‘‘oxidative Mannich
reaction’’.15
Despite the modest yields and restricted substrate scope, this
reaction significantly enhances previous methodologies in both
metal- and organocatalysis. By using an organic cocatalyst,
simple non-activated ketones become accessible substrates
for oxidative coupling reactions.16 By utilising oxidative
vanadium catalysis, cyclic amines become accessible for
Mannich-type reactions that are usually the domain of open-
chain imines. Further studies to extend the scope of these
reactions are ongoing in our laboratory.
Scheme 3 Results of asymmetric reactions using chiral organocatalysts.
This could help to explain the low enantioselectivities
observed in the product. Obviously, asymmetric oxidative
coupling reactions with tertiary amines11 remain a challenge,
as amino-carbonyl compounds have previously only been
made racemic and the best ee achieved with other coupling
partners is 74% in an alkynation.12 Although the use of
enantiopure proline seems superfluous in non-asymmetric
reactions, it still is the cocatalyst of choice in this reaction.
Its high activity, low price and ease of use as a bench-stable
crystalline solid make it superior to other organocatalysts we
have tested.
We are grateful for financial support from the Alexander
von Humboldt-foundation (research fellowship to D. S.) and
from Prof. Benjamin List. We thank Esther Boß for the
synthesis of some starting materials.
Notes and references
1 For recent overviews, see: (a) F. Kakiuchi and T. Kochi, Synthesis,
2008, 3013; (b) C.-J. Li, Acc. Chem. Res., 2009, 42, 335.
2 K. R. Campos, Chem. Soc. Rev., 2007, 36, 1069.
3 (a) J.-I. Yoshida, S. Suga, S. Suzuki, N. Kinomura, A. Yamamoto
and K. Fujiwara, J. Am. Chem. Soc., 1999, 121, 9546; (b) T. Chiba
and Y. Takata, J. Org. Chem., 1977, 42, 2973; (c) T. Shono,
Y. Matsumura and K. Tsubata, J. Am. Chem. Soc., 1981, 103,
1172; (d) T. Naota, T. Nakato and S.-I. Murahashi, Tetrahedron
Lett., 1990, 31, 7475; (e) S.-I. Murahashi, N. Komiya, H. Terai and
T. Nakae, J. Am. Chem. Soc., 2003, 125, 15312;
(f) S.-I. Murahashi, T. Nakae, H. Terai and N. Komiya, J. Am.
Chem. Soc., 2008, 130, 11005; (g) G. Rousselet, P. Capdevielle and
M. Maumy, Tetrahedron Lett., 1995, 36, 4999.
4 (a) Z. Li and C.-J. Li, J. Am. Chem. Soc., 2005, 127, 3672; (b) Z. Li
and C.-J. Li, Eur. J. Org. Chem., 2005, 3173; (c) Z. Li, D. S. Bohle
and C.-J. Li, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 8928;
(d) O. Basle and C.-J. Li, Green Chem., 2007, 9, 1047; (e) L. Zhao
and C.-J. Li, Angew. Chem., Int. Ed., 2008, 47, 7075; (f) Z. Li and
C.-J. Li, J. Am. Chem. Soc., 2005, 127, 6968.
Since both the formation of amine N-oxides from tertiary
amines13 and their conversion to iminium ions14 are reported
to be catalysed by vanadium complexes, we initially assumed
that the reaction mechanism involved amine N-oxides as
key intermediates. However, the preformed N-oxide of 13
failed to react under our reaction conditions. Also, it could
not be detected as an intermediate, even when the reaction
was performed in the absence of a ketone and proline.
Alternatively, iminium ions could be formed by a radical
mechanism, involving a stepwise removal of electrons and a
proton. The existence of radical intermediates is supported by
the fact that the synthesis of hygrine failed if one equivalent of
the radical inhibitor 2,4,6-tri-tert-butyl phenol is added.
As a working model, we propose a dual catalytic cycle
involving an iminium ion A, formed by oxidation through a
radical pathway from the substrate amine, and a nucleophilic
enamine B, formed from the ketone and proline, which attacks
5 For recent overviews on enamine catalysis, see: (a) S. Mukherjee,
J. W. Yang, S. Hoffmann and B. List, Chem. Rev., 2007, 107, 5471;
(b) P. Melchiorre, M. Marigo, A. Carlone and G. Bartoli, Angew.
Chem., Int. Ed., 2008, 47, 6138; (c) D. W. C. MacMillan, Nature,
2008, 455, 304.
6 A. M. Mathieson and H. K. Welsh, Acta Crystallogr., 1952,
5, 599.
7 A. Sud, D. Sureshkumar and M. Klussmann, unpublished results.
8 E. B. Arevalo-Garcıa and J. C. Q. Colmenares, Tetrahedron Lett.,
2008, 49, 3995.
9 F. Galinovsky and H. Zuber, Monatsh. Chem., 1953, 84, 798.
10 J. Franzen, M. Marigo, D. Fielenbach, T. C. Wabnitz,
A. Kjærsgaard and K. A. Jørgensen, J. Am. Chem. Soc., 2005,
127, 18296.
11 For a highly enantioselective oxidative Mannich reaction with
secondary amines, see: I. Ibrahem, J. S. M. Samec, J. E. Backvall
and A. Cordova, Tetrahedron Lett., 2005, 46, 3965.
12 Z. Li and C.-J. Li, Org. Lett., 2004, 6, 4997.
13 (a) L. Kuhnen, Chem. Ber., 1966, 99, 3384; (b) M. N. Sheng and
J. G. Zajacek, J. Org. Chem., 1968, 33, 588.
14 D.-R. Hwang and B.-J. Uang, Org. Lett., 2002, 4, 463.
15 For recent reviews on organocatalytic Mannich reactions, see:
(a) J. M. M. Verkade, L. J. C. van Hemert, P. J. L.
M. Quaedflieg and F. P. J. T. Rutjes, Chem. Soc. Rev., 2008, 37,
29; (b) A. Ting and S. E. Schaus, Eur. J. Org. Chem., 2007, 5797.
16 After submission of this manuscript, a related report using non-
activated ketones appeared: Y. Shen, M. Li, S. Wang, T. Zhan,
Z. Tan and C.-C. Guo, Chem. Commun., 2009, 953.
Scheme 4 Mechanistic proposal (L = acetyl acetonate).
ꢁc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 3169–3171 | 3171