Variations on the Blaise Reaction: Synthesis of 3,5-Dioxopentanoates and 3-Amino-5-oxopent-3-enoates

Article abstract of DOI:10.1055/s-0035-1561662

We have achieved a facile and convenient synthesis of a variety of 3,5-dioxopentanoates (3,5-diketo esters) and 3-amino-5-oxopent-3-enoates by a zinc-mediated condensation of readily available 3-oxopropanenitriles (α-cyano ketones) with ethyl bromoacetate. The reaction is a variation on the classical Blaise reaction, tuned to the synthesis of trifunctional compounds having 3,5-diketo ester or 3-enamino 5-keto ester functional groups. Our studies revealed that the Blaise reaction on the nitrile occurs in preference to the Reformatsky reaction on the neighboring ketone when the two functional groups are in a geminal relationship, as found in α-cyano ketones, possibly due to zinc complexation leading to increased electrophilicity of the nitrile.

Full text of DOI:10.1055/s-0035-1561662

SYNLETT
0
9
3
6
-
5
2
1
4
1
4
3
7
-
2
0
9
6
© Georg Thieme Verlag Stuttgart · New York
2016, 27, A–E
letter
A
H. S. Rao, N. Muthanna
Letter
Synlett
Variations on the Blaise Reaction: Synthesis of 3,5-Dioxopentanoates
and 3-Amino-5-oxopent-3-enoates
H. Surya Prakash Rao*
Nandurka Muthanna
O
O
O
(ii) 3 N HCl
R
OEt
R: aryl, heteroaryl, t-Bu
13 examples; 55–81%
O
(i) BrCH2CO2Et
Zn, THF reflux
Department of Chemistry, Pondicherry University,
Pondicherry-605 014, India
hspr.che@pondiuni.edu.in
CN
O
NH2
O
R
(ii) aq K2CO3
(50% w/v)
R
OEt
R: aryl, heteroaryl, t-Bu
6 examples; 62–82%
exclusive formation of the Blaise reaction products
Received: 16.03.2016
Accepted after revision: 11.05.2016
Published online: 14.06.2016
thesis. For example, β-amino acids and numerous heterocy-
clic compounds of medicinal interest can be synthesized by
simple synthetic manipulations of the above intermediates.
Many variations on the Blaise reactants, for example with
nitriles and ethyl bromoacetate, can be envisaged for
achieving access to products with diverse functional
groups. We have been seeking to broaden the scope of the
Blaise reaction by varying the nitrile and bromo ester. For
example, we recently described a facile, high-yielding, and
scalable synthesis of a variety of 1,3-diketones 6 or 3-amino
enones 7 from nitriles 1 and 2-bromoethanones 5 (Scheme
1, eq. 2).3 We found that, simply by changing the tempera-
ture during hydrolysis of the Blaise intermediate, the reac-
tion provides either 1,3-diketones 6 (100 °C) or 3-amino
enones 7 (30 °C). To expand the scope of the Blaise reaction
further, we planned to employ 3-oxopropanenitriles (α-cy-
ano ketones) 8 instead of simple nitriles, so that the reac-
tion would provide 3,5-dioxopentanoates 9 or 3-amino-5-
oxopent-3-enoates 10 (Scheme 1, eq. 3). The α-cyano ke-
tones 8 possess a carbonyl group and a nitrile in a geminal
relationship; these functional groups should react with the
zinc enolate derived from ethyl bromoacetate by the Refor-
matsky pathway or the Blaise pathway, respectively. If the
reaction proceeds through the Blaise pathway, it would be
expected to provide 3,5-diketo esters (Scheme 1, eq. 3),
whereas if the reaction follows the Reformatsky pathway, it
would be expected to provide ethyl 4-cyano-3-hydroxybu-
tanoates 11 as the primary products and ethyl 4-cyanobut-
2-enoates 12 as the secondary products (Scheme 1, eq. 4).
When we conducted the reaction, we found that the reac-
tion followed the Blaise pathway to provide 3,5-dioxopen-
tanoates 9 as the exclusive products, even though the ke-
tone functional group is inherently more reactive than the
nitrile.
DOI: 10.1055/s-0035-1561662; Art ID: st-2016-d0182-l
Abstract We have achieved a facile and convenient synthesis of a vari-
ety of 3,5-dioxopentanoates (3,5-diketo esters) and 3-amino-5-oxo-
pent-3-enoates by a zinc-mediated condensation of readily available 3-
oxopropanenitriles (α-cyano ketones) with ethyl bromoacetate. The re-
action is a variation on the classical Blaise reaction, tuned to the synthe-
sis of trifunctional compounds having 3,5-diketo ester or 3-enamino 5-
keto ester functional groups. Our studies revealed that the Blaise reac-
tion on the nitrile occurs in preference to the Reformatsky reaction on
the neighboring ketone when the two functional groups are in a gemi-
nal relationship, as found in α-cyano ketones, possibly due to zinc com-
plexation leading to increased electrophilicity of the nitrile.
Key words Blaise reaction, diketo esters, aminooxopentenoates, cya-
no ketones, ethyl bromoacetate
The Blaise reaction is a classical C–C bond-forming reac-
tion, first described by E. E. Blaise in 1901.1 In this zinc-me-
diated reaction, an aliphatic or aromatic nitrile 1 reacts
with ethyl bromoacetate 2 to give the corresponding β-keto
ester 3 (Scheme 1, eq. 1). The Blaise reaction is closely relat-
ed to the more famous and more widely applied Refor-
matsky reaction,2 in which aldehydes or ketones react with
ethyl bromoacetate in presence of Zn to provide β-hydroxy
esters as the primary products and α,β-unsaturated esters
as the secondary products. Although it has not acquired as
much importance as the Reformatsky reaction, the Blaise
reaction is the more versatile of the two reactions because,
depending on the workup method, the reaction can provide
β-keto esters 3 (acidic workup) or β-amino α,β-unsaturated
esters 4 (basic workup; Scheme 1, eq. 1). Compounds bear-
ing β-keto ester or β-amino α,β-unsaturated ester function-
al groups have a wide range of applications in organic syn-
© Georg Thieme Verlag Stuttgart · New York — Synlett 2016, 27, A–E

B
H. S. Rao, N. Muthanna
Letter
Synlett
(2), and zinc in refluxing THF containing 3 mol% of TMSCl as
a zinc activator provided the 3,5-diketo ester 9a (R = Ph) as
the only product in moderate yield (37%) after acidic work-
up at room temperature with 3 N aqueous HCl (Scheme 2).
There was no trace of the Reformatsky products 11a or 12a
(R = Ph). Evidently, the condensation took place on the ni-
trile in preference to the neighboring ketone. To optimize
the yield of the desired 3,5-diketo ester 9a, we examined
several variations of the reagents and reaction conditions.
By conducting reactions at a 1 mmol scale of phenacyl cya-
nide (8a), we obtained a maximum yield of 9a (75%) by em-
ploying 2 equivalents of ethyl bromoacetate, 3 equivalents
of Zn, and 10 mol% of TMSCl in refluxing THF (Scheme 2).
The spectroscopic data for the 3,5-diketo ester 9a matched
the reported values.9
O
O
O
(ii) 3 N HCl
The Blaise reaction
R
OEt
OEt
R1
(i) BrCH2CO2Et
3
2
(1)
RCN
NH2
Zn, THF reflux
1
(ii) aq. K2CO3
(50% w/v)
R
4
O
O
Our previous work (ref. 3)
(i) BrCH2COR1
100 °C
R
6
5
(2)
RCN
NH2
O
Zn, THF reflux
1
R
O
R1
O
(ii) 3 N HCl
0 °C to rt
7
O
(ii) 3 N HCl
This work
R
OEt
OEt
(i) BrCH2CO2Et
O
9
2
(3)
CN
(i) BrCH2CO2Et (2 equiv)
O
O
O
O
O
NH2
O
R
Zn, THF reflux
2
(ii) aq. K2CO3
(50% w/v)
CN
8
R
R
OEt
R
Zn (3 equiv), TMSCl (10 mol%)
THF reflux, 3–5 h
8a–h
9a–h
10
(ii) 3 N HCl, 0 °C to rt, 55–81%
Anticipated Reformatsky reaction
(i) BrCH2CO2Et
8a: R = C6H5
9a: R = C6H5, 4 h (75%)
8b: R = 4-FC6H4
8c: R = 4-ClC6H4
9b: R = 4-FC6H4, 4 h (71%)
9c: R = 4-ClC6H4, 4 h (74%)
9d: R = 4-MeC6H4, 4 h (70%)
9e: R = 4-MeOC6H4, 4 h (72%)
9f: R = 1-naphthyl, 5 h (68%)
9g: R = 1-pyrenyl, 5 h (67%)
9h: R = 2-furyl, 3 h (78%)
9i: R = 2-thienyl, 3 h (77%)
9j: R = N-tosyl-3-pyrrolyl, 5 h (65%)
9k: R = N-tosyl-3-indolyl, 5 h (71%)
9l: R = ferrocenyl, 5 h (55%)
9m: R = t-Bu, 4 h (75%)
COOEt
OH
COOEt
CN
O
2
8d: R = 4-MeC6H4
8e: R = 4-MeOC6H4
8f: R = 1-naphthyl
8g: R = 1-pyrenyl
8h: R = 2-furyl
CN
(4)
CN
R
R
R
Zn, THF reflux
(ii) H3O+
8
12
11
Scheme 1 The Blaise and the Reformatsky reactions and their possible
variants
8i: R = 2-thienyl
8j: R = N-tosyl-3-pyrrolyl
8k: R = N-tosyl-3-indolyl
8l: R = ferrocenyl
8m: R = t-Bu
The 3,5-dioxopentanoates 9, which contain three inter-
connected functional groups, are useful precursors for the
synthesis of a large number of heterocyclic compounds4 or
in the production of cholesterol-lowering drugs such as at-
arvostatin.5 Moreover, they are intermediates in the biosyn-
thesis of several secondary metabolites that are generated
through the polyketide pathway.6 Only a limited number of
methods are available for the synthesis of 3,5-diketo es-
ters.7 The most common procedure is the regioselective
Knoevenagel condensation of acetoacetates with esters un-
der basic conditions.8 A recent variation on this method is
the condensation of 1,3-bis(silyl enol ethers) with acid
chlorides.9 Although, the Knoevenagel condensation and its
variations can deliver 3,5-diketo esters in one step, the re-
actions have serious limitations, such as moderate yields,
restriction to β-keto esters with simple aliphatic or aromat-
ic substituents, and stringent reaction conditions. Further-
more, such condensation reactions do not deliver 3-amino-
5-oxopent-3-enoates 10.
For our initial studies, we selected the zinc-mediated
condensation of 3-oxo-3-phenylpropanenitrile (phenacyl
cyanide; 8a, R = Ph) with ethyl bromoacetate (2) to give the
corresponding two-carbon homologated product 9 (Blaise
product) or 11 (Reformatsky product). Initially, we con-
ducted the experiments under the conditions that we pre-
viously described for the synthesis of 3-aminoenones 7.3
The reaction of phenacyl cyanide (8a), ethyl bromoacetate
Scheme 2 Synthesis of 3,5-diketo esters 9a–m from corresponding α-
cyano ketones 8a–m
To generalize and expand the scope of the synthesis of
the 3,5-dioxopentanoates 9, we conducted the zinc-mediat-
ed condensation of various α-cyano ketones 8b–m with
ethyl bromoacetate under the optimized conditions
(Scheme 2). We carefully selected 3-oxo-3-arylpropane-
nitriles 8b–e to permit the evaluation of the influence of
electronic factors relating to the benzene ring on the out-
come of the reaction. Reactions conducted with a cyano ke-
tone substituted with an electron-withdrawing para-fluoro
group (8b), a moderately electron-withdrawing but ortho–
para-directing para-chloro group (8c), an electron-donat-
ing para-methyl group (8d), or a strongly electron-donating
para-methoxy substituent (8e) gave the corresponding 3,5-
diketo esters 9b–e in about 70% yield. These results indicat-
ed that the electron density in the benzene ring does not
have much bearing on the outcome of the reaction. The
substrate scope was further extended by placing a 3-oxo-
propanenitrile group on a polycyclic aromatic ring in the
naphthalene 8f and the pyrene 8g; these gave the corre-
sponding 3,5-diketo esters 9f and 9g, respectively, albeit in
lower yields. Next, we conducted the reaction of heteroaro-
matic compounds possessing a 3-oxopropanenitrile func-
tional group, such as the 2-furyl derivative 8h, the 2-thienyl
derivative 8i, the N-tosyl-3-pyrrolyl derivative 8j, and the
© Georg Thieme Verlag Stuttgart · New York — Synlett 2016, 27, A–E

C
H. S. Rao, N. Muthanna
Letter
Synlett
N-tosyl-3-indolyl derivative 8k; these also gave the corre-
sponding 3,5-diketo esters 9h–j in good yields. However,
the reaction of 3-oxo-3-pyridin-3-ylpropanenitrile did not
proceed, as the nitrile did not survive the reaction condi-
tions. The reaction of 3-oxo-3-ferrocenylpropanenitrile 8l
provided the 3,5-dioxopentanoate 9l in good yield. Finally,
we subjected 4,4-dimethyl-3-oxopentanenitrile 8m to the
Blaise reaction to evaluate the influences of the aliphatic
tert-butyl group and the concomitant steric effects on the
outcome of the reaction. The reaction proceeded without
difficulty to provide the 3,5-diketo ester 9m. The Blaise
variant did not, however, work with 3-oxooctanenitrile, in-
dicating that alternate deprotonation α to the carbonyl
group might affect the outcome of the reaction.
(i) BrCH2CO2Et (2 equiv)
O
O
NH2
O
2
CN
R
R
OEt
Zn (3 equiv), TMSCl (10 mol%),
THF, reflux, 3–5 h
(ii) aq K2CO3 (50% w/v), 0 °C to rt
62–82%
8a, c, 8e, g, k, m
10a–f
8a: R = C6H5
10a: R = C6H5 , 4 h (77%)
8c: R = 4-ClC6H4
8e: R = 4-MeOC6H4
8g: R = 1-pyrenyl
8k: R = N-tosyl-3-indolyl
8m: R = t-Bu
10b: R = 4-ClC6H4, 4 h (79%)
10c: R = 4-MeOC6H4, 5 h (72%)
10d: R = 1-pyrenyl, 3 h (62%)
10e: R = N-tosyl-3-indolyl, 3 h (67%)
10f: R = t-Bu, 4 h (82%)
Scheme 3 Synthesis of δ-keto β-amino β,γ-unsaturated esters 10a–f
It became clear from these results that the Blaise reac-
tion on 8 takes place ahead of the Reformatsky reaction. To
probe the reaction further, we conducted competition ex-
periments between 3-phenylpropanenitrile (13) and ace-
tophenone (15) with ethyl bromoacetate (2) under the opti-
mized conditions. Initially, when we carried out the reac-
tion between 13 and 2, the reaction provided the Blaise
product, β-keto ester 14 (Scheme 4, eq. 1). The zinc-mediat-
ed reaction of 15 with 2 provided the Reformatsky product,
β-hydroxy ester 16 (Scheme 4, eq. 2). Subsequently, we
conducted a reaction by employing a mixture of 1 mmol
each of 3-phenylpropanenitrile (13) and acetophenone (15)
with 1 mmol (1 equivalent) of 2; this reaction provided the
Reformastky product 16 exclusively (Scheme 4, eq. 3). This
set of reactions clearly demonstrated that the Reformatsky
reaction takes place more readily than the Blaise reaction.
However, when the ketone and nitrile are in a geminal rela-
tion, the Blaise product is favored.
A possible mechanism for the Blaise reaction of α-cyano
ketones is shown in Scheme 5. Oxidative insertion of acti-
vated Zn into the C–Br bond of ethyl bromoacetate (2) leads
to an organozinc intermediate that rearranges to the zinc
enolate of the acetic acid ester 17.10 Nucleophilic addition
of the zinc enolate 17 to activated nitrile 18 leads to the
Blaise intermediate 19, which is in equilibrium with enam-
ine 20. Hydrolysis of the intermediate 20 under acidic con-
ditions leads to diketo ester 9 via the geminal amino alcohol
Having successfully explored the scope of the Blaise re-
action, we reasoned that basic, rather than acidic, hydroly-
sis of the reaction mixture might provide 3-amino-5-oxo-
pent-3-enoates 10. Indeed, basic workup of the reaction
mixture from 8a with 50% aqueous K2CO3 provided ethyl
(3Z)-3-amino-5-oxo-5-phenylpent-3-enoate (10a) in 77%
yield (Scheme 3). The structure of 10a was confirmed by
means of analytical and spectroscopic data (IR, 1H NMR, 13
C
NMR, DEPT-135, COSY, HSQC, and HMBC spectra). The con-
figuration of the double bond in 10a was assigned as Z by
taking into consideration the intramolecular hydrogen
bonding of the primary amine with the two carbonyl
1
groups. The H and 13C NMR spectroscopic correlations ex-
hibited in the HMBC spectrum clearly showed that the dou-
ble bond is present between the C3 and C4 atoms. For ex-
ample, the HMBC spectra revealed cross peaks for the ole-
finic proton (δ = 5.71 ppm) and the carbonyl carbon (δ =
189.9 ppm), as well as for the enamine carbon (δ = 157.7
ppm); these show that the enamine double bond is conju-
gated to the carbonyl group rather than the ester. The trans-
formation proved to be general, and we synthesized five
more 3-amino-5-oxopent-3-enoates 10b–f. In general,
enamines 10a–f were unstable at room temperature (30 °C)
or in CDCl3, but were stable at –5 °C for over a year.
O
O
(i) BrCH2CO2Et (2 equiv), Zn (3 equiv),
TMSCl (10 mol%), THF, reflux, 3 h
CN
(1)
OEt
(ii) 3 N HCl, 0 °C to rt, 91%
13
14
OH
Me
O
O
OEt
OEt
(i) BrCH2CO2Et (2 equiv), Zn (3 equiv),
TMSCl (10 mol%), THF reflux, 3 h
Me
Me
(2)
(3)
O
(ii) 3 N HCl, 0 °C to rt, 94%
15
16
OH
Me
CN
(i) BrCH2CO2Et (2 equiv), Zn (3 equiv),
TMSCl (10 mol%), THF reflux, 3 h
O
15
13
16
(ii) 3 N HCl, 0 °C to rt, 94%
Scheme 4 Competition reactions between acetophenone and 3-phenylpropanenitrile
© Georg Thieme Verlag Stuttgart · New York — Synlett 2016, 27, A–E

D
H. S. Rao, N. Muthanna
Letter
Synlett
Zn2+
O
O
CN
R
N
Zn2+
N
8
+
Zn(0)
R
nucleophilic
addition
O
OEt
18
COOEt
ZnBr
NH2
Br
COOEt
R
O
19
2
17
O
O
O
COOEt
Zn2+
H3O+
OH–
R
R
O
9
H2O
O
NH
COOEt
R
NH2
COOEt
OH
21
R
COOEt
20
11
Scheme 5 Proposed mechanism for the formation of 3,5-diketo esters
21. Hydrolysis under basic conditions, on the other hand,
provides the 3-amino-5-oxopent-3-enoates 11.
Blaise reaction is preferred when the keto and nitrile func-
tional groups are in a geminal relationship and if there is a
possibility for enolization.
Crucial to success of the Blaise reaction on α-cyano ke-
tones is the zinc-mediated activation of the nitrile, as
shown in 18 (Scheme 5), wherein zinc complexation leads
to increased electrophilicity of the nitrile. We reasoned that
if enolization could be arrested, the Blaise pathway would
be difficult and the reaction of α-cyano ketones might take
the Reformatsky pathway. To validate this hypothesis, we
carried out a reaction between 2,2-dimethyl-3-oxo-3-
phenylpropanenitrile (22) and ethyl bromoacetate (2)
(Scheme 6). The reaction provided ethyl benzoylacetate
(23) exclusively, evidently through the Reformatsky path-
way. Thus, after the addition of the zinc enolate to the car-
bonyl carbon, elimination of an isobutyronitrile α-anion
takes place to leave the β-keto ester 23 (Scheme 6).
Acknowledgment
H.S.P.R. thanks the University Grants Commission (UGC), Special As-
sistance Program (SAP), Major Research Project (MRP), and Depart-
ment of Science and Technology Fund for Improvement of S&T
Infrastructure in Universities and Higher Educational Institutions
(DST-FIST) programs. N.M. thanks the UGC for a fellowship. The au-
thors thank the Central Instrumentation Facility (CIF) and the Depart-
ment of Chemistry at Pondicherry University for instrumentation
facilities.
Supporting Information
Supporting information for this article is available online at
(i) BrCH2CO2Et (2 equiv), Zn (3 equiv)
TMSCl (10 mol%), THF, reflux, 2 h
O
O
O
http://dx.doi.org/10.1055/s-0035-1561662.
S
u
p
p
ortiInfogrmoaitn
S
u
p
p
o
nrtogI
f
rmoaitn
CN
OEt
(ii) 3 N HCl, 0 °C to rt, 30 min, 91%
Me Me
References
22
23
Scheme 6 Generation of β-keto ester 23, the Reformatsky product,
from 2,2-dimethyl-3-oxo-3-phenylpropanenitrile (22)
(1) (a) Blaise, E. E. C. R. Hebd. Seances Acad. Sci. 1901, 132, 478.
(b) Blaise, E. E. C. R. Hebd. Seances Acad. Sci. 1901, 132, 987.
(c) Blaise, E. E.; Courtot, A. Bull. Soc. Chim. Fr. 1906, 35, 589. For a
review, see: (d) Rao, H. S. P.; Rafi, S.; Padmavathy, K. Tetrahedron
2008, 64, 8037.
(2) (a) Reformatsky, S. Ber. Dtsch. Chem. Ges. 1887, 20, 1210.
(b) Reformatsky, S. Zh. Russ. Fiz.-Khim. O-va. 1890, 22, 44. For
reviews, see: (c) Shriner, R. L. Org. React. 1942, 1, 1. (d) Rathke,
M. W. Org. React. 1975, 22, 423.
(3) Rao, H. S. P.; Muthanna, N. Eur. J. Org. Chem. 2015, 1525.
(4) (a) Douglas, C. J.; Sklenicka, H. M.; Shen, H. C.; Mathias, D. S.;
Degen, S. J.; Golding, G. M.; Morgan, C. D.; Shih, R. A.; Mueller, K.
L.; Seurer, L. M.; Johnson, E. W.; Hsung, R. P. Tetrahedron 1999,
55, 13683. (b) Dong, Y.; Shi, Q.; Nakagawa-Goto, K.; Wu, P.-C.;
Morris-Natschke, S. L.; Brossi, A.; Bastow, K. F.; Lang, J.-Y.; Hung,
M.-C.; Lee, K.-H. Bioorg. Med. Chem. 2010, 18, 803. (c) Schmidt,
D.; Conrad, J.; Klaiber, I.; Beifuss, U. Chem. Commun. 2006, 4732.
(5) Roth, B. D. Prog. Med. Chem. 2002, 40, 1.
In conclusion, the zinc-mediated condensation of 3-
aryl-3-oxopropanenitriles with ethyl bromoacetate fur-
nishes the corresponding 3,5-diketo esters. This reaction is
a useful variant for two-carbon homologation of readily
available α-keto nitriles to give a range of 3,5-dioxopenta-
noates.11 A variety of 3,5-diketo esters having aromatic, het-
eroaromatic, ferrocenyl, or aliphatic entities in the C5 posi-
tion were prepared by this method. Acidic workup of the
reaction furnished 3,5-dioxopentanoates, whereas basic
workup furnished ethyl (3Z)-3-amino-5-oxo-5-pent-3-
enoates. Thus, the method permits the direct and regiose-
lective synthesis of 3,5-dioxopentanoates or the corre-
sponding enamines. Mechanistic probing revealed that the
(6) (a) Koskinen, A. M. P.; Karisalmi, K. Chem. Soc. Rev. 2005, 34,
677. (b) Staunton, J.; Weissman, K. J. Nat. Prod. Rep. 2001, 18,
380.
© Georg Thieme Verlag Stuttgart · New York — Synlett 2016, 27, A–E

E
H. S. Rao, N. Muthanna
Letter
Synlett
(7) (a) Hegde, S. G. In e-EROS, Encyclopedia of Reagents for Organic
Synthesis [Online]; Fuchs, P. L., Ed.; Wiley: Hoboken, 2001, ; DOI:
10.1002/047084289X.re040. (b) Yuste, F.; Breña, F. K.; Barrios,
H.; Sánchez-Obregón, R.; Ortíz, B.; Walls, F. Synth. Commun.
1988, 18, 735.
(8) Huckin, S. N.; Weiler, L. Can. J. Chem. 1974, 52, 1343.
(9) Rahn, T.; Nguyen, V. T. H.; Dang, T. H. T.; Ahmed, Z.; Methling,
K.; Lalk, M.; Fischer, C.; Spannenberg, A.; Langer, P. J. Org. Chem.
2007, 72, 1957.
under reduced pressure, the residue was diluted with CH2Cl2
(20 mL), and the resulting solution was stirred with 50% aq
K2CO3 (5 mL) for 15 min. The organic layer was separated,
washed sequentially with H2O (2 × 20 mL) and brine (10 mL),
dried (Na2SO4), filtered, and concentrated under reduced pres-
sure to give the crude (3Z)-3-amino-5-oxo-5-pent-3-enoate,
which was purified by column chromatography [silica gel (100–
200 mesh), 15% EtOAc–hexane].
Ethyl 5-(4-Chlorophenyl)-3,5-dioxopentanoate (9c)
(10) Dekker, J.; Boersma, J.; van der Kerk, G. J. M. J. Chem. Soc., Chem.
Commun. 1983, 553.
(11) 3,5-Dioxopentanoates 9a–m and 3-Amino-5-oxopent-3-
enoates 10a–f; General Procedure
Prepared by the general procedure (Method A) from 3-(4-chlo-
rophenyl)-3-oxopropanenitrile (8c; 503 mg, 2.79 mmol) and
ethyl bromoacetate (2; 932 mg, 5.58 mmol) in the presence of
Zn powder (544 mg, 8.38 mmol) and TMSCl (30 mg, 10 mol %)
in THF (10 mL) for 4 h followed by hydrolysis with 3 N HCl (2
mL) as a pale-yellow liquid; yield: 553 mg (74%); Rf = 0.5 (10%
EtOAc). IR: 2983, 2934, 1741, 1719, 1607, 1318, 1267, 1038, 817
A solution of TMSCl (10 mol%) in THF (1 mL) was added to a sus-
pension of Zn powder (1.2 mmol, 3 equiv) in THF (8 mL), and
the mixture was refluxed with vigorous stirring for 25 min. A
solution of the appropriate α-cyano ketone 8 (0.4 mmol, 1
equiv) in THF (2 mL) and a solution of the α-bromo ester (0.8
mmol) in THF (2 mL) were simultaneously added dropwise over
30 min by using two syringes. After the addition, the pale-green
mixture was refluxed until all the α-cyano ketone was con-
sumed (TLC; 3–6 h), by which time the color had changed to
brown. The mixture was then cooled to r.t. and centrifuged at
700 rpm. The upper solution was collected, and the residual
solid was washed with THF (2 × 2 mL) and the solvent was col-
lected after centrifugation.
cm–1 1H NMR (400 MHz, CDCl3 + CCl4, 1:1): δ = 15.73 (s, 1 H),
.
7.82 (d, J = 8.6 Hz, 2 H), 7.42 (d, J = 8.6 Hz, 2 H), 6.24 (s, 1 H), 4.22
(q, J = 7.2, 2 H), 3.47 (s, 2 H), 1.30 (t, J = 7.1 Hz, 3 H). 13C NMR
(100 MHz, CDCl3 + CCl4, 1:1): δ = 189.3, 181.4, 167.3, 139.1,
132.8, 129.1, 128.6, 96.6, 61.6, 45.9, 14.3. HRMS (ESI): m/z [M +
H] calcd for C13H14ClO4: 269.0581; found: 269.0580.
Ethyl (3Z)-3-Amino-5-oxo-5-phenylpent-3-enoate (10a)
Prepared by the general procedure (Method B) from 3-oxo-3-
phenylpropanenitrile (8a; 502 mg, 3.44 mmol) and ethyl bro-
moacetate (2; 1.15 g, 6.89 mmol) in the presence of Zn powder
(672 mg, 10.34 mmol) and TMSCl (38 mg, 10 mol %) in THF (10
mL) for 4 h followed by hydrolysis with 50% aq K2CO3 (5 mL) as a
pale-yellow liquid; yield: 627 mg (77%); Rf = 0.3 (10% EtOAc–
hexane). IR: 3413, 3186, 2982, 2927, 1734, 1616, 1527, 1282,
Method A: 3,5-Diketo Esters 9a–m
The pH of the combined THF solutions was adjusted to 2 with 3
M HCl (~2 mL) and the mixture was stirred for 30 min at r.t. The
THF was removed on a rotary evaporator under reduced pres-
sure, and the residue was diluted with CH2Cl2 (20 mL) and H2O
(20 mL). The organic layer was separated, washed sequentially
with H2O (2 × 20 mL) and brine (10 mL), dried (Na2SO4), filtered,
and concentrated under reduced pressure to give the crude 3,5-
diketo ester, which was purified by column chromatography
[silica gel (100–200 mesh), 10% EtOAc–hexane].
1175, 1060, 737 cm–1 1H NMR (400 MHz, CDCl3 + CCl4, 1:1):
.
δ = 10.14 (br s, 1 H), 7.85–7.82 (m, 2 H), 7.42–7.36 (m, 3 H), 6.17
(br s, 1 H), 5.71 (s, 1 H), 4.21 (q, J = 7.1 Hz, 2 H), 3.26 (s, 2 H),
1.29 (t, J = 7.1 Hz, 3 H). 13C NMR (100 MHz, CDCl3 + CCl4, 1:1):
δ = 189.9, 169.0, 157.7, 140.0, 131.1, 128.3, 127.3, 93.0, 61.7,
41.0, 14.3. HRMS (ESI): m/z [M + H] calcd for C13H16NO3:
234.1130; found: 234.1128.
Method B: (3Z)-3-Amino-5-oxo-5-pent-3-enoates 10a–f
The solvent was removed from the combined THF solutions
© Georg Thieme Verlag Stuttgart · New York — Synlett 2016, 27, A–E