Selective synthesis of N-substituted 1,2-dihydropyridines from furans by copper-induced concurrent tandem catalysis

Article abstract of DOI:10.1021/ja1006614

A novel transformation in which mono- or dialkyl-substituted furans are converted into 1,2-dihydropyridines upon reaction with PhI - NTs at room temperature is reported. The reaction is catalyzed by complexes of general formula Tp^xM (M = Cu, Ag) and consists of a one-pot procedure with four consecutive catalytic cycles. Furan aziridination is followed by aziridine ring-opening, transimination reaction, inverse-electronic-demand aza-Diels-Alder reaction, and a final hydrogen elimination reaction. The mechanism of the overall transformation is proposed where the metal complex displays a crucial role along the reaction pathway.

Full text of DOI:10.1021/ja1006614

Published on Web 03/15/2010
Selective Synthesis of N-Substituted 1,2-Dihydropyridines
from Furans by Copper-Induced Concurrent Tandem Catalysis
†
‡
,†
´
Manuel R. Fructos, Eleuterio Alvarez, M. Mar D´ıaz-Requejo,* and
Pedro J. Pe´rez*,†
Laboratorio de Cata´lisis Homoge´nea, Departamento de Qu´ımica y Ciencia de los Materiales,
Unidad Asociada al CSIC, Campus de El Carmen s/n, UniVersidad de HuelVa, 21007-HuelVa, Spain,
and Instituto de InVestigaciones Qu´ımicas, Centro de InVestigaciones Isla de La Cartuja, AVda
Americo Vespucio 49, 41092 SeVilla, Spain
Received June 29, 2009; E-mail: perez@dqcm.uhu.es; mmdiaz@dqcm.uhu.es
Abstract: A novel transformation in which mono- or dialkyl-substituted furans are converted into
1,2-dihydropyridines upon reaction with PhIdNTs at room temperature is reported. The reaction is catalyzed
by complexes of general formula TpxM (M ) Cu, Ag) and consists of a one-pot procedure with four
consecutive catalytic cycles. Furan aziridination is followed by aziridine ring-opening, transimination reaction,
inverse-electronic-demand aza-Diels-Alder reaction, and a final hydrogen elimination reaction. The
mechanism of the overall transformation is proposed where the metal complex displays a crucial role along
the reaction pathway.
Scheme 1. Nitrene Transfer Reactions
Introduction
The metal-catalyzed nitrene transfer reaction to saturated or
unsaturated substrates constitutes an emerging methodology in
organic synthesis.1,2 Thus, the addition of a nitrene moiety NR
to an olefin leads to the formation of aziridine rings (Scheme
1a). In the case of saturated substrates, the insertion of such
unit into a C-H bond affords a new carbon-nitrogen bond
(Scheme 1b), a reaction of enormous current interest.3 The role
of the catalyst in these transformations is the generation of
transient metallonitrene species (Scheme 1c) from nitrene
precursors, PhIdNTs being by far the most employed to date.
Such intermediates have scarcely been detected or isolated due
to their high reactivity.4
previously found that complexes of formulas TpxCu (Tpx )
hydrotrispyrazolylborate ligand)6 efficiently catalyze the olefin
aziridination reaction.7 On the other hand, we have also
described the use of TpxM (M ) Cu, Ag) complexes as catalysts
for the insertion of NTs groups into the C-H bond of several
substrates, including benzene, alkylaromatics, and plain alkanes.8,9
As a follow up of these findings, we thought it would be
We have been interested in the development of group 11
metal-based catalysts for the above transformations5 and have
† Universidad de Huelva.
‡ IIQ-CSIC.
(1) For reviews on this topic, see: (a) Zalatan, D. N.; Du Bois, J. Top.
Curr. Chem. 2010, DOI: 10.1007/128_2009_19. (b) Collet, F.; Dodd,
R. H.; Dauban, P. Chem. Commun. 2009, 5061. (c) Davies, H. M. L.;
Manning, J. R. Nature 2008, 451, 417. (d) Dick, A. R.; Sanford, M. S.
Tetrahedron 2006, 62, 2439. (e) Modern Rhodium-Catalyzed Organic
Reactions; Evans, P. A., Ed.; Wiley-VCH: Weinheim, 2005; p 379.
(f) Halfen, J. A. Curr. Org. Chem. 2005, 9, 657. (g) Mu¨ller, P.; Fruit,
C. Chem. ReV. 2003, 103, 2905.
(4) (a) Badiei, Y. M.; Dinescu, A.; Dai, X.; Palomino, R. M.; Heinemann,
F. W.; Cundari, T. R.; Warren, T. H. Angew. Chem., Int. Ed. 2008,
47, 9961. (b) Badiei, Y. M.; Krishnaswamy, A.; Melzer, M. M.;
Warren, T. H. J. Am. Chem. Soc. 2006, 128, 15056. (c) Leung, S. K.-
Y.; Tsui, W.-M.; Huang, J.-S.; Che, C.-M.; Liang, J.-L.; Zhu, N. J. Am.
Chem. Soc. 2005, 127, 16629. (d) Au, S.-M.; Huang, J.-S.; Yu, W.-
Y.; Fung, W.-H.; Che, C.-M. J. Am. Chem. Soc. 1999, 121, 9120.
(5) D´ıaz-Requejo, M. M.; Pe´rez, P. J. Chem. ReV. 2008, 108, 3379.
(6) (a) Trofimenko, S. Scorpionates, The Coordination Chemistry of
Polypyrazolylborate Ligands; Imperial College Press: London, 1999.
(b) Pettinari, C. Scorpionates II: Chelating Borate Ligands; Imperial
College Press: London, 2008.
(7) Mairena, M. A.; D´ıaz-Requejo, M. M.; Belderrain, T. R.; Nicasio,
M. C.; Trofimenko, S.; Pe´rez, P. J. Organometallics 2004, 23, 253.
(8) (a) Fructos, M. R.; Trofimenko, S.; D´ıaz-Requejo, M. M.; Pe´rez, P. J.
J. Am. Chem. Soc. 2006, 128, 11784. (b) D´ıaz-Requejo, M. M.;
Belderra´ın, T. R.; Nicasio, M. C.; Trofimenko, S.; Pe´rez, P. J. J. Am.
Chem. Soc. 2003, 125, 12078.
(2) For some key contributions in the area, see: (a) Liang, C.; Collet, F.;
Robert-Peillart, F.; Mu¨ller, P.; Dodd, R. H.; Dauban, P. J. Am. Chem.
Soc. 2008, 130, 343. (b) Williams, K.; Dubois, J. J. Am. Chem. Soc.
2007, 129, 562. (c) Lebel, H.; Huard, K.; Lectard, S. J. Am. Chem.
Soc. 2005, 127, 14198. (d) Espino, C. G.; Du Bois, J. Angew. Chem.,
Int. Ed. 2001, 40, 598. (e) Yu, X.-Q.; Huang, J.-S.; Zhou, X.-G.; Che,
C.-M. Org. Lett. 2000, 2, 2233. (f) Guthikonda, K.; Du Bois, J. J. Am.
Chem. Soc. 2002, 124, 13672. (g) Dauban, P.; Sanie`re, L.; Tarrade,
A.; Dodd, R. H. J. Am. Chem. Soc. 2001, 123, 7707. (h) Mu¨ller, P.;
Baud, C.; Jacquier, Y. Tetrahedron 1996, 52, 1543. (i) Evans, D. A.;
Faul, M. M.; Bilodeau, M. T. J. Am. Chem. Soc. 1994, 116, 2742.
(3) (a) Ley, S. V.; Thomas, A. W. Angew. Chem., Int. Ed. 2003, 42, 5400.
(b) Buchwald, S. L.; Mauger, C.; Mignani, G.; Scholz, U. AdV. Synth.
Catal. 2006, 348, 23.
9
4600 J. AM. CHEM. SOC. 2010, 132, 4600–4607
10.1021/ja1006614  2010 American Chemical Society

Conversion of Furans into 1,2-Dihydropyridines
A R T I C L E S
Scheme 2. Expected Products Derived from the Reaction of
2,5-Dimethylfuran and PhIdNTs
Figure 1. X-ray structure of 2. Thermal ellipsoids are drawn at the 30%
probability level.
interesting to study the effect that the presence of a heteroatom
in the substrate (the above systems were applied to saturated
of unsaturated hydrocarbons without any substituent) would
exert in these transformations, when both insertion or aziridi-
nation routes could be, in principle, accessible. With this idea
in mind, we screened the use of furans as substrates and found
an unexpected, novel transformation: the formation of 1,2-
dihydropyridine rings. In this paper, we describe this novel
process, which could be classified in the area of concurrent
tandem catalysis,10 the metal catalyst being crucial in several
steps of the overall transformation.
of silica gel, a clean, quantitative transformation of the bicyclic
1 into another compound took place. The NMR spectrum of
this compound was deceptively simple in terms of the number
of resonances: four singlets with integrals of 3 H each appeared
in the 2.0-1.5 ppm range, a AX spin system for two protons
(5.36, 3.31 ppm, J ) 8 Hz), two broad singlets (5.08, 4.57 ppm),
accounting for one proton each, and the set of resonances typical
for a tosyl group. Again, the overall 23 proton NMR spectrum
was in agreement with the formation of a product formed from
two molecules of 2,5-dimethylfuran and one NTs unit, but
different from 1. Single crystals were obtained by slow
evaporation of methylene chloride solutions, some of them being
suitable for a X-ray study.11 The structure of the molecule of 2
is shown in Figure 1 and has led to the identification of this
compound as the 1,2-dihydropyridine derivative 1-(4-acetyl-
1,2-dihydro-3,6-dimethyl-1-tosylpyridin-2-yl)propan-2-one), be-
ing consistent with the spectroscopic data available. In this
structure, the C4′ is a stereogenic carbon, showing the enanti-
omer R in the asymmetric unit, but the enantiomer S is also
present in the unit cell for symmetry of the inversion center of
the spatial group P1-, appearing this compound in the crystal
as a racemate as expected. The transformation of 1 into 2 (eq
2) formally consists of the elimination of a hydrogen atom
vicinal to the acetyl group. This step will be discussed in more
detail in the mechanistic proposal.
Results and Discussion
Reaction of 2,5-Dimethylfuran with PhIdNTs. We first
studied the reaction of 2,5-dimethylfuran with PhIdNTs in the
presence of the complex TpBr3Cu(NCMe) as the catalyst,
expecting the formation of the aziridine and/or the tosylamide
(Scheme 2) derived from the addition or the insertion of the
nitrene group in that substrate, respectively. This copper complex
has been already described as a good catalyst for both
transformations.7,8 The reaction was carried out in CH2Cl2 as
the solvent, at room temperature, with a 1:20:400 ratio of [Cu]/
[PhIdNTs]/[2,5-dimethylfuran] being employed. The nitrene
precursor dissolved within 1 h to give a yellow solution that
turned reddish-orange with time. After 20 h of stirring, volatiles
were removed to give a reddish oil that was investigated by
NMR, showing the existence of one unique set of resonances.
Isolation of this new compound 1 as an off-white solid was
achieved by silica gel chromatography. Extensive multinuclear,
1D, and 2D NMR studies have led to the proposal of a bicyclic
structure that has been characterized as 1-(3a,4,7,7a-tetrahydro-
2,5-dimethyl-4-tosylfuro[3,2-b]pyridin-7-yl)ethanone. Such com-
position formally derives from two molecules of 2,5-dimeth-
ylfuran and one NTs unit (eq 1):
Catalyst Screening. After the observation of the formation
of dihydropyridine derivatives from 2,5-dimethylfuran and
PhIdNTs in the presence of TpBr3Cu(NCMe), we evaluated the
effect of the transition metal complex in this transformation. A
series of complexes of formula TpxM (Scheme 3) were
employed following the above procedure, with 2,5-dimethyl-
furan as the substrate. In all cases, the bicyclic compound 1
was obtained upon stirring at room temperature for 20 h.
However, the use of the asymmetric 2-methylfuran as the
Interestingly, during the process of purification, we unexpect-
edly observed that when neutral alumina was employed instead
(9) Go´mez-Emeterio, B. P.; Urbano, J.; D´ıaz-Requejo, M. M.; Pe´rez, P. J.
Organometallics 2008, 27, 4126.
(11) CCDC 713086 contains the supplementary crystallographic data for
compound 2. These data can be obtained free of charge from the
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
(10) Wasilke, J.-C.; Obrey, S. J.; Baker, R.; Baza´n, G. C. Chem. ReV. 2005,
105, 1001.
9
J. AM. CHEM. SOC. VOL. 132, NO. 13, 2010 4601

A R T I C L E S
Fructos et al.
Scheme 3. Ligands Employed in This Work
mixtures of compounds, not only at room temperature but also
at -10 °C. At this temperature, in addition to other species, a
minor set of signals similar to those observed for the dihydro-
pyridines 2 and 4 were observed in the 1H NMR spectrum, but
accounting for less than 10% of the overall mixture.
2-Ethylfuran provided very interesting information, since the
presence of methylenic protons (CH2) could offer a reaction
site for the insertion of the NTs moiety to occur. However, a
clean transformation took place, and a mixture of the bicyclic
and dihydropyridine compounds 5 and 6 was obtained, favoring
the latter in a ratio similar to that previously observed with
2-methylfuran. Therefore, this result shows that the system is
not restricted to furans with methyl groups as substituents but
also is compatible with more reactive methylenic groups.
As already discussed, 2,5-dimethylfuran gave only the
bicyclic compound 1, further transformed into the dihydropy-
ridine 2 with the aid of neutral alumina. Another disubstituted
furan has also been investigated, that with two methyl groups
located at positions 2 and 3. Again, only the related bicyclic
compound 7 was quantitatively obtained. However, at variance
with 1, all attempts to convert 7 into a dihydropyridine skeleton
failed (prolonged heating or treatment with alumina).
Two more furans were studied, ethyl furan-2-carboxylate and
benzofuran. In both cases, the reaction did not take place toward
the expected products. In contrast, all of the PhIdNTs reagent
decomposed into TsNH2. Several N-Me-pyrrolidines and me-
thylthiophenes have been also screened under similar conditions,
but a distinct reactivity has been observed. Pyrrolydines
underwent insertion into the C-H bond of the N-Me moiety,
whereas in the case of 2-methylthiophene or 2,5-dimethylth-
iophene, products derived from the insertion into the methyl
groups were obtained. This is in agreement with the strong
delocalization in the pyrrol- and thiophene rings, in contrast
with that of furans. The transformation reported herein requires
loss of aromaticity of the starting heterocycle, and therefore,
those with strong delocalization (pyrrole, thiophene) are less
prone to undergo this reaction.
substrate has provided very interesting information. In a manner
similar to that for the disubstituted furan, two compounds have
been obtained from the reaction of 2-methylfuran and PhIdNTs
(eq 3), the bicyclic 3 and the 1,2-dihydropyridine 4, i.e., the
analogues to 1 and 2, respectively. But at variance with that,
after 20 h of stirring at room temperature, mixtures of 3 and 4
were obtained in a variable ratio depending of the catalyst
employed (Table 1). Thus, TpMsCu(NCMe) induced the quan-
titative formation of 4, whereas in the case of the Tp*-, TpBr3-,
or Tp*,Br-copper catalysts, a certain, although minor (5-10%),
amount of 3 could be detected. More interestingly, this
selectivity could be reversed by using silver instead of copper:
the complex Tp*,BrAg provided a 90:10 ratio of 3 and 4,
respectively, allowing the full characterization of the former. It
is worth mentioning that these values correspond to identical
reaction times (20 h) and that when prolonged for several days,
the dihydropyridine 4 was the sole product isolated. In addition,
any uncompleted reaction after those 20 h could be instanta-
neously completed toward 4 by treatment with neutral alumina,
similar to the aforementioned 1f2 conversion. Reaction times
can be dramatically diminished by increasing the temperature.
Thus, the reaction of 2-methylfuran with PhIdNTs in the
presence of TpBr3Cu(NCMe) as the catalyst carried out at 60
°C was completed in 2 h, with quantitative conversion into 4.
On the other hand, when the reaction was performed at 5-10
°C, a mixture of 3 and 4 with 3 as the major product (ca. 75%)
was obtained.
Compounds 2, 4, and 6 are 1,2-dihydropyridine derivatives
and have not been previously described. The 1,4-dihydropyridine
skeleton is found in a large number of pharmaceuticals and
therefore has been largely studied.12 In contrast, the related 1,2-
dihydropyridines have not been so far developed in the same
manner. The number of synthetic procedures reported to date
is small,13,14 with only a few being known to generate
1,2-dihydropyridines in a regioselective manner.15 Compared
to them, the method described in this contribution is quite simple
and is performed under very mild conditions. The synthetic
potential of this system is yet to be developed, although it is
important to note at this point that the existence of acetyl groups
in the easily accessible compounds 2, 4, and 6 provides an
additional entry for further derivatization.
Extension to Other Furans and Heterocycles. Since the
previous results were obtained with substrates bearing methyl
substituents at the position vicinal to oxygen, we wondered
if that would be a restriction of the system. Therefore, a series
of different furans, shown in Scheme 4, were employed as
starting materials, under identical reaction conditions, with
TpBr3Cu(NCMe) as the catalyst and at room temperature. The
use of unsubstituted, plain furan as the substrate led to intractable
Mechanistic Studies. In order to gain useful information about
the mechanism that governs this novel transformation, we first
Table 1. Selectivity of the Reaction of 2-Methylfuran and PhIdNTs
1
in the Presence of TpxM Complexes As Catalystsa
monitored by H NMR the reaction of 2,5-dimethylfuran with
catalyst
3 (%)
4 (%)
(12) (a) Lavilla, R. J. Chem. Soc., Perkins Trans. 1 2002, 1141. (b) Gaudio,
A. C.; Korolskovas, A.; Takahata, Y. J. Pharm. Sci. 1994, 83, 1110.
(13) (a) Stout, D. M.; Meyers, A. I. Chem. ReV. 1982, 82, 233. (b) Comins,
D. L.; Hong, H.; Salvador, J. M. J. Org. Chem. 1991, 56, 7197.
(14) (a) Ichikawa, E.; Suzuki, M.; Yabu, K.; Albert, M.; Kanai, M.;
Shibasaki, M. J. Am. Chem. Soc. 2004, 126, 11808. (b) Brunner, B.;
Stoigaitis, N.; Lautens, M. Org. Lett. 2006, 8, 3473.
TpBr3Cu(NCMe)
Tp*Cu
5
5
<1
10
90
95
95
>99
90
TpMsCu(NCMe)
Tp*,BrCu(NCMe)
Tp*,BrAg
10
a Reaction conditions: 0.0125 mmol of the catalyst, 0.25 mmol of
PhIdNTs, 10 mmol of 2-methylfurane, 2 mL in CH2Cl2. Temperature:
20 °C. Reaction time: 20 h.
(15) (a) Motamed, N.; Bunnelle, E. M.; Singaram, S. W.; Sarpong, R. Org.
Lett. 2007, 9, 2167. (b) Wan, J.-P.; Gan, S.-F.; Sun, G.-L.; Pan, Y.-J.
J. Org. Chem. 2009, 74, 2862.
9
4602 J. AM. CHEM. SOC. VOL. 132, NO. 13, 2010

Conversion of Furans into 1,2-Dihydropyridines
Scheme 4. Reactivity of Several Furans with PhIdNTs in the Presence of TpBr3Cu(NCMe) as the Catalysta b
A R T I C L E S
,
a Reaction conditions: 0.0125 mmol of the catalyst, 0.25 mmol of PhIdNTs, 10 mmol of the furan, 2 mL in CH2Cl2. Temperature: 20 °C. Reaction time:
20 h. All PhIdNTs was converted into products; percentages correspond to distribution of products. b nro: no reaction observed. All PhINTs decomposed
into TsNH2.
Figure 2. 1H NMR spectrum (CDCl3, 400 MHz) of the reaction of
2-methylfuran and PhIdNTs (equimolar ratio) using TpBr3Cu(NCMe) as
the catalyst after 10 min of stirring at room temperature.
PhIdNTs, using TpBr3Cu(NCMe) as the catalyst. Unfortunately,
only the resonances of 1 were observed along the reaction time,
with no other intermediate being clearly detected (the aromatic
Figure 3. Monitoring of the reaction mixture of 2-methylfuran and
PhIdNTs (equimolar ratio) using TpBr3Cu(NCMe) as the catalyst with time
(downfield region).
region is crowded in this chemical system). We were more
fortunate when moving to 2-methylfuran as the substrate. In
this case, an NMR-scaled reaction (0.0125 mmol of the catalyst,
initial mixture was constituted by equimolar amounts of
2-methylfuran and PhIdNTs).
0.25 mmol of PhIdNTs, 0.25 mmol of 2-methylfuran, 1 mL of
CDCl3) with equimolar amounts (eq 4) of the furan and
PhIdNTs showed that upon dissolution of PhIdNTs (10 min)
the reaction mixture was constituted by two major species, (2Z)-
4-tosyliminopent-2-enal (OI1) and (3E)-5-tosyliminopent-3-en-
2-one (OI2), the exact geometry being assigned from the values
of the coupling constants (see the Experimental Section). Figure
2 shows the NMR spectrum of the reaction mixture after 10
min of reaction. Some residual 2-methylfuran was also observed,
along with PhI derived from PhIdNTs. Monitoring of the
reaction mixture showed a smooth transformation of the
aldehyde OI1 into the imine OI2 (Figure 3), the latter being
the unique species detected after 3 h at room temperature (the
Once demonstrated that the aldehyde OI1 and the imine OI2
are intermediates in this transformation, we decided to inves-
tigate the next step, i.e., the reaction of the imine with a second
equivalent of 2-methylfuran that was added to the reaction
mixture. However, the reaction was extremely slow. After some
experiments in which a 5- or 10-fold excess of 2-methylfuran
was added to the imine solution, we found that at least 20 equiv
9
J. AM. CHEM. SOC. VOL. 132, NO. 13, 2010 4603

A R T I C L E S
Fructos et al.
intermediate previous to OI1 in the catalytic cycle, we believe
that aziridination occurred as the first step in this transformation.
This proposal is based in the reactivity already described for
the oxidation of furan with TS1-H2O2,16 peracids,17 or diox-
iranes,18 for which the mechanism shown in Scheme 5a has
been proposed. In these cases, an epoxide intermediate, which
has not been detected, undergoes a rearrangement to yield
unsaturated dicarbonylic species.16-18 In our system, the addition
of the nitrene moiety would lead to the formation of a bicyclic
aziridine that could be also unstable. However, given the
asymmetry of the 2-methylfuran molecule, aziridination could
in principle take place in both double bonds of the ring, leading
to AZ1 and AZ2 (Scheme 5b). Subsequent, spontaneous ring
opening, similar to the aforementioned epoxide, would lead to
the open intermediates OI1 and OI2, respectively. It is worth
mentioning that OI1 would form from the more sterically
hindered but also more electron-rich substituted double bond.
This could explain that at the early stages of the reaction, OI1
predominates. Unfortunately, the picture in Scheme 5b cannot
explain the experimentally observed conversion of OI1 into OI2,
unless an additional process arises. This could be a transimi-
nation reaction,19 shown in Scheme 6, in which the presence
of adventitious water should be responsible for such conversion
in a catalytic manner. Alternatively, the acid nature of the TpxM
active catalyst could also induce this transformation. We cannot
rule out that coordination of the intermediate OI1 to the metal
center could also have a certain influence to activate OI1 and
trigger this transformation.
Figure 4. Monitoring of the disappearance of the imine OI2 with (square)
and without (circle) additional catalyst. Inset: plot of Ln[OI2] vs time (three
half-lifes).
of the furan should be added in order to induce the transforma-
tion at an acceptable reaction rate. Under those conditions,
complete transformation of the imine into the final product 4
was observed, with a steady-state concentration of the bicyclic
3 detected along the reaction time. In the case of the silver
catalyst Tp*,BrAg, accumulation of 3 was observed, the 3f4
conversion being significantly lower than for the copper-based
catalysts (see Table 1).
(b) Step 2: Formation of the Bicyclic Compound 3. The
second step of this transformation can be explained as an aza-
Diels-Alder reaction (ADAR) with inverse-electron-demand20,21
between the intermediate OI2 and a second molecule of
2-methylfuran(Scheme7).Theexclusiveformationof1-(3a,4,7,7a-
tetrahydro-2-methyl-4-tosylfuro[3,2-b]pyridin-7-yl)ethanone (3),
along with the lack of observance of the product derived from
the ADAR reaction with the more hindered double bond of the
furan, evidence the existence of a certain steric effect in this
step. The same behavior was found when 2-ethylfuran or 2,3-
dimethylfuran were employed as the substrates, the substituents
occupying the external double bond in the final bicyclic products
5 and 7, respectively. As inferred from the results shown in
Figure 4, this step seems to be affected by the presence of the
metal-center, the TpxM fragments displaying a certain degree
of Lewis basicity. As a precedent for this metal-induced ADAR,
very recently, Carretero and co-workers have reported22 the use
of a nickel-mediated ADAR with inverse-electron-demand in
which the metal catalyst also displays such acidic behavior.
(c) Step 3: Generation of the Dihydropyridine 4. The use of
2-methyl- or 2-ethylfuran as substrates has provided dihydro-
A question that arises from these observations is the role of
the metal center in this step, which could be mediated by the
TpBr3Cu moiety, a 16-electron fragment that would display a
certain degree of Lewis acidity. We have tried to isolate the
imine OI2 from the reaction mixture of 2-methylfuran and
PhIdNTs, with no success: decomposition was observed in all
cases. Independent synthesis16 has also failed, and therefore,
the direct reaction of a pure sample of the imine with
2-methylfuran in the presence and in the absence of the catalyst
could not be performed. However, we have done a related
experiment in the following manner (see the Experimental
Section). The imine OI2 was generated as above (2-methylfuran
+ PhIdNTs, in the presence of TpBr3Cu(NCMe) as the catalyst),
and the solution was divided in two identical parts. Additional
TpBr3Cu(NCMe) was added to one of the aliquots, and then 40
equiv of 2-methylfuran was added to both solutions. The
disappearance of the imine resonance (and the concomitant
1
appearance of the dihydropyridine 4) was monitored by H
NMR. Figure 4 shows the decrease of the concentration of imine
with time, where it is clearly observable that the concentration
of the catalyst seems crucial: the reaction with a higher
concentration of the copper complex consumed the imine at a
higher rate (25% increase). We interpret these results as a proof
of the involvement of the copper complex in the formation of
the bicyclic compound 3.
Mechanistic Proposal. (a) Step 1: Formation of the
Aldehyde OI1 and the Imine OI2. The intermediate (2Z)-4-
tosyliminopent-2-enal (OI1) corresponds to the addition of a
molecule of 2-methylfuran and the NTs unit, although there is
no obvious, direct reaction that could explain the formation of
such molecule. In spite of the lack of observation of any
(17) Kobayashi, Y.; Katsuno, F.; Sato, F. Chem. Lett. 1983, 1771.
(18) (a) Adger, B. M.; Barret, C.; Brennan, J.; McKervey, M. A.; Murray,
R. W. J. Chem. Soc., Chem. Commun. 1991, 1553. (b) Adger, B. M.;
Barret, C.; Brennan, J.; McKervey, M. A.; Tarbit, B. J. Chem. Soc.,
Chem. Commun. 1993, 1220.
(19) Hogg, J. L.; Jenks, D. A.; Jenks, W. P. J. Am. Chem. Soc. 1977, 99,
4772.
(20) (a) Schnermann, M. J.; Boger, D. L. J. Am. Chem. Soc. 2005, 127,
15704. (b) Hong, J.; Boger, D. L. J. Am. Chem. Soc. 1998, 120, 1218.
(c) Boger, D. L.; Cassidy, K. C.; Nakahara, S. J. Am. Chem. Soc.
1993, 115, 10733. (d) Boger, D. L.; Corbett, W. L.; Curran, T. T.;
Kasper, A. M. J. Am. Chem. Soc. 1991, 113, 1713. (e) Boger, D. L.;
Corbett, W. L.; Wiggins, J. M. J. Org. Chem. 1990, 55, 2999. (f)
Boger, D. L.; Kasper, A. M. J. Am. Chem. Soc. 1989, 111, 1517.
(21) Behforouz, M.; Ahmadian, M. Tetrahedron 2000, 56, 5259.
(22) Esquivias, J.; Go´mez-Arraya´s, R.; Carretero, J. C. J. Am. Chem. Soc.
2007, 129, 1480.
(16) We have employed a procedure similar to that reported for the synthesis
of unsaturated 1,4-diketones. See: Whalen, J.; Moens, B.; De Vos,
D. E.; Alsters, P. L.; Jacobs, P. A. AdV. Synth. Catal 2004, 346, 333.
9
4604 J. AM. CHEM. SOC. VOL. 132, NO. 13, 2010

Conversion of Furans into 1,2-Dihydropyridines
A R T I C L E S
Scheme 5. (a) Instability of Epoxides Derived from Furans. (b) Formation of OI1 and OI2 through Aziridine Intermediates
Scheme 6. Transimination Pathway That Converts OI1 into OI2
new enol (EN2) which further tautomerizes to the more stable
ketone 4. This would be in good agreement with the lack of
reactivity of 7, due to the presence of a methyl group as R3.
Interestingly, the use of copper- or silver-based catalysts has
provided a different selectivity at the end of the reaction, the
silver one leading to the bicyclic compound 3 as the major
product. Since the unique difference between the experiments
was the catalyst, we believe that the metal center is affecting
this last elimination step. Elimination reactions are easily
induced by acids. Therefore, the differences must stand on the
distinct acidic nature of the metal center, i.e., Cu(I) or Ag(I)
ions in a tetrahedral environment. A simple estimation of the
acidity based in Z2/r values23 (rCu(I) ) 74 pm, rAg(I) ) 114 pm)24
allows us to propose a higher acidity for the copper center, with
the corresponding effect in the conversion 3 f 4. On the other
hand, the acidity of Ag(I), moderate compared with that of Cu(I),
was not enough to promote such conversion at a significant rate,
compound 3 remaining as the major component of the reaction
mixture after 24 h. The need of neutral alumina to promote the
conversion of 1 into 2 also favors the proposal of an acid-
catalyzed elimination step: in this case, the presence of the
additional methyl groups probably provides additional stability
to the bicyclic compound 1.
pyridines 4 and 6 as the final products. The related compound
1, derived from 2,5-dimethylfuran, was converted into the six-
membered ring with the aid of neutral alumina. On the other
hand, the bicylic compound 7, formed from 2,3-dimethylfuran
as the substrate, has not been converted into a dihydropyridine
derivative in any case. The only difference in this case is the
existence of a methyl group instead of a hydrogen as R3, vicinal
to the acetyl group. Therefore, a plausible explanation for the
formation of the dihydropyridines should invoke the participa-
tion of that H atom at R3, in the form of an elimination reaction.
Scheme 8 shows a possible reaction pathway for this step. An
enol (EN1) would form at the first stage, followed by a
ꢀ-elimination ring-opening mechanism that would end into a
The Global Mechanism. On the basis of the above, the general
mechanism for the reaction of 2-methylfuran and PhIdNTs in
the presence of TpxM (M ) Cu, Ag) complexes is shown in
Scheme 9. Four consecutive catalytic cycles I-IV take place,
Scheme 7. Proposal for the Conversion of the Imine Intermediates into the Bicyclic Compounds 3, 5, and 7
9
J. AM. CHEM. SOC. VOL. 132, NO. 13, 2010 4605

A R T I C L E S
Fructos et al.
Scheme 8. Final Elimination Step to Convert 3 into 4
four concurrent catalytic cycles in a completely unprecedented
transformation. Work aimed to expand the scope of this
transformation is currently underway in our laboratory.
Experimental Section
General Methods. All preparations and manipulations were
carried out under an oxygen-free nitrogen atmosphere using
conventional Schlenk techniques or inside a drybox. All the
substrates were purchased from Aldrich. Substrates and solvents
were rigorously dried previously to their use. The copper and silver
complexes employed as catalysts were prepared according to the
literature.7-9 PhIdNTs was also prepared following the previously
reported methods.25 NMR experiments were run in a Varian
Mercury 400 MHz spectrometer. GC data were collected with a
Varian GC-3900 with a TSD detector.
Reaction of 2,5-Dimethylfuran with PhIdNTs in the Pres-
ence of TpBr3Cu(NCMe). Synthesis of 1-(3a,4,7,7a-Tetrahydro-
2,5,7a-trimethyl-4-tosylfuro[3,2-b]pyridin-7-yl)ethanone (1). A
0.025 mmol (25.6 mg) portion of the copper complex was dissolved
in 2.0 mL of 2,5-dimethylfuran and 2 mL of CH2Cl2. To the
colorless stirred solution was added 186 mg (0.5 mmol) of PhIdNTs
as yellow crystals. The mixture became greenish in a few seconds,
and the color slowly turned orange while the nitrene precursor
dissolved into the solution. After 2 h of stirring at room temperature,
a homogeneous, red solution was obtained. Volatiles were removed
after 20 h of stirring under vacuum, and the residue was washed
several times with petroleum ether to give compound 1 as a reddish,
somewhat oily solid in 78% yield. This procedure was employed
for the catalyst screening set of experiments showed in Table 1.
1H NMR (CDCl3, 400 MHz): 7.70 (d, 2H, J ) 8 Hz), 7.30 (d, 2H,
J ) 8 Hz), 5.36 (d, 1H, J ) 8 Hz), 5.08 (br s, 1 H), 4.57 (br s,
1H), 3.31 (d, 1 H, J ) 8 Hz), 2.41 (s, 3H, Me), 2.0 (s, 3H, Me),
1.92 (s, 3H, Me), 1.70 (s, 3H, Me), 1.50 (s, 3H, Me). 13C{1H} NMR
(CDCl3, 100 MHz): 13C{1H} NMR (CDCl3, 100 MHz): 143.4,
137.6, 130.5 (quaternary), 129.7, 127.7, (aromatic, C-H), 113.0,
55.8 (olefinic C-H), 69.4 (C-H), 55.8 (C-H), 29.7 (COCH3), 22.2,
21.7, 13.9 (CH3). Anal. Calcd for C19H23SO4N: C, 63.13; H, 6.41;
N, 3.88; S, 8.87. Found: C, 63.18; H, 6.49; N, 3.81; S, 8.89.
Conversion of 1-(3a,4,7,7a-Tetrahydro-2,5,7a-trimethyl-4-
tosylfuro[3,2-b]pyridin-7-yl)ethanone (1) into 1-(4-Acetyl-1,2-
dihydro-3,6-dimethyl-1-tosylpyridin-2-yl)propan-2-one (2). The
product obtained in the previous reaction (0.4 mmol) was dissolved
in methylene chloride and passed through a column filled with
neutral alumina. The resulting red solution was concentrated and
cooled at -20 °C to give off-white crystals of 1-(4-acetyl-1,2-
dihydro-3,6-dimethyl-1-tosylpyridin-2-yl)propan-2-one (2). Slow
evaporation of a CH2Cl2/petroleum ether solutions gave single
Scheme 9. Overall Mechanism for the Concurrent Tandem
Catalytic System to Convert 2-Methaylfuran and PhIdNTs into
Dihydropyridine 4
at least three of them promoted by the metal catalyst. Cycle (I)
corresponds to furan aziridination and ring-opening to yield
aldehyde OI1. Cycle II, for which either adventitious water or
the metal complex could act as the catalyst, converts OI1 into
the imine OI2. Cycle III corresponds to a metal-catalyzed aza-
Diels-Alder reaction with inverse electronic demand, which
yields the bicyclic compound 3. Finally, the latter undergoes a
metal-catalyzed elimination reaction to afford the 1,2-dihydro-
pyridine 4. According to Baker, Bazan, and co-workers,10 the
overall mechanism can be designed under the term “concurrent
tandem catalysis” on the basis of the existence of cooperative
action of four catalytic cycles in a single reactor. The mechanism
can be applied with success to the series of furans discussed in
the previous sections.
1
crystals suitable for a X-ray study. H NMR (CDCl3, 400 MHz):
7.57 (d, 2H, J ) 8 Hz), 7.21 (d, 2H, J ) 8 Hz), 5.93 (s, 1H), 4.96,
2.52, 2.41 (ABM spin system, δA ) 2.41, δB ) 2.52, δM ) 4.96,
JAB ) 15 Hz, JAM) 10 Hz, JBM ) 5 Hz), 2.36 (s, 3H, CH3), 2.25
(s, 3H, COCH3), 2.25 (s, 3H, COCH3), 1.83 (s, 3H, CH3), 1.56 (s,
3H, CH3). 13C{1H} NMR (CDCl3, 100 MHz): 144.3, 138.3, 135.5,
132.9, 129.9 (quaternary C), 127.3, 127.1 (aromatic, C-H), 116.4
(olefinic C-H), 58.2 (C-H), 43.9 (CH2), 30.4, 29.9 (COCH3), 22.9,
21.7, 19.3 (CH3). Anal. Calcd for C19H23SO4N: C, 63.13; H, 6.41;
N, 3.88; S, 8.87. Found: C, 63.27; H, 6.51; N, 3.81; S, 9.05.
CCDC 713086 contains the supplementary crystallographic data
for compound 2. These data can be obtained free of charge from
the Cambridge Crystallographic Data Centre via www.ccdc.ca-
m.ac.uk/data_request/cif.
Conclusions
We have discovered that complexes TpxCu(NCMe) or related
TpxM (M ) Cu, Ag) catalyze the reaction of several mono- or
disubstituted dialkylfurans with PhIdNTs, at room temperature,
to provide novel 1,2-dihydropyridines derived from a series of
Reaction of 2-methylfuran with PhIdNTs in the Presence
of TpBr3Cu(NCMe). A 0.025 mmol (25.6 mg) portion of the copper
complex was dissolved in 1.85 mL of 2-methylfuran and 2 mL of
CH2Cl2. To the colorless stirred solution was added 186 mg (0.5
mmol) of PhIdNTs as yellow crystals. The mixture became
(23) Huheey, J. E.; Keiter, E. A.; Keiter, R. L. Inorganic Chemistry:
Principles of Structure and ReactiVity, 4th ed.; Prentice Hall: Upper
Saddle River, NJ, 1997.
(24) Wiberg, N.; Holleman, A.; Wiberg, E. Inorganic Chemistry; Academic
Press: New York, 2001.
(25) Yamada, Y.; Yamamoto, T.; Okawara, M. Chem. Lett. 1975, 361.
9
4606 J. AM. CHEM. SOC. VOL. 132, NO. 13, 2010

Conversion of Furans into 1,2-Dihydropyridines
A R T I C L E S
greenish in a few seconds, and the color slowly turned orange while
the nitrene precursor entered the solution in a few minutes. The
mixture was stirred for 20 h. Volatiles were removed under vacuum,
and the residue was investigated by NMR showing the formation
of 1-(3a,4,7,7a-tetrahydro-2-methyl-4-tosylfuro[3,2-b]pyridin-7-yl)-
ethanone (3) and 1-(4-acetyl-1,2-dihydro-1-tosylpyridin-2-yl)pro-
pan-2-one (4) (ratio 3/4 5:95). Attempts to separate both compounds
failed, both by crystallization or chromatography. In fact, passing
the mixture through silica gel or neutral alumina led to the
quantitative conversion of 3 into 4. The same result was achieved
by stirring the reaction mixture for 3-4 days at room temperature.
However, the use of the silver complex Tp*,BrAg as the catalyst
provided a 90:10 mixture of 3 and 4, respectively, from which 3
could be completely characterized by NMR spectroscopy.
nary C), 130.0, 126.9 (aromatic, C-H), 128.8, 126.3, 105.7 (olefinic
C-H), 50.0 (C-H), 48.3 (CH2), 36.5, 30.6 (COCH3), 8.3, 7.7 (CH3).
Mass spectrum (CI): m/z calcd for C19H23NO4S 362.1426, found
362.1419.
Reaction of 2,3-Dimethylfuran with PhIdNTs in the Pres-
ence of TpBr3Cu(NCMe). With the above procedure, the reaction
was performed with 2,3-dimethylfuran, leading to the isolation of
an oily residue that has been identified as 1-(3a,4,7,7a-tetrahydro-
2,3,7-trimethyl-4-tosylfuro[3,2-b]pyridin-7-yl)ethanone (7). This
bicyclic compound could not be further transformed into a
dihydropyridine by treatment with alumina or by prolonged heating.
1H NMR (CDCl3, 400 MHz): 7.8 (d, 2H, J ) 8 Hz), 7.26 (d, 2H,
J ) 8 Hz), 6.50 (d, 1H, J ) 8 Hz), 5.18 (d, 1H, J ) 8 Hz), 5.08
(d, 1H, J ) 8 Hz), 4.95 (d, 1H, J ) 8 Hz), 2.4 (s, 3H, CH3), 1.75
(s, CH3), 1.71 (s, CH3), 1.35 (s, CH3), 1.27 (s, CH3). 13C{1H} NMR
(CDCl3, 100 MHz): 208.0 (carbonylic CO),151.4, 148.2, 143.5,
139.6, 132.7 (quaternary C), 130.0, 126.9 (aromatic, C-H), 128.5,
118.6, (olefinic C-H), 85.6, 65.9 (C-H), 27.8, 21.6, 11.6, 11.4, 10.5
(CH3). Mass spectrum (CI): m/z calcd for C19H23NO4S 362.1426,
found 362.1440.
1-(3a,4,7,7a-Tetrahydro-2-methyl-4-tosylfuro[3,2-b]pyridin-7-
1
yl)ethanone (3). H NMR (CDCl3, 400 MHz): 7.8 (d, 2H, J ) 8
Hz), 7.26 (d, 2H, J ) 8 Hz), 6.75 (d, 1H, J ) 8 Hz), 5.25 (d, 1H,
J ) 8 Hz), 4.98 (d, 1H, J ) 8 Hz), 4.78 (d, 1H, J ) 8 Hz), 3.4 (dd,
1H, J ) 8 Hz), 2.3 (s, 3H, CH3), 1.95 (s, 3H, COCH3), 1.78 (s,
3H, COCH3). 13C{1H} NMR (CDCl3, 100 MHz): 204.1 (carbonylic
CO), 158.5, 144.1, 136.2 (quaternary C), 129.8, 128.7 (aromatic,
C-H), 127, 104.8, 81.4 (olefinic C-H), 100.5 (C-H), 60.4 (C-H),
48.7 (CH), 28.4 (COCH3), 21.8 (CH3), 13.7 (CH3).
NMR Monitoring Experiments. A 0.0125 mmol portion of the
TpBr3Cu(NCMe), 0.25 mmol of PhIdNTs, and 0.25 mmol of
2-methylfuran were dissolved, inside of a drybox, in 1 mL of CDCl3
in a vial. After 5 min of stirring, a orange solution was obtained
that was transferred into an NMR tube. The reaction was monitored
1-(4-Acetyl-1,2-dihydro-1-tosylpyridin-2-yl)propan-2-one (4). 1H
NMR (CDCl3, 400 MHz): 7.67 (d, 2H, J ) 8 Hz), 7.28 (d, 2H, J
) 8 Hz), 6.72 (d, 1H, J ) 8 Hz), 5.98 (d, 1H, J ) 8 Hz), 6.36,
1
by H NMR (Figures 2 and 3). From the first spectrum, 10 min
5.10, 3.09, 2.82 (ABMX spin system, δA ) 2.82, δB ) 3.09, δM
)
after the beginning of the reaction, OI1 and OI2 could be detected.
Selected 1H NMR resonances for (2Z)-4-tosyliminopent-2-enal
(OI1): δ 9.32 (d, H3, J ) 9.3 Hz), 6.52 (dd, H2, J ) 11.3, 9.3 Hz),
5.10, δX ) 6.36, JAB ) 18 Hz, JAM ) 4 Hz, JBM ) 8 Hz, JMX ) 6
Hz), 2.41 (s, 3H, CH3), 2.15 (s, 3H, COCH3), 2.15 (s, 3H, COCH3).
13C{1H} NMR (CDCl3, 100 MHz): 205.6, 195.5 (carbonylic CO),
144.3, 136.2, 133.1 (quaternary C), 129.7, 126.4 (aromatic, C-H),
129.6, 126.1, 105.0 (olefinic C-H), 49.7 (C-H), 49.4 (CH2), 30.2,
25.1 (COCH3), 21.5 (CH3). Mass spectrum (FAB): m/z [M+ + 1]
334, [M+ + Na] 356, [M+ + K] 372.
1
6.83 (d, H1, J ) 11.3 Hz). See Figure 2 for labeling. Selected H
NMR resonances for (3E)-5-tosyliminopent-3-en-2-one (OI2): δ
8.71 (d, H6, J ) 9 Hz), 6.81 (d, H4, J ) 15 Hz). H5 is obscured by
aromatic resonances. See Figure 2 for labeling.
Effect of the Metal Catalyst in the Reaction of OI2 with
2-Methylfuran. A solution identical to that described in the
previous paragraph was prepared and strred until only OI1 was
present. At that point, the solution was divided in two identical
aliquots of 0.5 mL each, and 0.5 mL of CDCl3 was added to re-
establish the initial concentration values in both samples. Additional
TpBr3Cu(NCMe) (0.0125 mmol) was added to one of the aliquots.
Reaction of 2-Ethylfuran with PhIdNTs in the Presence of
TpBr3Cu(NCMe). The procedure was analogous to that previously
described for 2-methylfuran. Again, two products were obtained
after 20 h of stirring, in a 95:5 ratio. The major product 6 was
identified as 1-(1,2-dihydro-4-propionyl-1-tosylpyridin-2-yl)butan-
2-one. Minor amounts of the bicylic intermediate 1-(2-ethyl-
3a,4,7,7a-tetrahydro-4-tosylfuro[3,2-b]pyridin-7-yl)propan-1-one (5)
could be identified on the basis of a set of resonances similar to
those of compound 3.
1
Both samples were monitored by H NMR as shown in Figure 4.
Acknowledgment. Financial support of this work by MEC
(CTQ2008-00042/BQU and Consolider Ingenio 2010, Grant No.
CSD2006-0003) is acknowledged. We thank Prof A. Echavarren
(ICIQ) for helpful comments and suggestions.
1-(1,2-Dihydro-4-propionyl-1-tosylpyridin-2-yl)butan-2-one (6).
1H NMR (CDCl3, 400 MHz), 7.67 (d, 2H, J ) 8 Hz), 7.28 (d, 2H,
J ) 8 Hz), 6.72 (d, 1H, J ) 8 Hz), 5.98 (d, 1H, J ) 8 Hz), 6.36,
5.10, 3.09, 2.82 (ABMX spin system, δA ) 2.82, δB ) 3.09, δM
)
5.10, δX ) 6.36, JAB ) 18 Hz, JAM) 4 Hz, JBM ) 8 Hz, JMX ) 6
Hz), 2.38 (s, 3H, CH3), 2.40 (q, 2H, COCH2CH3, J ) 7.8 Hz),
2.35 (q, 2H, COCH2CH3), 1.02 (t, 3H, COCH2CH3, J ) 7.8 Hz),
0.98 (t, 3H, COCH2CH3, J ) 7.8 Hz). 13C{1H} NMR (CDCl3, 100
MHz): 208.5, 198.5 (carbonylic CO), 144.5, 136.5, 132.7 (quater-
Supporting Information Available: Crystallographic data for
compound 2 (CIF). This material is available free of charge
via the Internet at http://pubs.acs.org.
JA1006614
9
J. AM. CHEM. SOC. VOL. 132, NO. 13, 2010 4607