Nucleophilic β-oniovinylation: Concept, mechanism, scope, and applications

Article abstract of DOI:10.1021/ja071316a

Insertion of an electron-deficient alkyne A-C≡C-A (A = CO ₂Me) into the C-L⁺ bond of an acyl-onio salt R-C(O)-L ⁺ (R = Ar, OAlk; L = 4-dimethylaminopyridine, PPh₃) has for the first time been achieved in the presence of catalytic amounts of the nucleophile L. For R = OMe, a second insertion of the alkyne was observed. X-ray structures were obtained for a number of such β-oniovinylation products. Depending on reaction conditions, preferentially E- or Z-stereochemistry was observed, the Z-isomer being the thermodynamically more stable. A mechanism for this novel insertion reaction is presented which accounts for the topology of the products and rationalizes the observed stereochemistry. The β-onio-activated Michael systems thus generated represent a virtually unexplored class of compounds. The onio substituent in such compounds can be selectively replaced by a number of nucleophiles. Thus a series of Michael systems with donor functions in the β-position is easily synthesized. These compounds represent a source for useful further transformations, for example, cyclizations to quinolones, thiochromones, and pyrazoles.

Full text of DOI:10.1021/ja071316a

Published on Web 03/15/2008
Nucleophilic â-Oniovinylation: Concept, Mechanism, Scope,
and Applications
Robert Weiss,*,† Matthias Bess,† Stefan M. Huber,*,† and Frank W. Heinemann‡
Department of Chemistry and Pharmacy, Organic Chemistry, Friedrich-Alexander-UniVersita¨t
Erlangen-Nu¨rnberg, Henkestrasse 42, D-91054 Erlangen, Germany, and Department of
Chemistry and Pharmacy, Inorganic Chemistry, Friedrich-Alexander-UniVersita¨t
Erlangen-Nu¨rnberg, Egerlandstrasse 1, D-91058 Erlangen, Germany
Received February 24, 2007; E-mail: weiss@chemie.uni-erlangen.de; huber205@umn.edu
Abstract: Insertion of an electron-deficient alkyne A-CtC-A (A ) CO2Me) into the C-L+ bond of an
acyl-onio salt R-C(O)-L+ (R ) Ar, OAlk; L ) 4-dimethylaminopyridine, PPh3) has for the first time been
achieved in the presence of catalytic amounts of the nucleophile L. For R ) OMe, a second insertion of
the alkyne was observed. X-ray structures were obtained for a number of such â-oniovinylation products.
Depending on reaction conditions, preferentially E- or Z-stereochemistry was observed, the Z-isomer being
the thermodynamically more stable. A mechanism for this novel insertion reaction is presented which
accounts for the topology of the products and rationalizes the observed stereochemistry. The â-onio-activated
Michael systems thus generated represent a virtually unexplored class of compounds. The onio substituent
in such compounds can be selectively replaced by a number of nucleophiles. Thus a series of Michael
systems with donor functions in the â-position is easily synthesized. These compounds represent a source
for useful further transformations, for example, cyclizations to quinolones, thiochromones, and pyrazoles.
Chart 1. Structural Examples for Systems 1 and 2, Which Can Be
Activated with Regard to Their d2-Reactivity by Nucleophiles L (3,
Introduction
R,â-Unsaturated carbonyl compounds 1 and 2 count among
the most important substrates in organic synthesis. Their d0-,
a1-, and a3-reactivities1 belong to the classical repertoire of
organic synthesis. However, d2-reactivity at the R-position (cf.
1, 2; Chart 1) of these carbonyl compounds is completely
underdeveloped and usually remains dormant. In their pioneering
work, Morita et al.,2 Baylis and Hillman,3 and Rauhut and
Carrier4 as well as many subsequent workers5 have demonstrated
that d2-reactivity can be brought to the fore in these (and
electronically related) compounds by addition of catalytic
amounts of nucleophiles L (normally tertiary amines and
phosphines). E1cb-type intermediates 3 and 4 act as d2-reagents
in subsequent reactions with electrophiles. The vast majority
of such reactions have been performed with systems 1;
comparatively few have been performed with systems 2.
4)
been performed with d2-reagents of type 3 and 4. The most
obvious problem with this latter reaction category is the danger
of rapid irreversible consumption of L by substitution reactions
involving L and the substitutable electrophilic center in question,
before d2-nucleophiles such as 3 and 4 are formed to any
reasonable amount in equilibrium concentrations. Thus a
conventional attempt to acylate intermediates 3 and 4 by using
acid halides as electrophiles is problematic, since the latter are
much stronger electrophiles than Michael systems and will
successfully compete for L under formation of acyl-onio salts.
We envisaged that this difficulty might be overcome by reacting
stoichiometric amounts of presynthesized acyl-onio salts7 with
Without exception, all products thus obtained derive from
addition reactions6 of intermediates of type 3 or 4. R-Hydroxy-
alkylation products are thus obtained with aldehydes and ketones
as standard electrophiles. To the best of our knowledge, no
substitution reactions on appropriate electrophilic centers have
† Department of Chemistry and Pharmacy, Organic Chemistry.
‡ Department of Chemistry and Pharmacy, Inorganic Chemistry.
(1) Seebach, D. Angew. Chem. 1979, 91 (4), 259-278.
(2) Morita, K.; Suzuki, Z.; Hirose, H. Bull. Chem. Soc. Jpn. 1968, 41, 2815.
(3) Baylis, A. B.; Hillman, M. E. D. German Patent 2155113, 1972; Chem.
Abstr. 1972, 77, 34174q.
(6) (a) Winterfeldt, E. Chem. Ber. 1965, 98, 1581-1587. (b) Winterfeldt, E.
Angew. Chem., Int. Ed. Engl. 1967, 6, 423-434. (c) Nozaki, K.; Sato, N.;
Ikeda, K.; Takaya, H. J. Org. Chem. 1996, 61, 4516-4519. (d) Johnson,
A. W.; Tebby, J. C. J. Chem. Soc. 1961, 2126-2130. (e) Nair, V.; Rajesh,
C.; Vinod, A. U.; Bindu, S.; Sreekanth, A. R.; Mathen, J. S.; Balagopal, L.
Acc. Chem. Res. 2003, 36, 899-907. (f) Nair, V.; Menon, R. S.; Sreekanth,
A. R.; Abhilash, N.; Biju, A. T. Acc. Chem. Res. 2006, 39, 520-530.
(7) Steglich, W.; Hoefle, G. Angew. Chem., Int. Ed. Engl. 1969, 8 (12), 981.
(4) Rauhut, M.; Currier, H. U.S. Patent 3074999, 1963; Chem. Abstr. 1963,
58, 11224a.
(5) Basavaiah, D.; Rao, A. J.; Satyanarayana, T. Chem. ReV. 2003, 103, 811-
891.
9
4610
J. AM. CHEM. SOC. 2008, 130, 4610-4617
10.1021/ja071316a CCC: $40.75 © 2008 American Chemical Society

Nucleophilic
â
-Oniovinylation
A R T I C L E S
Scheme 1. General Reaction Scheme for Catalytic Insertion of an
Electron-Deficient Alkyne 2 into an Onio-Activated Acyl Compound
5 by Employing a3- and d2-Reactivity of a Michael System
Scheme 3. General Reaction Equation for the Formation of
Vinylonio Compoundsa
a A ) CO2Me; OTf- (triflate) ) CF3SO3-; L ) PPh3, 4-dimethylami-
nopyridine (DMAP).
Scheme 2. Formation of 9
Table 1. Different Combinations of Reagents in the NOV Reaction
Shown in Scheme 3
variations
L
catalyst
type A
type B
type C
PPh3
DMAP
DMAP
PPh3
PPh3
DMAP
Scheme 4. Formation of 10Za
Michael systems in the presence of catalytic amounts of L
according to Scheme 1.
This reaction mode would maintain a stationary concentration
of the nucleophilic catalyst L, as its attack on the strong
electrophile 5 would regenerate L in a self-exchange reaction.
Consequently, a unique situation arises in that L serves the
double purpose of simultaneously increasing the a1-reactivity
of the acylating agent7 and the d2-reactivity of the Michael
system. Furthermore, this type of reaction would be both
stoichiometric and catalytic in L.
a L ) DMAP, A ) CO2Me. (i) CH2Cl2, PPh3 (cat.), room temperature,
16 h, 84%.
The reaction shown in Scheme 2 is considered proof of
principle for a novel synthetically useful type of nucleophilic
substitution that amounts to an insertion of an electron-deficient
alkyne into the C-L bond of a highly electrophilic R-C(O)-
L+ system under the mediating influence of nucleophile L. To
the best of our knowledge, this reaction type is without
precedence in the literature. We have termed it nucleophilic (â)-
oniovinylation (NOV).
As there are obviously no stereoisomers in the example
presented above, it is not possible at this point to discuss
stereochemical aspects (syn vs anti addition to the triple bond)
of this novel reaction. These and other mechanistic aspects of
this reaction are addressed below.
The NOV protocol was successfully applied to a number of
other onio-activated electrophiles. A general reaction equation
is shown in Scheme 3. Three types of general reaction
procedures can be distinguished, depending on substrate and
catalyst (Table 1). One example for type B is formation of the
â-oniovinylation product 10Z from benzoyl(4-dimethylami-
nopyridinio) triflate (Scheme 4; Figure S2, Supporting Informa-
tion). This reaction is useful for a first view on the addition
pattern (to the acetylene) being applied during the NOV reaction,
as in 10Z the substituents at the initial d2-position of the
acetylene can be distinguished.
As can be seen from the structural plot, the stereochemistry
of 10Z indicates anti addition to the acetylene. NMR data prove
that the reaction product consists of only one isomer. In this
case stereochemistry is clearly specified as the result of an anti
addition to DMAD. The two ester groups are almost coplanar
with the ethylene unit (twisted only by 2° and 14°), but bond
lengths for the DMAD backbone are almost the same as in 9.
The torsional angle between the π-systems of ethylene and
benzoyl is 86°. The exocyclic C-NMe2 bond indicates a high
degree of double-bond character (1.333 Å), at the expense of
the aromaticity of the pyridinium substituent.
Formation of 6 amounts to an insertion of an electron-
deficient alkyne into the C-L+ bond of an electron-deficient
acyl-onio component. To the best of our knowledge, this type
of reaction has so far never been reported. The presumed product
6 is of considerable interest in itself since it represents a
vinylogous acyl-onio compound, a virtually unexplored type of
electrostatically activated Michael acceptors.
We have addressed this problem by reaction of structurally
defined acyl-onio salts as substitutable electrophiles with
representatives of d2-nucleophiles of type 4 (generated in situ
from the corresponding electron-deficient alkyne 2) in prototypi-
cal model reactions. Subsequently we report on results and
insights obtained in the course of these investigations.
Results and Discussion
As a first model investigation for the insertion reaction
considered in Scheme 1, we synthesized the (methoxycarbonyl)-
triphenylphosphonium triflate 7 according to the SASAPOS
protocol8 and reacted it with dimethyl acetylenedicarboxylate
8 (DMAD) in the presence of a catalytic amount (2%) of PPh3.
Under mild conditions (CH2Cl2, room temperature, <1 h) the
desired insertion reaction could indeed be observed, the â-onio-
substituted Michael system 9 being the sole reaction product
(Scheme 2). In the absence of PPh3, the product did not form.
Even though the cation of compound 9 has been made before
by Shaw et al.,9 this new synthetic method is a progressive path,
as no delicate methylation process involving highly toxic
reactants is needed anymore. The reaction product 9 was fully
characterized (cf. Experimental Section), including an X-ray
structural analysis (Figure S1, Supporting Information).
(8) Weiss, R.; Pu¨hlhofer, F. Angew. Chem., Int. Ed. 2002, 41 (20), 3815-
3817.
A number of further substrates could be shown to react in
the same way (cf. Table 2). Some of the products mentioned
(9) Shaw, M. A.; Tebby, J. C.; Ronayne, J.; Williams, D. H. J. Chem. Soc. C
1967, 10, 944-947.
9
J. AM. CHEM. SOC. VOL. 130, NO. 14, 2008 4611

A R T I C L E S
Weiss et al.
Table 2. Summary of Successful NOV Reactions (see Scheme 3)
are useful for a closer look at the reaction characteristics of
NOV.
flate could be isolated as a pure E (15E) or Z compound (15Z)
by simple variation of the catalyst (PPh3 yielding E vs DMAP
yielding Z; Scheme 5). Compounds 15E and 15Z have been
fully characterized, including X-ray analysis.
Experimental results suggest that short reaction times and
PPh3-catalysis are the key for the formation of anti addition
products. By application of this method to benzoyl(4-dimethy-
laminopyridinio) triflate, the E-compound 10E could be obtained
(Scheme 6; Figure S3, Supporting Information) and fully
characterized. However, it was mixed with the corresponding
Z-compound 10Z.
For some products of the â-oniovinylation, 13C NMR and
X-ray analysis proved the exclusive formation of one of the
two configurational isomers. To a certain extent these results
classify the â-oniovinylation as a stereospecific anti addition
with respect to DMAD 8. However, there are results contradict-
ing this simple proposition: In some cases mixtures of syn (E)
and anti (Z) addition product and even pure compounds of the
syn addition pattern were obtained. N-[2-(2-thienylcarbonyl)-
1,2-di(methoxycarbonyl)ethenyl]-4-dimethylaminopyridinio tri-
9
4612 J. AM. CHEM. SOC. VOL. 130, NO. 14, 2008

Nucleophilic
â
-Oniovinylation
A R T I C L E S
Scheme 5. Selective Formation of Two Stereochemically
Scheme 8. Catalytic Cycle for Formation of 10Z
Alternative Vinylonio Products from the Same Substratea
a Reactions were carried out in CH3CN at room temperature. (i) PPh3
(2.3 mol %), 2 h, 21%; (ii) DMAP (8.8 mol %), 3 days, 68%. A ) CO2Me.
There is some evidence that the initial product from the
catalytic cycle in Scheme 8 is the syn addition product 10E.
This could be due to a SN2-type phosphine/pyridine exchange
reaction (instead of addition-elimination). Studies by Rap-
poport10 suggest that such substitutions at a Michael position
are very fast and thus SN2-like. For that reason, inversion at
the reaction center has to be expected.
In self-exchange experiments it could be shown that the anti
addition product is the thermodynamic product. It is very likely
that the occurrence of the anti addition product is due to some
free DMAP in the reaction mixture because of impurity of the
precursor. This is how both appearance and disappearance of
E-products like 15E and 10E can be explained.
Scheme 6. Formation of 10Ea
a L ) DMAP, A ) CO2Me. (i) CH2Cl2, PPh3 (cat.), room temperature,
10 min, 84%. The short reaction time allows us to obtain the major part as
the E-product 10E (E/Z ) 3:1). Overall yield is 79%.
Scheme 7. Proposed Catalytic Cycle for Formation of 9
Bifunctional and Repetitive NOV. The described NOV
reaction pattern has also been applied both to bifunctional aroyl-
onio compounds and to products of NOV (cf. Table 2). Some
compounds could be obtained that showed two vinylonio
functions or proved repetitive NOV reaction. One of them
is Z,Z-1,4-phenylenebis[[2-(4-dimethylaminopyridinio)-1,2-di-
(methoxycarbonyl)ethenyl]carbonyl] bistriflate 16. It can be
obtained from 1,4-phenylenebis[(4-dimethylaminopyridinio)-
carbonyl] bistriflate 22, applying reaction conditions of type C
(Scheme 9). The bis(vinylonio) product 16 was fully character-
ized, including X-ray analysis.
Compound 23 is unique, as it is the only electrophile for NOV
with a non-acylic reaction center (cf. Scheme 1 vs Scheme 3).
Related research by Nair et al.11 shows that carbonyl groups,
not Michael positions, are the appropriate reaction partners for
composite nucleophiles like 20. The reactivity of 23 is probably
due to the deactivation of the carbonyl function in the ester
groups combined with the high electron deficit of the ethene.
Compound 23 is thus the only compound so far to which the
vinylonio reaction could be applied one more time, forming the
product 18 (Scheme 10). This reaction can be considered as
similar to the nucleophilic (e.g., anionic) induced polymerization
of acetylenes. However, 18 (Figure S4, Supporting Information)
does not react with DMAD anymore. No product with three
(or more) DMAD units can be found. This is quite remarkable,
as intuition suggests that further insertions should be even easier
the longer the chain gets: the conjugated π-system should attract
reactions because of its reduced lowest unoccupied molecular
Mechanistic Considerations. First we refer to the most
simple NOV reaction (type A) shown in Scheme 2. For this
reaction we propose the reaction mechanism given in Scheme
7. The initial step is the attack of the primary nucleophile PPh3
on the acetylene 8, generating 20 and activating d2-reactivity
of the acetylene. The intermediate 20 reacts fast, attacking the
electrophilic carbonyl center of the acyl-phosphonio precursor
7. In a proposed addition-elimination process, the product 9 is
formed and the catalyst PPh3 is liberated. (The liberated PPh3
molecule is not identical to the one that initiated the reaction!)
This reaction is significantly faster than the possible side reaction
of 20 with another PPh3 molecule forming a stable 2:1
compound of PPh3 and DMAD previously described by Shaw
et al.9
A different proposition has to be made for the reaction
sequence and especially the role of the catalyst PPh3 in the case
of DMAP as onio ligand (type B). The reaction sequence is
similar to type A, up to the formation of a vinylphosphonio
intermediate 21 and liberation of the initial onio ligand DMAP.
At least one more reaction step has to be explained as the
primary nucleophile (catalyst PPh3) cannot keep its position after
the NOV reaction and is replaced by the liberated pyridine: The
C-N bond is lower in energy compared to the C-P bond, thus
the substitution fits expectations. Consequently, the final product
is 10Z instead of 21. The catalyst is regenerated for further
catalytic cycles (Scheme 8).
(10) (a) Rappoport, Z. Acc. Chem. Res. 1981, 14, 7-15. (b) Rappoport, Z. Acc.
Chem. Res. 1992, 25, 474-479.
(11) (a) Nair, V.; Vinod, A. U.; Nair, J. S.; Sreekanth, A. R.; Rath, N. P.
Tetrahedron Lett. 2000, 41, 6675-6679. (b) Nair, V.; Mathen, J. S.; Viji,
S.; Srinivas, R.; Nandakumar, M. V.; Varma, L. Tetrahedron 2002, 58,
8113-8118.
9
J. AM. CHEM. SOC. VOL. 130, NO. 14, 2008 4613

A R T I C L E S
Weiss et al.
Scheme 9. Repetitive (Bifunctional) â-Oniovinylationa
a A ) CO2Me, L ) DMAP. (i) CH3CN, DMAP (cat.), room temperature, 4 days, 77%.
Scheme 10. Repetitive â-Oniovinylation by Insertion of DMAD into an Already Existing Vinylonio Function
Scheme 11. Characteristic Nucleophilic Substitution Reactions in Some Vinylonio Compoundsa
a A ) CO2Me. (i) DMAP, 16 h; (ii) N-methylimidazole, 30 min; (iii) R ) 4-methoxyphenyl or thiophenol, 1 h, -DMAPH+; (iv) R ) 2-fluorophenyl or
aniline, 12 h, -DMAPH+; (v) R ) 2-thienyl or phenol, DBU, 3 days, - DMAPH+; (vi) R ) phenyl or p-phenylenediamine, 12 h, -DMAPH+. All
-
reactions were carried out at room temperature. Anion: CF3SO3
.
orbital (LUMO). But this assumption is invalid in case of 18,
as π-conjugation is very poor (cf. Figure S4). Additionally, the
steric situation of the desired reaction center does not offer a
trajectory for the bulky composite nucleophile.
in Scheme 11). The vinylonio systems are transformed into
neutral push-pull derivatives. As a byproduct, DMAPH+OTf-
is formed. These reactions yield a mixture of E and Z products.
This can be shown by doubled signal sets in all the NMR
spectra.
Applications/Simple Derivatizations. Due to the â-vinylonio
compounds being highly functionalized, a variety of subsequent
reactions can be envisaged, bringing new derivatives of Michael
systems into reach. A first impression of compounds accessible
by nucleophilic substitution of the onio substituent at the
Michael position is given by the examples in Scheme 11. Two
types of such derivatizations can be distinguished: (a) onio
exchange and (b) substitution with protic nucleophiles.
(a) Onio exchange (i and ii in Scheme 11): The phosphonio
substituent in the vinylphosphonio compound 9 can be ex-
changed against other nucleophiles like DMAP and N-meth-
ylimidazole. This has already been implied to explain the
catalytic function of PPh3 in the formation of vinylpyridinio
compounds according to Scheme 8. Thus compounds 23 and
24 are accessible. They have been fully characterized, including
X-ray analysis.
More Complex Derivatizations. In some cases the substitu-
tion of L+ described above is followed by a intramolecular
reaction. We thus could obtain N- and S-heterocyclic compounds
(pyrazoles, quinolones, and a thiochromone).
New Pyrazole Synthesis. 1-Phenylpyrazoles can be obtained
from vinylonio compounds 10Z and 16 by treatment with
phenylhydrazine. Compound 10Z gave a dimethyl diphenyl-
pyrazoledicarboxylate 30 (Scheme 12). The hydrazine substitutes
the onio ligand L followed by deprotonation and an intramo-
lecular condensation reaction of the intermediate 29, giving
dimethyl 1,5-diphenylpyrazole-3,4-dicarboxylate 30. NMR spec-
tra and X-ray analysis show that only one regioisomer is formed
(Figure S5, Supporting Information). This gives evidence that
the nucleophilic substitution of the onio ligand is significantly
faster than a conceivable formation of the corresponding
hydrazone. This mechanistic pathway is discredited because it
(b) Reaction of vinyl-DMAP-onio compounds with nucleo-
philes of type H-Nu (O-, S-, and N-based nucleophiles; iii-vi
9
4614 J. AM. CHEM. SOC. VOL. 130, NO. 14, 2008

Nucleophilic
â
-Oniovinylation
A R T I C L E S
Scheme 12. Reaction of 10Z with Phenylhydrazine
Scheme 13. Alternative Reaction Pathway of 10Z with Phenylhydrazine, Showing That the Isolated Product 30 Excludes This Mechanism
Because of Its Regiochemistry
Scheme 14. Reaction of 16 with Phenylhydrazine
Scheme 15. Quinolone Synthesis
would give dimethyl 1,3-diphenylpyrazole-4,5-dicarboxylate
rather than the actual product, 30 (Scheme 13).
The new protocol for pyrazole synthesis in Scheme 12 can
also be applied multiple times within the same molecule
(Scheme 14). After column chromatography (SiO2), only one
regioisomer (31) was detected. The position of the phenyl groups
was determined by 1H nuclear Overhauser effect (NOE) NMR
(500 MHz). Together with the crystallographic data provided
above, it proves the regioselectivity of this new pyrazole
synthesis. Due to the high reactivity of the vinylonio intermedi-
ates, pyrazole formation can be achieved under mild conditions.
Thus this new synthetic approach is likely to be useful for the
synthesis of substituted pyrazoles with delicate functional
groups.
New Quinolone and Thiochromone Synthesis. Further
reactions on NOV products were carried out with aniline or its
derivatives as nucleophiles. For instance, starting from the
2-fluorobenzoylvinylonio system 11/OTf, the corresponding
neutral product 26 could be obtained as an E/Z mixture in a
including an X-ray analysis (cf. Figure S6, Supporting Informa-
tion).
2:1 ratio (yield 82%, white solid). Refluxing of 26 in acetonitrile
for 2 days or its treatment with base (n-butyllithium) led to the
Using this procedure (albeit partly via isolating some of the
intermediate products), we also synthesized products 33-35 (cf.
Chart 2) by variation of the starting electrophile (33, overall
yield 33%) or the substituting nucleophile (34, yield 60%; 35,
yield 25%). Thiochromone 35 was also characterized by X-ray
analysis.
formation of quinolone 32 in high yields (reflux 92%, base 69%)
via nucleophilic aromatic substitution of the fluoride by the
amine group (cf. Scheme 15).
On the basis of these findings, we were able to derive a novel
one-pot quinolone synthesis. Starting from 2-fluorobenzoyl
chloride, the intermediate vinylonio species 11/Cl was generated
in situ by addition of (i) DMAP and (ii, after 1 h) DMAD with
a catalytic amount of DMAP. Further addition of aniline (or
one of its derivatives) and treatment with n-butyllithium gave
the quinolone product 32 (Z ) CO2Me) (yield 72%; isolation
via column chromatography). It was completely characterized,
When methyl propiolate was used as the acetylene component
(Z ) H), only moderate to low yields were obtained (one-pot
quinolone synthesis, 35%), compared to DMAD as electrophilic
substituent. This is probably due to the deprotonation of the
acidic acetylene proton by the intermediately formed composite
9
J. AM. CHEM. SOC. VOL. 130, NO. 14, 2008 4615

A R T I C L E S
Weiss et al.
Chart 2. Quinolones 33 and 34 and Thiochromone 35
though). With respect to aromatic carboxylic acid derivatives,
nucleophilic heteroatoms (like N) within the aromatic could lead
to unwanted intramolecular reactions. In addition, strongly
electron-donating groups or strongly electron-withdrawing sub-
stituents at the aromatic should be avoided, as they might lead
to Friedel-Crafts-like reactions with DMAD or might create
further centers for nucleophilic attack (respectively). An NOV-
like reactivity of alkenes 1 instead of DMAD could not be
provoked.
Further investigations into NOV will very likely reveal more
applications of this method and will equally likely develop a
more precise idea of its limitations.
nucleophile. Unfortunately, methyl 3-(trimethylsilyl)propiolate
(Z ) TMS) did not undergo any vinylonio reaction.
Yet, as the decarboxylation of quinolones of type 32 at C2
is known from the literature to be simple and highly effective,12
the above-mentioned method provides a novel synthetic ap-
proach to modern fluoroquinolone antibiotics (which bear a
hydrogen substituent at C2 as well as a carboxy group at C3).
These constitute the most important class of antiobiotics used
today, curing a wide variety of diseases.13 In comparison to the
currently used syntheses for fluoroquinolones, namely, the
Gould-Jacobs14 and especially the Grohe15 method, the syn-
thetic route introduced above features some distinctive advan-
tages: in addition to the mild reaction conditions, the one-pot
approach, and the favorable “atom economy”, it should be
possible to introduce very weakly nucleophilic amines at 11/
Cl, due to the strong electrostatic activation by the onio ligand
(which is also a much better leaving group compared to alkoxy
or dimethylamino substituents employed by the established
Grohe method15). Hence, in addition to the advantages, our
method has the potential to increase the range of accessible
fluoroquinolones (especially those with weakly nucleophilic
amines introduced at N1 and those bearing a substituent at C2
that can be derived from a carboxymethyl group), thus enabling
the development and testing of prospective structures for novel
antibiotics.
Experimental Section
Unless otherwise stated, all manipulations were performed under
an inert atmosphere of nitrogen via standard Schlenk techniques. Dry
solvents were employed throughout.
1,2,2-Tri(methoxycarbonyl)ethenyltriphenylphosphonio triflate
(9). The employed compound 7 was prepared in situ by addition of
1.404 g (5.35 mmol) triphenylphosphine to a solution of 0.39 mL (5.07
mmol) of methyl chloroformate and 0.92 mL (5.09 mmol) (trimethyl-
silyl)trifluoromethanesulfonate in CH2Cl2 at room temperature. On
addition of 0.62 mL (5.07 mmol) of dimethyl acetylenedicarboxylate,
the color of the solution turns to orange and considerable heat develops.
The product 9 precipitates on addition of Et2O. It is filtered and washed
with a mixture of CH2Cl2 and Et2O 1:1. Yield: 2.330 g (75%) colorless
powder. M (C27H24F3O9PS) ) 612.52. FAB-MS: 463 (9 - OTf-). IR
(KBr): 2959 (w), 1742 (s), 1439 (m), 1265 (br, vs), 1170 (w), 1104
(m), 1030 (m), 1011 (w), 900 (w), 863 (w), 755 (w), 723 (w), 691 (w),
637 (m), 572 (w), 517 (m). 1H NMR (CDCl3): 7.72 (m, 15H), 3.85 (s,
3H, OMe), 3.37 (s, 3H, OMe), 3.08 (s, 3H, OMe). 13C NMR (CDCl3):
161.77 (d, CO), 160.85 (d, CO), 160.28 (d, CO), 144.20, 136.12, 134.09,
1
131.79, 130.44, 120.72 [q, J(C-F) ) -321 Hz, CF3SO3-], 114.84,
54.49 (s, OMe), 54.11 (s, OMe), 53.78 (s, OMe). Anal. C27H24F3O9PS:
calcd C 52.95, H 3.95, S 5.23; found C 52.60, H 4.36, S 5.19.
Dimethyl 1,5-Diphenylpyrazole-3,4-dicarboxylate (30). A solution
of 639 mg (1.23 mmol) of (Z)-N-[2-benzoyl-1,2-di(methoxycarbonyl)-
ethenyl]-4-dimethylaminopyridinio triflate 10Z in 15 mL of CH2Cl2 is
treated with 0.12 mL (1.21 mmol) of phenylhydrazine at room
temperature. After 15 h the solvent was removed and the product was
isolated by column chromatography (mobile phase, hexane/ethyl acetate
1:1). Yield: 204 mg (49%) yellow crystals. M (C19H16N2O4) ) 336.34.
FAB-MS: 337 (30 + H+), 305 (30 - MeO-). IR (KBr): 1723 (m),
1500 (w), 1401 (br, vs), 1330 (w), 1317 (w), 1261 (w), 1225 (m), 1079
(m), 969 (w), 792 (w), 762 (w), 696 (w). 1H NMR (CDCl3): 7.34 (m,
br, 10H, Ar), 3.93 (s, 3H, OMe), 3.72 (s, 3H, OMe). 13C NMR
(CDCl3): 163.32, 162.14, 144.68, 143.00, 138.41, 129.87, 129.37,
128.82, 128.47, 128.22, 127.61, 125.45, 115.38, 52.48, 52.10. Anal.
C19H16N2O4: calcd C 67.85, H 4.79, N 8.33; found C 67.51, H 4.78,
N 8.17.
Dimethyl 4-Oxo-1-phenyl-1,4-dihydroquinoline-2,3-dicarboxylate
(32; R1 ) R2 ) H, Z ) CO2Me). (a) From 26 (R1 ) R2 ) H, Z )
CO2Me), by use of base (n-BuLi): A solution of 533 mg (1.49 mmol)
of 26 in 20 mL of CH2Cl2 was cooled to -78 °C and treated with 1.05
mL of a 1.6 M solution of n-butyllithium in hexanes (corresponding to
1.68 mmol of butyllithium). After being stirred for 60 min, the solution
was warmed to room temperature and stirred for another 15 min. The
solution was treated with 40 mL of H2O, and the layers were separated.
The organic layer was washed another three times with 40 mL of H2O
and dried over MgSO4. After removal of the solvent, the product was
isolated by column chromatography (mobile phase hexane/ethyl acetate
2:1, slowly more polar). Yield: 348 mg (69%), white powder. M
(C19H15NO5) ) 337.33. FAB-MS: 338 (32 + H+). IR (KBr): 3052
(w), 2954 (w), 1744 (s), 1714 (s), 1631 (s), 1604 (s), 1536 (m), 1492
(m), 1468 (s), 1436 (m), 1330 (m), 1297 (s), 1275 (m), 1225 (s), 1174
Conclusion
Nucleophilic â-oniovinylation (NOV), as presented in this
paper, constitutes a novel synthetic method for the preparation
of (formerly almost unknown) â-onio-activated Michael systems.
The products are obtained under mild conditions from onio
derivatives of carboxylic acids, DMAD, and catalytic amounts
of either DMAP or PPh3, making use of the alkyne’s implicit
d2-reactivity. As the products are highly functionalized, they
are themselves the starting points for further subsequent
reactions and syntheses. A few first examples were presented
in this paper, namely, the synthesis of pyrazoles and quinolones,
and a wide array of heterocycles may be accessible via this route.
During our investigations we also came upon some limitations
of NOV. With regard to aliphatic carboxylic acid derivatives
as substrates, acetic acid chloride (as well as trichloroacetic acid
chloride) proved to be incompatible with NOV, as a proton (or
chloro substituent) at the R carbon was attacked by the
composite nucleophile (methoxyacetic acid was compatible,
(12) Biere, H.; Seelen, W. Liebigs Ann. Chem. 1976, 1972-1981.
(13) (a) Crumplin, G. C., Ed. The 4-quinolones; Springer-Verlag, London, 1990.
(b) Shah, P. M. Pharm. i. u. Z. 2001, 5, 394-398. (c) Grohe, K. J. Prakt.
Chem. 1993, 335, 397-409.
(14) (a) Gould, R. G.; Jacobs, W. A. J. Am. Chem. Soc. 1939, 61, 2890-2895.
(b) Koga, H.; Murayama, S.; Suzue, S.; Irikura, T. J. Med. Chem. 1980,
23, 1358-1363.
(15) (a) Grohe, K.; Bayer, A. G. Chem. Abstr. 1980, 92, 41916w. (b) Grohe, K.
J. Prakt. Chem. 1993, 335, 397-409. (c) Grohe, K.; Bayer, A. G. Chem.
Abstr. 1986, 105, 226051r.
9
4616 J. AM. CHEM. SOC. VOL. 130, NO. 14, 2008

Nucleophilic
â
-Oniovinylation
A R T I C L E S
(w), 1158 (w), 1136 (m), 1076 (w), 1042 (m), 972 (m), 952 (w), 935
(w), 910 (w), 876 (w), 841 (w), 808 (w), 767 (s), 742 (m), 726 (w),
704 (m), 644 (w), 600 (m), 503 (w). 1H NMR (CDCl3): 8.50 [dd, 3J(H-
the text, sorted by the corresponding compound (as mentioned in the
text).The supplementary crystallographic data can be obtained free of
4
H) ) 8.1 Hz, J(H-H) ) 1.6 Hz, 1H, C5-H), 7.59 (m, 3H, C3′-H/
Table 3.
3
3
4
C4′-H), 7.50 [ddd, J(H-H) ) 7.1 Hz, J(H-H) ) 7.1 Hz, J(H-H)
compd
CCDC number
3
) 1.6 Hz, 1H, C7-H], 7.40 (m, 3 H, C6-H/C2′-H), 6.76 [d, J(H-H)
9
637381
637387
637378
637390
637377
637385
637379
637386
637384
637380
637383
637394
637382
637388
637393
637389
637391
) 8.1 Hz, 1H, C8-H], 3.92 (s, 3H, CO2Me), 3.52 (s, 3H, CO2Me). 13
C
10Z
10E
11
12
13E
14
15Z
15E
16
18
23
NMR (CDCl3): 174.3 (C4), 166.0 (CO2Me), 162.6 (CO2Me), 149.4
(C2), 141.0 (C8a), 137.2 (C1′), 132.9 (C7), 130.5 (C4′), 129.9 (C3′),
129.4 (C2′), 127.4 (C4a), 127.1 (C5), 125.5 (C6), 118.1 (C8), 110.7
(C3), 52.9 (CO2Me), 52.6 (CO2Me).
(b) From 26 (R1 ) R2 ) H, Z ) CO2Me), by use of heat: A yellow
solution of 139 mg (0.39 mmol) of 26 in 15 mL of CH3CN was heated
to reflux for 2 days. After removal of the solvent, the residue was
chromatographed on SiO2 (mobile phase hexane/ethyl acetate 1:1; Rf
ca. 0.15). Yield: 122 mg (93%), white powder. Analytical data as
above.
(c) One-pot synthesis from 2-fluorobenzoyl chloride: A solution of
0.8 mL (6.70 mmol) of 2-fluorobenzoyl chloride in 50 mL of CH2Cl2
was treated with 845 mg (6.92 mmol) of DMAP, resulting in a white
suspension. After the reaction mixture was stirred for 3 h, a solution
of 0.84 mL (6.83 mmol) of DMAD in 20 mL of CH2Cl2 was slowly
added dropwise, yielding a clear, red solution. After this solution was
stirred for 21 h, 0.92 mL (10.1 mmol) of aniline was added dropwise.
The resulting orange solution was again stirred for 2 days and cooled
to -78 °C, and 5.5 mL of a 1.6 M solution of n-butyllithium in hexanes
was added (corresponding to 8.8 mmol of n-butyllithium). After being
stirred for 30 min, the solution was warmed to room temperature and
stirred for another 22 h. The precipitate was filtered off, and after
addition of 130 mL CH2Cl2 to the filtrate, it was treated with 80 mL of
water. After separation of the layers, the organic phase was washed
twice with 50 mL of water and dried over MgSO4, and the solvent
was evaporated. The resulting residue was chromatographed on SiO2
(mobile phase hexane/ethyl acetate 3:2, slowly more polar until pure
ethyl acetate was reached; Rf ca. 0.05). Yield: 1.61 g (71% referring
to 2-fluorobenzoyl chloride), white powder. Analytical data as above.
Table 3 contains the CCDC numbers for all structures mentioned in
24
25
30
32
35
charge from the Cambridge Crystallographic Data Centre via www.
ccdc.cam.ac.uk/data_request/cif.
Acknowledgment. S.M.H. thanks the Studienstiftung des
Deutschen Volkes and the State of Bavaria (BayBFG) for
1
scholarships. We also thank Prof. W. Bauer for the H NOE
NMR measurements and several fruitful discussions. This work
is dedicated to the memory of Werner Reinhart.
Supporting Information Available: Further experimental
procedures and characterization of the products, crystallographic
files in CIF format. This material is available free of charge
via the Internet at http://pubs.acs.org.
JA071316A
9
J. AM. CHEM. SOC. VOL. 130, NO. 14, 2008 4617