Expanding Blaise-Type Reactions towards Indium-Mediated Transformations of α-Bromo-β-keto Esters with Nitriles

Article abstract of DOI:10.1002/ejoc.201700868

The aim of this work is the identification of mild reaction conditions for the Blaise-type transformation of brominated β-keto esters with nitriles to generate enamino-substituted keto esters. The best results were obtained when a combination of indium metal (0.7 equiv.) with indium trichloride (1.6 equiv.) were applied at 60 °C for 20 to 72 hours, and these conditions could be applied to a broad range of nitriles and a significant number of different β-keto esters. The transformation of aliphatic nitriles proved to be difficult and gave only moderate yields. However, aromatic nitriles gave good yields in many cases. The applicability range of β-keto esters is acceptable while some electron-deficient aryl-substituents on the keto ester were challenging substrates. Nevertheless, we were able to expand the scope of the Blaise-type reaction towards brominated β-keto esters significantly.

Full text of DOI:10.1002/ejoc.201700868

DOI: 10.1002/ejoc.201700868
Communication
Blaise-Type Reaction
Expanding Blaise-Type Reactions towards Indium-Mediated
Transformations of α-Bromo-ꢀ-keto Esters with Nitriles
Luomo Li,[a] Emre Babaoglu,[a] Klaus Harms,[a] and Gerhard Hilt*[a,b]
Abstract: The aim of this work is the identification of mild reac- phatic nitriles proved to be difficult and gave only moderate
tion conditions for the Blaise-type transformation of bromin- yields. However, aromatic nitriles gave good yields in many
ated ꢀ-keto esters with nitriles to generate enamino-substituted cases. The applicability range of ꢀ-keto esters is acceptable
keto esters. The best results were obtained when a combination while some electron-deficient aryl-substituents on the keto
of indium metal (0.7 equiv.) with indium trichloride (1.6 equiv.) ester were challenging substrates. Nevertheless, we were able
were applied at 60 °C for 20 to 72 hours, and these conditions to expand the scope of the Blaise-type reaction towards brom-
could be applied to a broad range of nitriles and a significant inated ꢀ-keto esters significantly.
number of different ꢀ-keto esters. The transformation of ali-
Introduction
2-aminonaphthalenes or naphtho[2,3-c]furanes were ob-
tained.[7]
The zinc-mediated Blaise reaction has long been known.[1] Its
limitation is mostly the application of bromo-substituted acetic
esters, which can be reacted with nitriles to generate either
aminoacrylates or ꢀ-keto esters depending on the work-up con-
ditions applied to the zinc intermediate 1 (Scheme 1). Although
established more than a century ago, recent advances in new
conditions to overcome the limitations of the traditional Blaise
reaction have to be mentioned. First of all, the Lee group
reported the application of α-bromoacetates in combination
with nitriles. In this case the primary intermediate 1 was de-
protonated with a strong base (nBuLi) and quenching of the
anion with an acyl chloride was reported to generate the formal
product 2 of a Blaise reaction starting from an α-bromo-ꢀ-keto
ester with a nitrile.[2] Shortly thereafter, the Lee group reported
the reaction of aminoacrylates with hydrazine derivatives to
generate pyrazoles,[3] whereas the Thomas group used the
formylation of the aminoacrylates and subsequent cyclisation
to generate interesting pyridine derivatives.[4] Additionally, the
reaction of the aminoacrylates with alkynes was reported to
generate polysubstituted 1-amino-substituted 1,3-dienes,
which led to a number of promising follow-up reactions.[5] De-
pending on substituents and the reaction conditions, different
types of products could be synthesised, including a palladium-
catalysed carbonylative cyclisation sequence for the forma-
tion of isoindolines.[6] Recently, the Fan group reported the
intramolecular version of a protocol in which an α-bromoester
was treated with a 2-alkynyl benzonitrile derivative, and either
Scheme 1. Blaise reaction and recent advances.
Some time ago, we were able to expand the scope of the
Blaise-type transformation towards α-bromo-malonates and to
realise carbon–carbon bond formations with nitriles to generate
products of type 3 as well as isocyanates to form 2-carbamoyl
malonates.[8] Although similar transformations of non-bromin-
ated starting materials with nitriles have been reported, we
continued our efforts to find mild reaction conditions for the
Blaise-type reactions with brominated ꢀ-keto esters instead of
malonates to enlarge the scope even further. Thereby, we
wanted to establish a versatile alternative to the Lewis acid
induced transformation of non-brominated ꢀ-keto esters with
nitriles, such as TiCl4 or SnCl4.[9]
[a] Fachbereich Chemie, Philipps-Universität Marburg,
Hans-Meerwein-Str. 4, 35043 Marburg, Germany
[b] Current address: Institut für Chemie, Universität Oldenburg,
Carl-von-Ossietzky-Str. 9-11, 26111 Oldenburg, Germany
E-mail: Gerhard.Hilt@uni-oldenburg.de
Results and Discussion
In this report, we focus on the carbon–carbon bond formation
of bromo-substituted ꢀ-keto esters with nitriles for the synthe-
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/ejoc.201700868.
Eur. J. Org. Chem. 2017, 4543–4547
4543
© 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
sis of enamino-substituted ꢀ-keto esters (5). As test system, we temperature, 60 °C was found to be the optimum temperature.
chose the ethyl 2-bromo-3-oxobutenoate (4, Scheme 2) and Below 50 °C the reaction is very slow, and at higher tempera-
benzonitrile as suitable starting materials. Like the reaction with tures (entry 12) side reactions were encountered.
bromomalonate for the formation of 3, the corresponding
transformation towards 5 with zinc powder as activating agent
failed.[8]
The product (5) was isolated in up to 83 % yield when in-
dium shots were applied whereas the yield dropped slightly to
75 % when indium powder was applied.
With the optimised conditions in hand, we then investigated
the application of a number of nitriles in the indium metal/
indium trichloride mediated Blaise-type reaction (Scheme 3).
The results of this series of experiments are summarised in
Table 2.
Scheme 2. Outline of the optimisation of the reaction conditions of 4 with
benzonitrile.
Also, when indium trichloride[10] was used without a reduc-
ing metal, only traces of product were observed. However,
when zinc powder was added as additive, 26 % yield of 5 could
be achieved. When the zinc powder was substituted with
indium metal, a significantly higher yield was obtained.[11] Ac-
cordingly, different reaction conditions were screened to
achieve reasonable turnovers of 4. The results of this series of
experiments are summarised in Table 1.
Scheme 3. Indium metal/indium trichloride-mediated transformation of α-
bromo-ꢀ-keto ester 4 with different nitriles.
For this set of experiments, a number of aromatic nitriles,
including electron-rich as well as electron-deficient aryl nitriles,
were applied successfully. On the other hand, the application
of aliphatic nitriles was more challenging and led to signifi-
cantly lower yields. Nevertheless, we would like to highlight the
finding that the application of cyclopropanecarbonitrile is of
mechanistic interest (entry 19).[12] The reaction does not seem
to proceed through a radical mechanism because ring-opened
products could not be detected and only product 5r was iso-
lated in 53 % yield.
The functional-group tolerance of the transformation is quite
acceptable; alkoxy groups, aryl halides, esters, aromatic nitro
groups as well as heterocyclic substrates were applicable. Only
in a few cases, functional groups seem to give lower yields,
such as nitro groups (entry 10) or pinacolboronic ester groups
(entry 7). Unfortunately, the application of the boronic ester
gave only low yields, though this would be an interesting build-
ing block for further transformations. Sterically hindered nitriles,
such as 2-methoxybenzonitrile, gave the desired product 5c in
only slightly diminished yield (entry 3). Also, 2- and 4-bromo-
as well as 4-fluorobenzonitrile are well accepted as starting ma-
terials and no dehalogenation of the arene moiety was ob-
served (entries 4–6). Moreover, heterocyclic nitriles were appli-
cable (entries 11 and 12) and it is noteworthy that in case of
product 5l the Boc protecting group was removed under the
reaction condition. On the other hand, the unprotected indole-
3-carbonitrile did not undergo the desired reaction. Unfortu-
nately, the application of aliphatic nitriles gave only diminished
yields of up to 23 % yield (entries 14–16) with the exception of
the cyclopropanecarbonitrile as already mentioned.
Table 1. Optimisation studies on the indium/indium trihalide mediated Blaise-
type reaction for the synthesis of 5.[a,b]
Entry
Solvent
InX3
Indium[c]
T [°C]
Yield[a]
1
2
3
4
5
6
7
8
(CH2Cl)2
(CH2Cl)2
(CH2Cl)2
(CH2Cl)2
(CH2Cl)2
(CH2Cl)2
(CH2Cl)2
(CH2Cl)2
(CH2Cl)2
(CH2Cl)2
(CH2Cl)2
(CH2Cl)2
THF
0.7 equiv. InCl3
1.0 equiv. InCl3
1.4 equiv. InCl3
1.6 equiv. InCl3
1.6 equiv. InBr3
1.6 equiv. InCl3
1.8 equiv. InCl3
1.6 equiv. InCl3
1.6 equiv. InCl3
1.6 equiv. InCl3
1.6 equiv. InCl3
1.6 equiv. InCl3
1.6 equiv. InCl3
1.6 equiv. InCl3
1.6 equiv. InCl3
0.7 equiv.
0.7 equiv.
0.7 equiv.
0.7 equiv.
0.7 equiv.
0.7 equiv.[e]
0.7 equiv.
0.5 equiv.
0.8 equiv.
0.9 equiv.
0.9 equiv.
0.7 equiv.
0.7 equiv.
0.7 equiv.
0.7 equiv.
60
60
60
60
60
60
60
60
60
60
60
70
60
60
50
58 %
74 %
76 %
83 %[d]
81 %
75 %
73 %
75 %
75 %
82 %
78 %
66 %
4 %
9
10
11
12
13
14
15
toluene
CH2Cl2
66 %
70 %
[a] Reaction conditions: ethyl 2-bromo-3-oxobutanoate (1.0 mmol), benzo-
nitrile (2.0 mmol), solvent (1.0 mL), sealed tube. [b] The yields were deter-
mined by 1H NMR with 1,3,5-trimethoxybenzene as internal standard. [c] Cut
indium shots were used. [d] The product was isolated in 83 % yield.
The optimisation process of the indium-mediated Blaise-type
reaction showed that indium alone led to a very low yield of the
desired product 5. Also, the application of indium trichloride in
the absence of any metal gave only trace amounts of 5. As
mentioned before, the results for the application of zinc powder
and indium trichloride furnished product 5 already in 26 %
yield. The replacement of zinc towards indium metal gave satis-
factory results when more than 0.5 equiv. of indium were used
while with more than 0.7 equiv. of indium (entries 9–11) no
further increase of the yield could be achieved. The optimum
amount of indium trichloride was 1.6 equiv. Higher as well as
lower amounts of InCl3 were found to be not beneficial (entries
1–3 and 7), and the replacement of InCl3 toward InBr3 gave no
In the next set of experiments, we focused our attention on
the application of different α-bromo-ꢀ-keto esters 6 (Scheme 4)
in the reaction with benzonitrile to afford enamino-substituted
keto esters 7. The results of these experiments are summarised
in Table 3.
The scope of the ꢀ-keto esters was evaluated by applying
significant change (entry 5). Also, the application of other sol- electron-rich (4-methoxyphenyl-substituted), electron-neutral
vents (entries 13–15) proved to be inferior to 1,2-dichloro- (phenyl) and electron-deficient (4-fluorophenyl-substituted) α-
ethane, which is the solvent of choice. Concerning the reaction bromo-ꢀ-keto esters as starting materials (entries 1–5). These
Eur. J. Org. Chem. 2017, 4543–4547
www.eurjoc.org
4544
© 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
Table 2. Results of the indium metal/indium trichloride mediated Blaise-type reaction of 4 with different nitriles.
[a] Reaction conditions: α-bromo-ꢀ-keto ester 4 (1.0 mmol), nitrile (2.0 mmol), cut indium shots (0.7 mmol), indium trichloride (1.7 mmol), DCE (1.0 mL), 60 °C,
[b]
sealed tube. Crystallographic data for 5a, see CCDC 1555178.
reactions revealed that electron-neutral as well as electron-rich 7e were isolated in 22 % and 21 %, respectively. Aliphatic ꢀ-
starting materials are well accepted (59 % for 7a and 54 % for keto esters gave moderate yields of around 40 % for 7g and
7b), while electron-deficient starting materials gave only a very 7h, while the cyclopropyl-substituted derivative 7f could be ob-
poor yield (7c: 14 %). The yields for the bromophenyl-substi- tained in a good yield of 70 %. As already observed for the
tuted starting materials were not significantly higher, as 7d and transformation of cyclopropanecarbonitrile (Table 2, entry 18),
Eur. J. Org. Chem. 2017, 4543–4547
www.eurjoc.org
4545
© 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
to be an excellent indicator. For the Z isomer Z-7, the CH3 pro-
tons gave a signal 1H NMR spectrum around δ = 1.0 ppm,
whereas for the E-isomer E-7, a resonance was found around
δ = 0.7 ppm. Thereby, an easy assignment of the two sets of
signals could be made (Figure 1).
Scheme 4. Indium metal/indium trichloride mediated reaction applying dif-
ferent α-bromo-ꢀ-keto esters of type 6.
Table 3. Results of the indium metal/indium trichloride mediated Blaise-type
reaction of α-bromo-ꢀ-keto esters 6 with benzonitrile.
Figure 1. Assignment of the E/Zisomers of 7 based on 1H NMR shifts.
´
When a single crystal, used for the X-ray analysis of Z-7a
(CCDC 1555179), was dissolved in CDCl3 and analysed by 1H
NMR, nearly the same ratio of E/Z isomers as in the original
sample was obtained. Accordingly, we assume that the E/Z
isomerisation takes place in situ and is not dependent on the
reaction mechanism or any possible induction by either the
indium metal or the indium Lewis acid applied.
Last but not least, we compared the reaction conditions opti-
mised for the application of α-bromo-ꢀ-keto esters with those
found previously for the reaction of diethyl α-bromo-malonate
with benzonitrile (Scheme 5).[8]
Scheme 5. Comparison of the indium metal mediated with the indium metal/
indium trichloride mediated synthesis of 3.
The overall amounts of indium metal and indium salts ap-
plied in these reactions are similar and the reaction conditions
concerning solvent, reaction temperature and reaction time are
comparable. However, the desired product 3 could be isolated
in nearly quantitative yield (95 %), representing a significant im-
provement compared to the earlier reported reaction condi-
tions.
Finally, the question of the activation mode of the α-bromo-
1
ꢀ-keto ester was investigated by H NMR experiments. The in-
terpretation of these spectra suggest that the Lewis acid coordi-
nates to the α-bromo-ꢀ-keto ester (A) and leads to the release
of a bromonium ion (Br+) and the formation of an indium-enol-
ate equivalent (B) (Scheme 6).
The equilibrium of this process is probably shifted irreversi-
bly towards the indium enolate when the Br+ ion is reduced by
[a] Reaction conditions: α-bromo-ꢀ-keto ester 6 (1.0 mmol), benzonitrile
(2.0 mmol), cut indium shots (0.7 mmol), indium trichloride (1.7 mmol), DCE
(1.0 mL), 60 °C, sealed tube.
no ring-opened side product could be detected in the case of
the formation of 7f.
In principle, the products of type 5 could be obtained as a
mixture of E and Z isomers. However, the products of type 5
(Table 2) were isolated as E isomers only. In contrast, the appli-
cation of aryl-substituted α-bromo-ꢀ-keto esters gave mixtures
of E/Z isomers of products 7 (see Table 3). For the assignment
of the E and Z isomer, the CH3 group of the ethyl ester proved
Scheme 6. Comparison of the indium metal mediated with the indium metal/
indium trichloride mediated synthesis of 3.
Eur. J. Org. Chem. 2017, 4543–4547
www.eurjoc.org
4546
© 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
hanna, Eur. J. Org. Chem. 2015, 1525–1532; f) Y. S. Chun, K. Y. Ryu, J. H.
Shin, S.-g. Lee, Org. Biomol. Chem. 2011, 9, 1317–1319; g) J. H. Kim, S. Y.
Choi, J. Bouffard, S.-g. Lee, J. Org. Chem. 2014, 79, 9253–9261.
[2] Y. S. Chun, K. K. Lee, Y. O. Ko, H. Shin, S.-g. Lee, Chem. Commun. 2008,
5098–5100.
the indium metal to generate InBr3, and the Lewis acid further
activates the nitrile (C) for the nucleophilic attack of the indium-
enolate B, which leads to the formation of the products.
[3] Y. O. Ko, Y. S. Chun, C.-L. Park, Y. Kim, H. Shin, S. Ahn, J. Hong, S.-g. Lee,
Org. Biomol. Chem. 2009, 7, 1132–1136.
[4] a) S. Kumar, A. A. Sawant, R. P. Chikhale, K. Karanjai, A. Thomas, J. Org.
Chem. 2016, 81, 1645–1653; b) Y. S. Chun, J. H. Lee, J. H. Kim, Y. O. Ko,
S.-g. Lee, Org. Lett. 2011, 13, 6390–6393.
Conclusions
With the results presented herein, we have been able to identify
mild reaction conditions for the successful application of α-
bromo-ꢀ-keto esters in indium metal/indium trichloride medi-
ated Blaise-type reactions with nitriles. A broad scope of the
reaction concerning different nitriles and various α-bromo-ꢀ-
keto esters could be demonstrated. A number of interesting
products was obtained, which open the way to explore promis-
ing follow-up reactions. These results justify the extra step to
generate the brominated keto ester compared to similar trans-
formations with strong Lewis acids because not all functional
groups will tolerate harsh reaction conditions.
[5] a) Y. S. Chun, Y. O. Ko, H. Shin, S.-g. Lee, Org. Lett. 2009, 11, 3414–3417;
b) Y. S. Chun, K. Y. Ryu, Y. O. Ko, J. Y. Hong, J. Hong, H. Shin, S. Lee, J.
Org. Chem. 2009, 74, 7556–7558; c) J. H. Kim, J. Bouffard, S. Lee, Angew.
Chem. Int. Ed. 2014, 53, 6435–6438; Angew. Chem. 2014, 126, 6553–6556;
d) J. H. Kim, S.-g. Lee, Org. Lett. 2011, 13, 1350–1353; e) Z. Xuan, J. H.
Kim, S.-g. Lee, J. Org. Chem. 2015, 80, 7824–7829; f) J. H. Kim, Y. S. Chun,
S.-g. Lee, J. Org. Chem. 2013, 78, 11483–11493.
[6] a) Z. Xuan, D. J. Jung, H. J. Jeon, S.-g. Lee, J. Org. Chem. 2016, 81, 10094–
10098. See also b) J. H. Kim, H. Shin, S.-g. Lee, J. Org. Chem. 2012, 77,
1560–1565; c) J. H. Kim, S.-g. Lee, Synthesis 2012, 44, 1464–1476; d) J. H.
Kim, Y. S. Chun, H. Shin, S.-g. Lee, Synthesis 2012, 44, 1809–1817; e) Y. S.
Chun, J. H. Kim, S. Y. Choi, Y. O. Ko, S.-g. Lee, Org. Lett. 2012, 14, 6358–
6361; f) Y. S. Chun, Z. Xuan, J. H. Kim, S.-g. Lee, Org. Lett. 2013, 15, 3162–
3165; g) M.-N. Zhao, H. Liang, Z.-H. Ren, Z.-H. Guan, Adv. Synth. Catal.
2013, 355, 221–226.
Experimental Section
[7] Y. He, X. Zhang, X. Fan, Chem. Commun. 2014, 50, 5641–5643.
[8] E. Babaoglu, K. Harms, G. Hilt, Synlett 2016, 27, 1820–1823.
[9] a) S. Pei, C. Xue, L. Hai, Y. Wu, RSC Adv. 2014, 4, 38055–38058; b) L. Tang,
D. Zhang-Negrerie, Y. Du, K. Zhao, Synthesis 2014, 46, 1621–1629.
[10] For selected examples, see: a) Z. Li, U. K. Sharma, Z. Liu, N. Sharma, J. N.
Harvey, E. V. Van der Eycken, Eur. J. Org. Chem. 2015, 3957–3962; b) P. R.
Krishna, E. R. Sekhar, Adv. Synth. Catal. 2008, 350, 2871–2876. For reviews,
see: c) F. Fringuelli, O. Piermatti, F. Pizzo, L. Vaccaro, Curr. Org. Chem.
2003, 7, 1661–1689; d) B. C. Ranu, Eur. J. Org. Chem. 2000, 2347–2356.
[11] For recent reviews, see: a) Y.-S. Yang, S.-W. Wang, Y. Long, Curr. Org. Chem.
2016, 20, 2865–2880; b) Z.-L. Shen, S.-Y. Wang, Y.-K. Chok, Y.-H. Xu, T.-P.
Loh, Chem. Rev. 2013, 113, 271–401; c) U. Schneider, S. Kobayashi, Acc.
Chem. Res. 2012, 45, 1331–1344. For recent applications, see: d) Y. H.
Chen, P. Knochel, Angew. Chem. Int. Ed. 2008, 47, 7648–7651; Angew.
Chem. 2008, 120, 7760; e) Y. H. Chen, M. Sun, P. Knochel, Angew. Chem.
Int. Ed. 2009, 48, 2236–2229; Angew. Chem. 2009, 121, 2270.
[12] K. Inoue, A. Sawada, I. Shibata, A. Baba, J. Am. Chem. Soc. 2002, 124,
906–907.
CCDC 1555178 (for 5a), 1555179 (for Z-7a) contain the supplemen-
tary crystallographic data for this paper. These data can be obtained
free of charge from The Cambridge Crystallographic Data Centre.
Supporting Information (see footnote on the first page of this
article): Experimental procedures for the synthesis of starting mate-
rials, reaction protocols, analytical data and NMR spectra of previ-
ously unknown compounds.
Keywords: Blaise reaction · Synthetic methods · Enamines ·
Indium · Lewis acids
[1] a) J. J. Li, Name reactions: a collection of detailed mechanisms and syn-
thetic applications, 3rd expanded ed., Springer, 2006, pp. 59–60; b) L.
Kürti, B. Czakó, Strategic applications of named reactions in organic syn-
thesis, 1st ed., Elsevier, 2005, p. 374; c) J. H. Kim, Y. O. Ko, J. Bouffard, S.-
g. Lee, Chem. Soc. Rev. 2015, 44, 2489–2507; d) H. S. P. Rao, S. Rafi, K.
Padmavathy, Tetrahedron 2008, 64, 8037–8043; e) H. S. P. Rao, N. Mut-
Received: June 19, 2017
Eur. J. Org. Chem. 2017, 4543–4547
www.eurjoc.org
4547
© 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim