Modular pyridine synthesis from oximes and enals through synergistic copper/iminium catalysis

Article abstract of DOI:10.1021/ja312346s

We describe here a [3+3]-type condensation reaction of O-acetyl ketoximes and α,β-unsaturated aldehydes that is synergistically catalyzed by a copper(I) salt and a secondary ammonium salt (or amine). This redox-neutral reaction allows modular synthesis of a variety of substituted pyridines under mild conditions with tolerance of a broad range of functional groups. The reaction is driven by a merger of iminium catalysis and redox activity of the copper catalyst, which would initially reduce the oxime N-O bond to generate a nucleophilic copper(II) enamide and later oxidize a dihydropyridine intermediate to the pyridine product.

Full text of DOI:10.1021/ja312346s

Communication
pubs.acs.org/JACS
Modular Pyridine Synthesis from Oximes and Enals through
Synergistic Copper/Iminium Catalysis
Ye Wei† and Naohiko Yoshikai*
Division of Chemistry and Biological Chemistry, School of Physical and Mathematical Sciences, Nanyang Technological University,
Singapore 637371, Singapore
S
* Supporting Information
Scheme 1
ABSTRACT: We describe here a [3+3]-type condensa-
tion reaction of O-acetyl ketoximes and α,β-unsaturated
aldehydes that is synergistically catalyzed by a copper(I)
salt and a secondary ammonium salt (or amine). This
redox-neutral reaction allows modular synthesis of a
variety of substituted pyridines under mild conditions
with tolerance of a broad range of functional groups. The
reaction is driven by a merger of iminium catalysis and
redox activity of the copper catalyst, which would initially
reduce the oxime N−O bond to generate a nucleophilic
copper(II) enamide and later oxidize a dihydropyridine
intermediate to the pyridine product.
ynergistic catalysis has attracted increasing interest in the
Sfield of chemical synthesis because it potentially enables a
previously unknown type of transformation or provides a
known reaction with unprecedented efficiency or chemo-,
regio-, and stereoselectivity.1a Among various catalyst combi-
nations, the combination of transition metal and organic
ketones such as acetophenone, propiophenone, and cyclo-
appeared straightforward, in our hands, reactions of simple
catalysts is attractive, where the complementary modes of
substrate activation could lead to diverse transformations.1 In
particular, the combination of transition metal and secondary
amine catalysts has been extensively practiced in carbonyl α-
functionalizations, where enamine catalysis is merged with
various modes of electrophile activation with transition metal
catalysts.2 In comparison, the use of secondary amines for
metal/iminium synergistic catalysis has thus far been limited to
asymmetric conjugate addition of silicon, boron, and carbon
nucleophiles to α,β-unsaturated aldehydes, where the organo-
metallic nucleophiles are activated uniformly through trans-
metalation.3 Here, we report on a copper(I)/secondary amine-
catalyzed [3+3]-type condensation reaction of O-acetyl oximes
and enals that features a unique merger of iminium catalysis
and redox copper catalysis, allowing modular synthesis of a
variety of substituted pyridines under mild conditions (Scheme
1a). The copper catalyst presumably plays roles (1) to reduce
the oxime N−O bond4,5 and generate a nucleophilic copper(II)
enamide, and (2) to oxidize a dihydropyridine intermediate to
the pyridine product. As such, the overall transformation is
redox-neutral and produces water and acetic acid as the only
byproducts.
hexanone with cinnamaldehyde and ammonium acetate all
failed to afford the desired pyridine products in more than 5%
yield but produced complex mixtures (Scheme 1b). This was
not at all surprising, because the scope of the classical pyridine
synthesis based on carbonyl condensation is severely limited
due to the difficulty in controlling multiple reaction steps, such
as imine/enamine formation and Michael addition, while
suppressing many undesirable competing pathways.7,8 We
reasoned that this intrinsic difficulty might be removed by
preinstalling a nitrogen atom as well as an internal oxidant in
the reactant instead of using a ketone, an ammonia source, and
an external oxidant separately.
With the above conjecture and the ability of copper(I) to
reduce oxime N−O bonds,4,5 we chose acetophenone O-acetyl
oxime 1a and cinnamaldehyde 2a as model substrates and
screened copper salts and other reaction conditions (Table 1).9
With only CuI (20 mol %), the reaction in DMSO at 60 °C
produced 2,4- and 2,6-diphenylpyridines 3aa and 3aa′ in low
yield (11%) with a ca. 1:1 ratio (entry 1). The addition of
pyrrolidinium perchlorate (20 mol %) dramatically accelerated
the reaction and improved the regioselectivity, affording 3aa as
the sole product in 78% yield (entry 2). Similar effects were
observed with ammonium salts of piperidine, morpholine, and
The development of the present condensation reaction was
conceived originally from our need for a convenient method for
preparing pyridine derivatives for other purposes.6 While
oxidative condensation of a ketone, an enal, and ammonia
Received: December 18, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja312346s | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
a
Table 1. Screening of Reaction Conditions
Table 2. Condensation of Various Oximes with
Cinnamaldehyde
b
b
entry
R2NH(·HX)
GC yield (%)
11
3aa:3aa′
1
2
3
4
5
6
none
53:47
>99:1
>99:1
>99:1
>99:1
>99:1
−
>99:1
>99:1
>99:1
99:1
c
pyrrolidine·HClO4 (20 mol %)
piperidine·HClO4 (20 mol %)
morpholine·HClO4 (20 mol %)
Et2NH·HClO4 (20 mol %)
iPr2NH·HClO4 (20 mol %)
pyrrolidine·HClO4 (20 mol %)
pyrrolidine (20 mol %)
78
46
65
72
8
d
7
0
8
23
1
d
9
pyrrolidine (2 equiv)
10
11
iPr2NH (20 mol %)
55
c
iPr2NH (2 equiv)
91
a
b
Reaction was performed on a 0.2 mmol scale. Determined by GC
c
d
using n-tridecane as an internal standard. Isolated yield. CuI was
omitted from the reaction.
Et2NH (entries 3−5), while the ammonium salt of iPr2NH was
not very effective (entry 6). The ammonium salt alone did not
catalyze the reaction (entry 7). Among other copper salts,
CuBr·SMe2 showed a comparable catalytic activity, while
Cu(OAc)2 and CuOTf were less effective.9
Upon further examination, we noted that pyrrolidine itself
was not a very effective cocatalyst (entry 8). The use of 2 equiv
of pyrrolidine almost completely shut down the reaction (entry
9). In contrast, iPr2NH, in its neutral form, showed a significant
positive effect, affording 3aa in 55% and 91% yield with 20 mol
% and 2 equiv loadings, respectively (entries 10 and 11). We
speculate that coordination of pyrrolidine to the copper catalyst
killed its catalytic activity, while bulky iPr2NH did not cause
such interference. Note that the use of O-pentafluorobenzoyl
oxime or free oxime instead of 1a resulted in a lower yield or no
reaction, respectively.
Based on the above optimization study, we defined the
methods using pyrrolidinium salt (20 mol %) and iPr2NH (2
equiv) as methods A and B, respectively, and explored their
scope in the condensation of various oximes with 2a (Table 2).
A variety of oximes derived from aryl methyl ketones
participated in the reaction with both of the methods, affording
the corresponding 2,4-disubstituted pyridines 3aa−3ma, with
tolerance of functional groups, including bromo, iodo, cyano,
and nitro groups as well as pyridine, furan, and thiophene
heterocycles. The reaction of the parent acetophenone oxime
1a could be scaled up to 10 mmol scale without a problem.
While most of the oximes exclusively afforded the 2,4-
disubstituted pyridines, reaction of the oxime derived from
2′-methylacetophenone was accompanied by the 2,6-disub-
stituted isomer 3ia′. Oximes derived from other aryl ketones
such as propiophenone, 2-phenylacetophenone, and tetralone
afforded the products 3na−3pa, respectively, in good yields.
The difference in the scope of methods A and B became clear
upon further exploration. Thus, oximes derived from
benzylideneacetone, acyclic and cyclic aliphatic ketones, and
ethyl pyruvate afforded the pyridine derivatives 3qa−3wa in
moderate to good yields with method B, while method A was
much less effective (<20% yield except for 3sa).
a
Unless otherwise noted, the reaction was performed on a 0.2 mmol
b
c
scale. 10 mmol scale. Method A afforded ca. 20% yield (GC).
d
Method A afforded less than 10% yield (GC).
We next explored the scope of α,β-unsaturated aldehydes
(Table 3). As was the case with cinnamaldehyde, β-aryl enals
were amenable to the condensation reaction using both
methods A and B, affording 2,4-diarylpyridines 3ab−3ae, 3bf,
3bg, and 3ah in good yields. Method B showed a better
performance than method A in the reaction of methacrolein
(see 3pi), and also allowed acrolein to take part in the reaction,
albeit in a modest yield (see 3pj). In contrast, β-alkyl and α,β-
disubstituted enals exclusively produced the expected pyridine
derivatives 3pk and 3pl, respectively, with method A, while
substantial amounts of their regioisomers were observed with
method B (see the Supporting Information). α,β-Unsaturated
ketones such as benzylideneacetone and chalcone reacted
rather sluggishly with both methods A and B, affording the
corresponding pyridine products in less than 5% yield.
Control experiments supported the intermediacy of an
iminium ion in the present reaction (with method A). Thus,
the reaction of 1a and iminium salt 4a (1.5 equiv) derived from
B
dx.doi.org/10.1021/ja312346s | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
a
condensation of 1a and 2a, affording 3aa in 79% yield (Scheme
2b). Note that the reaction of the iminium salt of
benzylideneacetone (4m) was rather sluggish (Scheme 2c),
which was consistent with the poor reactivity of enones in the
present reaction (vide supra). Unfortunately, the same control
experiments could not be performed for the reaction with
method B because attempts to prepare the corresponding
isolated iminium salt of iPr2NH were unsuccessful with
conventional and other methods.10
Table 3. Condensation with Various Enals
It is natural to assume that the oxime and the copper catalyst
give rise to a nucleophilic reaction partner for the iminium ion.
On the basis of previous studies on copper catalysis of oxime
derivatives,4,5d,e we consider that oxime 1 and CuI produce an
iminylcopper(II) species 6 through a sequential single-electron
reduction of the N−O bond (Scheme 3a).11 Tautomerization
Scheme 3. (a) Possible Pathway for the Activation of Oxime
with Copper(I) Catalyst and (b) Proposed Mechanism
Involving Copper(I)/Iminium Catalysis
a
b
The reaction was performed on a 0.2 mmol scale. 40 mol % of CuI
c
d
e
was used. p-An = 4-MeOC6H4. 2 equiv of methacrolein was used. 5
equiv of acrolein was used. Not examined with method A. 3 equiv of
f
enal was used. Method B produced a mixture of regioisomers (see the
Supporting Information).
cinnamaldehyde and pyrrolidine, in the presence of 20 mol %
CuI, afforded 3aa in 81% yield (Scheme 2a). No pyridine
product was obtained in the absence of CuI. Iminium salts
derived from piperidine and morpholine also participated in the
reaction, albeit in lower yields (57% and 55%, respectively). In
addition, the iminium salt 4a served as a cocatalyst for the
Scheme 2. Control Experiments Using Iminium Salt
of 6 would then give a copper(II) enamide 7,5e,12 which would
serve as a nucleophile toward the iminium ion. With this
speculation, we formulate a possible mechanism as illustrated in
Scheme 3b for the reaction with method A. Michael addition of
7 to the iminium ion 4 produces an enamine intermediate 8,13
which is then protonated to give an iminium ion 9.
Deaminative cyclization of 9 affords dihydropyridine 10 and
regenerates the ammonium catalyst. Oxidation of 10 with
copper(II) furnishes the pyridine product 3 along with acetic
acid (HX), while regenerating the copper(I) catalyst. Thus, the
synchronized and intersecting copper and iminium catalytic
cycles represent a signature of synergistic catalysis.1a The
formation of the pyridine regioisomer 3′ (see Table 2, 3ia′)
may result from Michael addition of the N atom of 7 (or 6) to
4 (see Scheme S1 in the Supporting Information).
We speculate that the reaction with method B also involves
the same type of iminium activation mechanism. The much
C
dx.doi.org/10.1021/ja312346s | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
76, 6159. (e) Ren, Z.-H.; Zhang, Z.-Y.; Yang, B.-Q.; Wang, Y.-Y.;
Guan, Z.-H. Org. Lett. 2011, 13, 5394.
(6) (a) Gao, K.; Yoshikai, N. J. Am. Chem. Soc. 2011, 133, 400.
(b) Gao, K.; Lee, P.-S.; Fujita, T.; Yoshikai, N. J. Am. Chem. Soc. 2010,
132, 12249.
(7) Jones, G. In Comprehensive Heterocyclic Chemistry; Katritzky, A.
R., Rees, C. W., Eds.; Pergamon: Oxford, 1984; pp 395−510.
(8) For such condensations to be practically feasible, modification of
the ketone with a less convenient α-leaving group (e.g., pyridinium,
benzotriazolyl) or an additional carbonyl group (i.e., 1,3-dicarbonyl
improved reactivity with iPr2NH than with its ammonium salt
(see entries 6 and 10 in Table 1) may suggest that the
concentration of neutral amine is more crucial than that of
active proton for the formation of an iminium ion from this
bulky and less nucleophilic secondary amine.
In summary, we have successfully combined the redox
activity of copper and the iminium activation strategy to
construct pyridines from O-acetyl oximes and α,β-unsaturated
aldehydes. With the operational simplicity, modularity, and
functional group compatibility, the present reaction has
substantially expanded the scope of pyridine derivatives
accessible from readily available carbonyl compounds.
Furthermore, in light of the accessible substitution patterns,
the present method not only effectively complements other
existing and emerging methods for pyridine synthesis14−18 but
also enables, in combination with methods for the direct
functionalization of the pyridine core,19 the construction of a
diverse array of polysubstituted pyridines. Further improvement
and extension of this synergistic catalysis are currently in
progress.
derivative) is typically required: (a) Krohnke, F. Synthesis 1976, 1.
̈
(b) Tewari, R. S.; Awasthi, A. K. Synthesis 1981, 314. (c) Katritzky, A.
R.; Abdel-Fattah, A. A. A.; Tymoshenko, D. O.; Essawy, S. A. Synthesis
1999, 2114. (d) Borthakur, M.; Dutta, M.; Gogoi, S.; Boruah, R. C.
Synlett 2008, 3125. (e) Lieby-Muller, F.; Allais, C.; Constantieux, T.;
Rodriguez, J. Chem. Commun. 2008, 4207. (f) Allais, C.; Constantieux,
T.; Rodriguez, J. Chem.Eur. J. 2009, 15, 12945.
(9) See the Supporting Information for the full results.
(10) (a) Leonard, N. J.; Paukstelis, J. V. J. Org. Chem. 1963, 28, 3021.
(b) Childs, R. F.; Dickie, B. D. J. Am. Chem. Soc. 1983, 105, 5041.
(c) Lakhdar, S.; Tokuyasu, T.; Mayer, H. Angew. Chem., Int. Ed. 2008,
47, 8723.
(11) A single-step oxidative addition of the N−O bond to Cu(I) may
not be excluded (see ref 5b).
(12) (a) Tan, Y.; Hartwig, J. F. J. Am. Chem. Soc. 2010, 132, 3676.
(b) Takai, K.; Katsura, N.; Kunisada, Y. Chem. Commun. 2001, 1724.
(13) Alternatively, the intermediate 8 may form through nucleophilic
attack of the N atom of 7 to the iminium carbon of 4 followed by
[3,3]-sigmatropic rearrangement.
ASSOCIATED CONTENT
■
S
* Supporting Information
Detailed experimental procedures and characterization data.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(14) (a) Hill, M. D. Chem.Eur. J. 2010, 16, 12052. (b) Henry, G.
D. Tetrahedron 2004, 60, 6043.
AUTHOR INFORMATION
■
(15) For [2+2+2] cycloadditions, see: (a) Varela, J. A.; Saa, C. Chem.
Rev. 2003, 103, 3787. (b) Heller, B.; Hapke, M. Chem. Soc. Rev. 2007,
36, 1085. (c) Onodera, G.; Shimizu, Y.; Kimura, J.; Kobayashi, J.;
Ebihara, Y.; Kondo, K.; Sakata, K.; Takeuchi, R. J. Am. Chem. Soc.
2012, 134, 10515.
Corresponding Author
nyoshikai@ntu.edu.sg
Present Address
†College of Pharmacy, Third Military Medical University,
Chongqing 400038, China.
(16) For [4+2]-type reactions, see ref 5c and the following:
(a) Movassaghi, M.; Hill, M. D.; Ahmad, O. K. J. Am. Chem. Soc.
2007, 129, 10096. (b) Colby, D. A.; Bergman, R. G.; Ellman, J. A. J.
Am. Chem. Soc. 2008, 130, 3645. (c) Parthasarathy, K.; Jeganmohan,
M.; Cheng, C.-H. Org. Lett. 2008, 10, 325. (d) Hyster, T. K.; Rovis, T.
Chem. Commun. 2011, 47, 11846. (e) Too, P. C.; Noji, T.; Lim, Y. J.;
Li, X.; Chiba, S. Synlett 2011, 2789. (f) Chen, M. Z.; Micalizio, G. C. J.
Am. Chem. Soc. 2012, 134, 1352. (g) Yamamoto, S.; Okamoto, K.;
Murakoso, M.; Kuninobu, Y.; Takai, K. Org. Lett. 2012, 14, 3182.
(17) For [3+3]-type reactions, see: (a) Bagley, M. C.; Glover, C.;
Merritt, E. A. Synlett 2007, 2459. (b) Manning, J. R.; Davies, H. M. L.
J. Am. Chem. Soc. 2008, 130, 8602. (c) Wang, Y. F.; Chiba, S. J. Am.
Chem. Soc. 2009, 131, 12570.
(18) For cyclization of linear precursors, see: (a) Movassaghi, M.;
Hill, M. D. J. Am. Chem. Soc. 2006, 128, 4592. (b) Trost, B. M.;
Gutierrez, A. C. Org. Lett. 2007, 9, 1473. (c) Barluenga, J.; Fernandez-
Rodriguez, M. A.; Garcia-Garcia, P.; Aguilar, E. J. Am. Chem. Soc. 2008,
130, 2764. (d) Rizk, T.; Bilodeau, E. J. F.; Beauchemin, A. M. Angew.
Chem., Int. Ed. 2009, 48, 8325. (e) Nakamura, I.; Zhang, D.; Terada,
M. J. Am. Chem. Soc. 2010, 132, 7884.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the National Research Foundation Singapore (NRF-
RF-2009-05) and Nanyang Technological University for
financial support.
REFERENCES
■
(1) (a) Allen, A. E.; MacMillan, D. W. C. Chem. Sci. 2012, 3, 633.
(b) Shao, Z.; Zhang, H. Chem. Soc. Rev. 2009, 38, 2745.
́
(2) For selected examples, see: (a) Ibrahem, I.; Cordova, A. Angew.
Chem., Int. Ed. 2006, 45, 1952. (b) Allen, A. E.; MacMillan, D. W. C. J.
Am. Chem. Soc. 2010, 132, 4986. (c) Ikeda, M.; Miyake, Y.;
Nishibayashi, Y. Angew. Chem., Int. Ed. 2010, 49, 7289. (d) Van
Humbeck, J. F.; Simonovich, S. P.; Knowles, R. R.; MacMillan, D. W.
C. J. Am. Chem. Soc. 2010, 132, 10012. (e) Allen, A. E.; MacMillan, D.
W. C. J. Am. Chem. Soc. 2011, 133, 4260. (f) Skucas, E.; MacMillan, D.
W. C. J. Am. Chem. Soc. 2012, 134, 9090.
(19) (a) Nakao, Y. Synthesis 2011, 3209. (b) Bull, J. A.; Mousseau, J.
J.; Pelletier, G.; Charette, A. B. Chem. Rev. 2012, 112, 2642.
(3) (a) Ibrahem, I.; Santoro, S.; Himo, F.; Cor
Catal. 2011, 353, 245. (b) Ibrahem, I.; Breistein, P.; Cor
Angew. Chem., Int. Ed. 2011, 50, 12036. (c) Afewerki, S.; Breistein, P.;
Pirttila, K.; Deiana, L.; Dziedzic, P.; Ibrahem, I.; Cor
dova, A. Chem.
Eur. J. 2011, 17, 8784. (d) Ibrahem, I.; Ma, G.; Afewerki, S.; Cordova,
́
dova, A. Adv. Synth.
́
dova, A.
́
́
A. Angew. Chem., Int. Ed. 2013, 52, 878.
(4) Narasaka, K.; Kitamura, M. Eur. J. Org. Chem. 2005, 4505.
(5) (a) Tanaka, K.; Kitamura, M.; Narasaka, K. Bull. Chem. Soc. Jpn.
2005, 78, 1659. (b) Liu, S.; Yu, Y.; Liebeskind, L. S. Org. Lett. 2007, 9,
1947. (c) Liu, S.; Liebeskind, L. S. J. Am. Chem. Soc. 2008, 130, 6918.
(d) Too, P. C.; Chua, S. H.; Wong, S. H.; Chiba, S. J. Org. Chem. 2011,
D
dx.doi.org/10.1021/ja312346s | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX