Indium-catalyzed 2-alkenylation of 1,3-dicarbonyl compounds with unactivated alkynes

Article abstract of DOI:10.1021/ja0702014

1,3-Dicarbonyl compounds add to unactivated alkynes in the presence of a catalytic amount of indium(lll) trifluoromethanesulfonate in high to excellent yield to give 2-alkenylated 1,3-dicarbonyl compounds with exclusive regioselectivity as to the position of C-C bond formation on the acetylene moiety. In most of the cases, the reaction requires less than 1-mol % loading of the catalyst and does not require solvent. The reaction tolerates a wide variety of functional groups including ester, ether, allylic halide, furan, thiophene, and protected amine. Experimental and theoretical studies suggested that the reaction proceeds via a concerted carbometalation reaction of an indium(lll) enolate with the acetylene, where indium - acetylene interaction is important.

Full text of DOI:10.1021/ja0702014

Published on Web 03/28/2007
Indium-Catalyzed 2-Alkenylation of 1,3-Dicarbonyl
Compounds with Unactivated Alkynes
Kohei Endo, Takuji Hatakeyama,† Masaharu Nakamura,*,† and Eiichi Nakamura*,‡
Contribution from the Department of Chemistry, The UniVersity of Tokyo, Hongo, Bunkyo-ku,
Tokyo, 113-0033, Japan
Received January 10, 2007; E-mail: nakamura@chem.s.u-tokyo.ac.jp
Abstract: 1,3-Dicarbonyl compounds add to unactivated alkynes in the presence of a catalytic amount of
indium(III) trifluoromethanesulfonate in high to excellent yield to give 2-alkenylated 1,3-dicarbonyl compounds
with exclusive regioselectivity as to the position of C-C bond formation on the acetylene moiety. In most
of the cases, the reaction requires less than 1-mol % loading of the catalyst and does not require solvent.
The reaction tolerates a wide variety of functional groups including ester, ether, allylic halide, furan, thiophene,
and protected amine. Experimental and theoretical studies suggested that the reaction proceeds via a
concerted carbometalation reaction of an indium(III) enolate with the acetylene, where indium-acetylene
interaction is important.
Introduction
the aza-allyic π-orbital of the enolate and also stabilizes the
carbanionic product through formation of a chelate organozinc
Aldol addition to a carbonyl compound, Michael addition to
an unsaturated carbonyl compound, and alkylation with an alkyl
halide are the three major categories of C-C bond formation
reactions of metal enolates. What has remained unexplored1 is
the addition of a metal enolate to an unactivated olefin and
acetylene.2 It has been widely considered that an enolate anion
is entirely unreactive to an unactivated alkene, because of well
established thermodynamic and kinetic reasons: Among those
issues, major concerns are the unfavorable thermodynamics of
the formation of a carbanion from a stable enolate anion and
the high-lying LUMO of an unactivated alkene. We reported
sometime ago that zinc(II) enolates or enamides add to unac-
tivated olefins in high yield with remarkable facility.3 The use
of the soft zinc countercation expedites the addition through
favorable interaction between the π-orbital of the olefin and
compound.3c While our effort to utilize the zinc chemistry for
catalytic addition to alkenes has so far met with little success,
we found that zinc and indium enolates do add to an unactivated
alkyne in a catalytic manner.4 We report herein an indium
catalyzed addition of a 1,3-dicarbonyl compound to a terminal
alkyne that provides us with an efficient synthetic route to
2-alkenyl-1,3-dicarbonyl compounds (eq 1).5,6 The reaction
possesses several synthetically attractive features: (1) simple
procedure allowing large scale reactions, (2) high catalytic
efficiency, (3) high yield, (4) high regioselectivity, (5) no
requirement of solvent, (6) good chemoselectivity, and (7) high
atom economy. These features make the new reaction particu-
larly attractive among a few other related reactions of similar
nature reported recently (e.g., catalysis using rhenium,5a pal-
ladium,7 gold,8 and potassium9).
† Present address: International Research Center for Elements Science,
Institute for Chemical Research, Kyoto University, Uji, Kyoto 611-0011,
Japan.
‡ ERATO, Japan Science and Technology Agency.
(1) To alkenes: (a) Rodriguez, A. -L.; Bunlaksananusorn, T.; Knochel, P. Org.
Lett. 2000, 2, 3285-3287. To alkynes: (b) Koradin, C.; Rodriguez, A.
-L.; Knochel, P. Synlett 2000, 1452-1454. (c) Bertrand, M. T.; Courtois,
G.; Miginiac, L. Tetrahedron Lett. 1975, 36, 3147-3150. (d) Bertrand,
M. T.; Courtois, G.; Miginiac, L. Tetrahedron Lett. 1974, 23, 1945-1948.
(e) Schulte, K. E.; Rucker, G.; Feldkamp, J. Chem. Ber. 1972, 105, 24-
33.
(4) Preliminary report: Nakamura, M.; Endo, K.; Nakamura, E. J. Am. Chem.
Soc. 2003, 125, 13002-13003.
(2) Reviews: (a) Fallis, A. G.; Forgione, P. Tetrahedron 2001, 57, 5899-
5913. (b) Asao, N.; Yamamoto, Y. Bull. Chem. Soc. Jpn. 2000, 73, 1071-
1087. (c) Marek, I.; Normant, J. F. In Metal-catalyzed Cross-coupling
Reactions; Diederich, F., Stang, P. J., Eds.; WILEY-VCH: Weinheim, 1998;
pp 271-337.
(3) (a) Nakamura, M.; Hatakeyama, T.; Hara, K.; Fukudome, H.; Nakamura,
E. J. Am. Chem. Soc. 2004, 126, 14344-14345. (b) Nakamura, M.;
Hatakeyama, T.; Nakamura, E. J. Am. Chem. Soc. 2004, 126, 11820-11825.
(c) Nakamura, M.; Hatakeyama, T.; Hara, K.; Nakamura, E. J. Am. Chem.
Soc. 2003, 125, 6362-6363. (d) Nakamura, M.; Hara, K.; Sakata, G.;
Nakamura, E. Org, Lett. 1999, 1, 1505-1507. (e) Nakamura, E.; Sakata,
G.; Kubota, K. Tetrahedron Lett. 1998, 39, 2157-2158. (f) Nakamura, E.;
Kubota, K.; Sakata, G. J. Am. Chem. Soc. 1997, 119, 5457-5458. (g)
Kubota, K.; Nakamura, E. Angew. Chem., Int. Ed. Engl. 1997, 36, 2491-
2493. (h) Nakamura, E.; Kubota, K. Tetrahedron Lett. 1997, 38, 7099-
7102.
(5) (a) Kuninobu, Y.; Kawata, A.; Takai, K. Org. Lett. 2005, 7, 4823-4825.
(b) Hirase, K.; Iwahama, T.; Sakaguchi, S.; Ishii, Y. J. Org. Chem. 2002,
67, 970-973. (c) Arisawa, M.; Akamatsu, K.; Yamaguchi, M. Org. Lett.
2001, 3, 789-790. (d) Badanyan, Sh. O.; Chobanyan, Zh. A.; Tirakyan,
M. R.; Danielyan, A. O. Russ. J. Org. Chem. 1997, 33, 17-20 and
references cited therein.
(6) The alkenyl halide reagents are generally inert to an SN2 substitution
reaction. For the coupling reaction of alkenyl halide reagents and metal
enolates, see: (a) Pei, L.; Qian, W. Synlett 2006, 11, 1719-1723. (b)
Chieffi, A.; Kamikawa, K.; Ahman, J.; Fox, J. M.; Buchwald, S. L. Org.
Lett. 2001, 3, 1897-1900. Recently, vinylic substitution reaction via 1,4-
addition was reported; see: (c) Poulsen, T. B.; Bernardi, L.; Bell, M.;
Jørgensene, K. A. Angew. Chem., Int. Ed. 2006, 45, 6551-6554.
(7) Wang, X.; Widenhoefer, R. A. Chem. Commun. 2004, 660-661.
(8) Yao, X.; Li, C.-J. J. Am. Chem. Soc. 2004, 126, 6884-6885.
9
5264
J. AM. CHEM. SOC. 2007, 129, 5264-5271
10.1021/ja0702014 CCC: $37.00 © 2007 American Chemical Society

Indium-Catalyzed 2-Alkenylation of 1,3-Dicarbonyl Compds
A R T I C L E S
Table 1. Counteranion Effect on Catalytic Activity of Zinc
Table 2. Counteranion Effect of Indium(III) Salt in the Reaction
between 4 and 2a
Catalystsa
entry
catalyst (20 mol %)
% yieldb,c
entry
InX3 (5 mol %)
additive
time (h)
% yieldb
1
2
3
4
Et2Zn
27
3
1
2
3
4
5
6
InCl3
InCl3
no reaction
80
6
38
100
100
ZnCl2/Et3N
ZnI2/Et3N
Zn(OTf)2/Et3N
AgOTf
AgOTf
AgSbF6
12
12
12
12
6
10
76d
InCl3
In(OTf)3
In(NTf2)3
a The reaction of acetylacetone 1 and phenylacetylene 2 (5.0 equiv) in
the presence of catalyst was carried out at 100 °C for 16 h in a sealed tube.
bThe yield was determined by 1H NMR using bromoform as an internal
a The reaction (1.2 equiv of 2) was carried out in the presence of 5 mol
% InX3 and 15 mol % additive in a sealed tube at 40 °C for the time
indicated. bThe yield was determined by 1H NMR using bromoform as an
internal standard.
c
d
standard. The regioselectivity was >99:1. Isolated yield.
Results and Discussion
Among triflates of Cu(I), Cu(II), Mg(II), Zn(II), Hg(II), Sc(III),
La(III), Nd(III), Dy(III), Ho(III), Yb(III), Lu(III), Al(III), Ga(III),
In(III), and others, only Hg(OTf)2, Ga(OTf)3, and In(OTf)3 gave
the desired product 3 in 3%, 14%, and 100% yields, respectively.
The reaction in the presence of Hg(OTf)2 gave a small amount
of 5 accompanied by rapid decomposition of phenylacetylene.12
Many of the corresponding metal chlorides were entirely
ineffective. It is interesting to note that In(OTf)3 did not require
any base while Zn(OTf)2 was ineffective in the absence of amine
such as Et3N or 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU).13
TfOH (0.5 mol %) did not catalyze the addition at all.
In analogy with the stoichiometric reaction of zinc enamides
with alkenes, we initially examined a zinc(II)-catalyzed reaction
of acetylacetone 1 and phenylacetylene 2. For instance, the
reaction starting from 0.6 mmol of acetylacetone 1 was catalyzed
by 20 mol % of Zn(acac)2 to give 0.72 mmol of 3 (eq 2). The
Counterion effect of the indium catalysis was examined in
some details (Table 2). The reaction between 4 and 2 in the
presence of an indium salt and an additive at 40 °C showed an
interesting trend. InCl3 (5 mol %) was entirely ineffective (entry
1)14 but could be activated by addition of 15 mol % of AgOTf
(entry 2). The mixture of InCl3 and AgOTf produced a
precipitate of AgCl and probably produced an indium triflate.
AgOTf itself was far less active (entry 3). A mixture of 5 mol
% of InCl3 and 15 mol % of AgSbF6 that may produce an
indium hexafluoroantimonate showed moderate catalytic activity
(entry 4). In(OTf)3 was a very good catalyst (entry 5), and
indium(III) tris[bis(trifluoromethanesulfonyl)amide], In(NTf2)3,
was even better (entry 6). The latter reagent however is more
costly than the former and is not a preferred catalyst in the cases
where the former is effective.
product stoichiometry indicated that the zinc acetylacetonate
catalyst took part in the addition reaction. Optimization of the
reaction was carried out briefly as summarized in Table 1. The
reaction in the presence of 20 mol % of diethylzinc gave the
desired product 3 in 27% yield (entry 1). Among several zinc(II)
salts, zinc(II) trifluoromethanesulfonate, Zn(OTf)2, combined
with triethylamine (Et3N)10 showed the highest catalytic activity
to furnish 3 in 76% yield (entry 4). Zn(OTf)2 was manipulated
quickly under air atmosphere and dried in the reaction vessel.
The Zn(OTf)2/Et3N-catalyzed reaction was found to be
powerful enough to create a quaternary carbon center by the
reaction with less reactive â-ketoesters. Thus, the reaction of
ethyl 2-methyl-3-oxobutanoate 4 catalyzed by 20 mol %
Zn(OTf)2/Et3N at 100 °C proceeded smoothly to generate an
R-alkenylated diketone 5 (eq 3).11
The In(OTf)3-catalyzed reaction was found to be facile and
versatile, taking place with as little as 0.05 mol % of the catalyst
with no necessity of solvent. Table 3 summarizes the results of
the first set of experiments carried out on a combination of ethyl
2-methyl-3-oxobutanoate 4 and functionalized phenylacetylene
derivatives (carried out on a 0.2-4.0-mmol scale). The regio-
selectivity of the reaction was always such that C-C bond
formation takes place next to the benzene ring. The product 5
was obtained in 99% isolated yield upon simple heating of near
stoichiometric amounts of 4 and 2 in the presence of 0.05 mol
% In(OTf)3 at 140 °C for 90 min (entry 1). The reaction was
After further experimentation, however, we felt it necessary
to find metal countercations more powerful than zinc and
examined the reaction of ethyl 2-methyl-3-oxobutanoate 4 and
phenylacetylene 2 (1.2 equiv), which was carried out in the
presence of 0.5 mol % of a metal triflate at 100 °C for 3 h.
(9) Rodriguez, A. L.; Bunlallsananusorn, T.; Knochel, P. Org. Lett. 2000, 2,
3285-3287.
(12) (a) Nishizawa, M.; Yadav, V. K.; Skwarczynski, M.; Takao, H.; Imagawa,
H.; Sugihara, T. Org. Lett. 2003, 5, 1609-1611. (b) Nishizawa, M.; Takao,
H.; Imagawa, H.; Sugihara, T. Org. Lett. 2003, 5, 4563-4565. (c) Imagawa,
H.; Kurisaki, T.; Nishizawa, M. Org. Lett. 2004, 6, 3679-3681. (d)
Imagawa, H.; Iyenaga, T.; Nishizawa, M. Org. Lett. 2005, 7, 451-453.
(13) Zn(OTf)2 and Al(OTf)3 showed the catalytic activity in the presence of 1
equiv of amine base, such as DBU, to metal triflate to give the
corresponding product in 7% and 5%, respectively. This is due to the
assistance of the deprotonation reaction to generate the metal enolate
intermediate.
(10) (a) Frantz, D. E.; Fassler, R.; Carreira, E. M. J. Am. Chem. Soc. 1999,
121, 11245-11246. (b) Frantz, D. E.; Fassler, R.; Tomooka, C. S.; Carreira,
E. M. Acc. Chem. Res. 2000, 33, 373-381. (c) Frantz, D. E.; Fassler, R.;
Carreira, E. M. J. Am. Chem. Soc. 2000, 122, 1806-1807. (d) Anad, N.
K.; Carreira, E. M. J. Am. Chem. Soc. 2001, 123, 9687-9688. (e) Boyall,
D.; Frantz, D. E.; Carreira, E. M. Org. Lett. 2002, 4, 2605-2606.
(11) The reaction of ethyl 3-oxobutanoate and phenylacetylene in the presence
of catalytic or stoichiometric amounts of Zn(OTf)2 and Et3N gave a mixture
of products as an R,â-unsaturated form and a â,γ-unsaturated form; see;
Nakamura, M.; Fujimoto, T.; Endo, K.; Nakamura, E. Org. Lett. 2004, 6,
4837-4840.
(14) The high reaction temperature is required for the reaction in the presence
of InCl3; see: Zhang, J.; Blazecka, P. G.; Angell, P.; Lovdahl, M.; Curran,
T. T. Tetrahedron 61, 7807-7813.
9
J. AM. CHEM. SOC. VOL. 129, NO. 16, 2007 5265

A R T I C L E S
Endo et al.
Table 3. Indium-Catalyzed Addition of 4 to Aromatic Alkynesa
methoxycarbonylbenzene 12 provided the product 13 in 97%
yield in the presence of 0.05 mol % catalyst at 140 °C (entry
5). Because of the thermal instability of 1-ethynyl-2-methoxy-
carbonylbenzene 14, the reaction of 14 needed to be finished
quickly under higher catalyst loading and at a lower reaction
temperature (entry 6).
Intramolecular competition between the acetylene and the
acrylic ester moiety in 16 resulted in an exclusive reaction on
the acetylene to give the corresponding alkenylated product 17
in 85% yield (entry 7). This lack of reactivity toward the Michael
accepter moiety indicates that the present addition reaction is
not a simple nucleophilic addition reaction but rather an
electrophilic reaction involving coordination of the indium metal
to the acetylene π-bond.
The addition to 1-ethynyl-4-methoxybenzene 18 in toluene
afforded the product 19 in 98% yield (entry 8). The reaction
required a solvent (toluene), without which the reaction afforded
a large amount of what appears to be the products of an
intermolecular Friedel-Crafts reaction including the aromatic
ring and the acetylene moiety.15 The reaction of 1-ethynyl-2-
methoxybenzene 20 was not complicated by this side reaction
and afforded, without the use of the solvent, the product 21 in
97% yield (entry 9). A dimethylamino group entirely inhibited
the reaction (entry 10), probably because of the coordination
of the amino moiety to the In(III) atom.
Table 4 illustrates further examples of alkynes other than
phenylacetylene derivatives, including functionalized alkynes,
an enyne and a diyne. 1-Octyne 23 was less reactive than
phenylacetylene but still gave the desired alkenylated product
24 in 99% isolated yield with the regioselectivity being the same
as the one observed for phenylacetylenes (entry 1). An ether
moiety can be tolerated as shown by the synthesis of 26 in 94%
yield (entry 2). The reaction of a conjugated enyne 27 took place
exclusively on the acetylene moiety in the presence of 2 mol
% In(OTf)3 at 60 °C to afford the diene 28 in 90% yield (entry
3). An amino group can be tolerated if it is protected for instance
as a phthalimide as illustrated by the reaction of 29 that gave
the desired product 30 in 90% yield (entry 4). Benzyl propargyl
ether 31 was so labile under the reaction conditions that the
product was not obtained at all (entry 5). However, the addition
of 5 mol % Et3N furnished the desired product 32 in 80% yield
(entry 6).
Furan and thiophene are also acid sensitive and caused a
problem as illustrated for ethynylfuran 33 (entry 7). The use of
1 mol % of DBU, however, improved the addition to give the
desired product 34 in 79% yield (entry 8). Ethynylthiophene
35 gave the corresponding product 36 in 85% yield in the
presence of In(OTf)3 (entry 9), but the yield improved to 99%
by the use of DBU (entry 10). 1,3-Decadiyne 37 was a rather
sensitive substrate and gave the product 38 in 50% yield (entry
11), and the yield could not be improved by DBU (13% yield
as in entry 12). In both reactions, the C-C bond formation took
place with perfect selectivity on the internal carbon atom of
the terminal acetylene moiety.
a
The reaction was carried out in the presence of In(OTf)3 in a sealed
tube. No solvent was used except in entry 8, where toluene was used.
bIsolated yield. The regioselectivity was >99:1.
c
scaled up with equal ease to a 0.1-mol scale and afforded the
product in 95% yield after vacuum distillation of the reaction
mixture.
An electron-deficient acetylene, 1-ethynyl-4-trifluorometh-
ylbenzene 6, reacted more slowly and required 24 h for
completion but still gave the desired product 7 in 92% yield
(entry 2). 1-Bromo-4-ethynylbenzene 8 and 1-bromo-2-ethyn-
ylbenzene 10 participated in the reaction to give the corre-
sponding product, and they are also slow reacting substrates:
The reaction of the alkyne 8 required 20 h for completion in
the presence of 0.05 mol % catalyst to give the product 9 in
97% yield (entry 3). The alkyne 10 was even less reactive
requiring 0.5 mol % catalyst and 60 h to afford the product 11
in 96% yield (entry 4). We assume that the steric effect of the
o-bromo group in 11 retarded the reaction.
The vinylation reaction of ketoester 39 under an atmospheric
pressure of acetylene gas 40 was not an easy reaction at first.
Commercially available purified acetylene gas in the presence
(15) (a) Tsuchimoto, T.; Maeda, T.; Shirakawa, E.; Kawakami, Y. Chem.
Commun. 2000, 1573-1574. (b) Tsuchimoto, T.; Matsubayashi, H.;
Kaneko, M.; Shirakawa, E.; Kawakami, Y. Angew. Chem., Int. Ed. 2005,
44, 1336-1340.
Likewise, ester-bearing compounds 12 and 14 are not very
reactive but still high-yielding. The reaction of 1-ethynyl-4-
9
5266 J. AM. CHEM. SOC. VOL. 129, NO. 16, 2007

Indium-Catalyzed 2-Alkenylation of 1,3-Dicarbonyl Compds
A R T I C L E S
Table 4. Indium-Catalyzed Addition of Ketoesters 4 and 39 to a
cheaper and safer than the purified one but contains water,
acetone, etc., did not afford the product 41 even in the presence
of DBU additive. Ester hydrolysis and decarboxylation posed
problems (entry 15). We found molecular sieves much improved
the reaction that then gave the desired product 41 in 95% yield
(entry 16). The catalyst loading can be reduced to 5 mol % at
the expense of the reaction rate. The details of the investigation
have been described elsewhere.16
Variety of Alkynesa
The In(OTf)3/DBU catalytic system was inefficient for the
reaction of 1,3-diynes, but In(NTf2)3 was found to be effective.
The reaction of ethyl 2-methyl-3-oxobutanoate 4 and 1,3-
decadiyne 37 (1.05 equiv) in the presence of 1 mol % In(NTf2)3
at 60 °C for 48 h furnished the corresponding product 38 in
98% yield with perfect regioselectivity (eq 4). The resultant
conjugated enyne 38 can be used for the Pd(0)-catalyzed [4 +
2] benzannulation17 to obtain the densely functionalized 2-arylke-
tone 42 in 96% yield regioselectively in a single reaction vessel
without purification. The details of this tandem reaction have
been described elsewhere.18
The addition to a silylacetylene gave us information on the
stereochemistry of the reaction. The addition of 1 to ethy-
nyldimethylphenylsilane 43 in the presence of In(OTf)3 (5 mol
%) and DBU (6 mol %) gave the product 45 in 94% yield that
bears a trans-olefinic double bond (eq 5).19 The trans stereo-
a The reaction was carried out in the presence of In(OTf)3 and additives
in a sealed tube. Ketoester 4 was used in entries 1-12. Ketoester 39 was
used in entries 13-16. bIsolated yield. cThe regioselectivity was >99:1.
dNPht ) phthalamide. eA trace amount of a regioisomeric addition product
was detected by 1H NMR analysis. fThe reaction that took place exclusively
at the position indicated by an arrow was conducted in the dark because of
the sensitivity of the product to light. gThe commercially available dry
acetylene gas was used. hThe commercially available wet acetylene gas for
chemistry was deduced from the 1H NMR coupling constant of
3Jab 19.6 Hz.20 Assuming that the geminal-indium/silane bime-
tallic intermediate such as 44 is configurationally stable, we
i
welding use was used. The reaction vessel was connected to a balloon
j
filled with acetylene gas. MS 3A was activated at 200 °C under 5 Pa for
4 h prior to use.
(16) Nakamura, M.; Endo, K.; Nakamura, E. Org. Lett. 2005, 7, 3279-3281.
(17) (a) Gevorgyan, V.; Takeda, A.; Homma, M.; Sadayori, N.; Radhakrishnan,
U.; Yamamoto, Y. J. Am. Chem. Soc. 1999, 121, 6391-6402. (b)
Gevorgyan, V.; Takeda, A.; Yamamoto, Y. J. Am. Chem. Soc. 1997, 119,
11313-11314.
of 20 mol % In(OTf)3 provided the desired product 41 in 25%
yield and uncharacterized oligomeric products (entry 13). The
combined catalyst system, 20 mol % In(OTf)3/DBU, was
effective when the purified acetylene gas 40 was used and gave
the desired vinylated product 41 in 94% (entry 14). On the
contrary, the welding grade acetylene gas 40, which is much
(18) Nakamura, M.; Endo, K.; Nakamura, E. AdV. Synth. Catal. 2005, 347,
1681-1686.
(19) DBU is essential. The reaction without DBU promotes the desilylation
reaction to give a vinylated product.
9
J. AM. CHEM. SOC. VOL. 129, NO. 16, 2007 5267

A R T I C L E S
Endo et al.
could conclude that the addition took place in a cis manner.
The observed regioselectivity of addition is then typical of a
carbometalation reaction.21
as a pronucleophile. The reaction of 2,4-pentanedione 1 and
phenylacetylene 2 proceeded in the presence of 5 mol % n-BuLi
and Et3N as additives to give the product 3 in excellent yield
in its enol form. The use of n-BuLi and Et3N suppressed
byproduct formation.25 The reaction of 3-methyl-2,4-pentanedi-
one 77 gave the corresponding products 78 and 79 under the
same reaction conditions in 88% and 77% yields, respectively
(entries 28 and 29). The reaction without n-BuLi produced more
byproducts and afforded the product only in 65% yield (entry
27). Similarly, 2-acetylcyclopentanone 80 and 2-acetylcyclo-
hexanone 82 reacted with phenylacetylene 2 in the presence of
n-BuLi to give the products 81 and 83 in high to excellent yields
(entries 30 and 31). Although the exact roles of n-BuLi and
Et3N are unclear, the results indicate the necessity of an added
base in the 1,3-diketone reactions. We might speculate that
indium-catalyzed addition of the terminal alkyne to the carbonyl
group has caused the byproduct formation.26
Table 5 illustrates the diversity of 1,3-dicarbonyl compounds
to be used for the reaction. The allyl ester 46 gave the desired
vinylated compound (entries 1, 2, and 3), which can be used
then for a decarboxylative R-allylation reaction.22 The reaction
of 2-benzyl substituted substrate 50 gave the product in good
yield (entries 4, 5, and 6). The product from the 2-allyl
substituted substrate 54 was too labile under the In(OTf)3
conditions, but the presence of added Et3N afforded the desired
products 55 and 56 in high yield (entries 7, 8, and 9). The
reaction of ethyl cyclopentanone-2-carboxylate 57 and phenyl-
acetylene 2 in the presence of 1 mol % catalyst at 100 °C
afforded the corresponding product 58 in 94% yield, but lower
catalyst loading (0.05 mol %) and higher temperature (140 °C)
lowered the yield to 72% (entries 10 and 11). 1-Octyne 23 gave
a comparable result (entry 12). Ethyl cyclohexanone-2-carboxyl-
ate 60 and phenylacetylene 2 reacted in the presence of 2 mol
% catalyst at 100 °C to give the corresponding adduct 61 in
83% yield (entry 13). The more stable enolate of benzoylacetate
62 and a tetralone derivative 66 gave the products 63-67 in
excellent yields at the expense of reaction rate (entries 14-
17). Overall, cyclic dicarbonyl compounds are less reactive than
open chain compounds such as 4.
Ethyl 2-fluoro-3-oxobutanoate 68 and ethyl 2-chloro-3-
oxobutanoate 71 also took part in the addition reaction to
phenylacetylene 2 and 1-octyne 23 in moderate to high yield
under the In(OTf)3 or In(OTf)3/Et3N conditions (entries 18-
23). We consider it noteworthy that the products bear a fluorine
or a chlorine atom on a quaternary carbon center flanked by
two carbonyl groups and one alkenyl group. These compounds
are interesting for their chemical reactivities23 and their potential
biological activity.24
The experimental observations described above gave us some
insights into the mechanism of the reaction, and we suggest
the catalytic cycle shown in Figure 1. Thus, the reaction begins
with the generation of an indium enolate by the reaction of the
dicarbonyl compound with In(OTf)3. This process produces
TfOH, which is trapped by an alkyne to give a neutral alkenyl
triflate. The addition reaction of the indium(III) enolate to the
alkyne takes place via a concerted carbometalation pathway to
afford an alkenyl indium(III) intermediate. Protonation of the
alkenyl indium(III) intermediate by the starting dicarbonyl
compound completes the catalytic cycle.
The first piece of mechanistic evidence is the following: The
reaction requires a base that removes the acidic methylene or
methyne proton in the starting 1,3-dicarbonyl compound. In
some cases, Et3N and n-BuLi deprotonate the substrate, and
we found, in the phenylacetylene reaction, that the alkyne can
accept the proton through formation of an acetylene-TfOH
adduct, styryl triflate 85, (Scheme 1). Thus, the reaction between
4 and 2 in the presence of In(OTf)3 (20 mol %) at ambient
temperature gave the adduct 5 in 98% yield and acetophenone
86 in 17% yield (85% based on one indium atom). It is known
that trifluoromethanesulfonic acid reacts with phenylacetylene
2 to give styryl triflate 85 and that the latter hydrolytically
decomposes to give acetophenone.27 The 19F NMR spectrum
of the reaction mixture showed signals at 88.5, 87.3, and 85.3
ppm (C6F6 ) 0 ppm as an internal standard in C6D6, 1:trace:9
integration ratio of signals), of which the 88.5 ppm signal is
the one expected for 85 as compared with an authentic sample.
The remaining two signals may well be due to some indium(III)
enolate intermediates. We therefore surmise that an In(OTf)2
enolate is a reactive species.
Diethyl methylmalonate 74 also added to alkynes in excellent
yield (entries 24 and 25), although â-nitroketone, cyanoketone,
cyanoester, phosphonylester, and sulfonylester were rather
unreactive. 1,3-Diketones participate in the addition reaction
(20) The geometry of olefinic double bond was confirmed as trans by NOE
experiments and the value of the coupling constant between Ha and
Hb.
The second important observation is that alkyne is more reac-
tive than an acrylic ester as revealed by the competitive experi-
ment in entry 7, Table 3. This is a clear indication that the
reaction is not a simple nucleophilic 1,2-addition but the reaction
involves a metal-alkyne interaction as an important step. The
(21) The theoretical study of the regioselectivity of carbometalation reaction
was reported: Nakamura, E.; Miyachi, Y.; Koga N.; Morokuma, K. J. Am.
Chem. Soc. 1992, 114, 6686-6692.
(22) Nakamura, M.; Hajra, A.; Endo, K.; Nakamura, E. Angew. Chem., Int. Ed.
2005, 44, 7248-7251.
(23) For the cross coupling reaction of an R-chloroketone and an organozinc
reagent, see: (a) Malosh, C. F.; Ready, J. M. J. Am. Chem. Soc. 2004,
126, 10240-10241. For the Nozaki-Hiyama allylation using an allylhalide
reagent, see: (b) Xia, G.; Yamamoto, H. J. Am. Chem. Soc. 2006, 128,
2554-2555.
(24) (a) Biomedical Frontiers of Fluorine Chemistry; Ojima, I., McCarthy, J.
R., Welch, J. T., Eds.; ACS Symposium Series 639; American Chemical
Society: Washington, DC, 1996. (b) Asymmmetric Fluoroorganic Chem-
istry; Synthesis, Application and Future Directions; Ramachandran, P. V.,
Ed.; ACS Symposium Series 746; American Chemical Society: Washing-
ton, DC, 2000. For synthetic chemistry of R-fluorocarbonyl compounds,
see: (c) Rozen, S.; Filler, R. Tetrahedron 1985, 41, 1111-1153.
(25) The addition of n-BuLi decreased the generation of the major byproduct,
3,3-diphenylethynyl-2,4-pentandione. The addition of Et3N decreased the
generation of uncharacterized complex byproducts.
(26) (a) Takita, R.; Yakura, K.; Ohshima, T.; Shibasaki, M. J. Am. Chem. Soc.
2005, 127, 13760-13761. (b) Takita, R.; Fukuta, Y.; Tsuji, R.; Ohshima,
T.; Shibasaki, M. Org. Lett. 2005, 7, 1363-1366.
(27) (a) Stang, P. J.; Summerville, R. H. J. Am. Chem. Soc. 1969, 91, 4600-
4601. (b) Summerville, R. H.; Senkler, C. A.; Schleyer, P. v. R.; Dueber,
T. E.; Stang, P. J. J. Am. Chem. Soc. 1974, 96, 1100-1110.
9
5268 J. AM. CHEM. SOC. VOL. 129, NO. 16, 2007

Indium-Catalyzed 2-Alkenylation of 1,3-Dicarbonyl Compds
A R T I C L E S
Table 5. Indium-Catalyzed Addition of Various 1,3-Dicarbonyl Compounds to Alkynesa
b
c
a The reaction was carried out in the presence of In(OTf)3 and additives in a sealed tube. Isolated yield. The regioselectivity was >99:1.
effectiveness of soft Lewis acids, indium(III) and zinc(II),
supports this conjecture. Third, the addition reaction to a
silylacetylene exhibits stereo- and regioselectivities typical for
1,2-carbometalation reactions (eq 5). The cis-selectivity was also
proved through deuterium labeling experiments as described in
the next paragraph.
The deuterium-labeled experiment was examined first as
shown in eq 6. The reaction of ethyl 2-methyl-3-oxobutanoate
9
J. AM. CHEM. SOC. VOL. 129, NO. 16, 2007 5269

A R T I C L E S
Endo et al.
used a stoichiometric amount of In(OTf)3 and Et3N and obtained
the product 67-D with 83% deuterium incorporation at trans-
position (Hb) together with 23% deuterium at the cis-position
(Ha). We therefore conclude that the cis-carbometalation is likely
the key step of the reaction.
The final evidence has come from theoretical analysis. We
carried out density functional calculations to examine if it is
indeed possible to add an indium enolate to acetylene. Thus,
we obtained some transition state (TS) structures for a few
simple model systems. These calculations afforded a consensus
structure (Figure 3A) that lies on a reaction coordinate between
the starting π-complex and the vinyl indium product (as shown
in Figure 2). The forming C-In bond and the forming C-C
bond in TS are almost the same (2.356 Å and 2.373 Å,
respectively). Judging by the final lengths of the corresponding
bonds in the vinyl indium product (2.144 Å and 1.555 Å,
respectively, Figure 3), the formation of the C-In bond is
apparently much more advanced than that of the C-C bond.
Different courses of the rehybridization of two acetylene carbons
also demonstrate the advanced nature of the C-In bond
formation over that of the C-C bond (the H-Cγ-Câ and Cγ-
Câ-H angles are 134.2° and 163.3°, respectively). Such
structural characteristics strongly suggest the importance of the
interaction between acetylene and the indium metal center in
the earlier stage of the carboindation step. One interesting feature
of this structure is that the acetylene is located just above the
center of the pseudo-C2 symmetry plane of the indium enolate
of a 1,3-dicarbonyl compound and that the two asterisked C*-
H* bonds are parallel to each other. This special geometry can
maximize the orbital interaction between the indium enolate and
acetylene and may be a part of the reasons why certain substrates
such as cyclic ketone substrates (entries 10-17, Table 5) are
less reactive than acyclic substrates.
Figure 1. A plausible mechanism of indium-catalyzed addition reaction.
Scheme 1. Generation of acetophenone 86 via deprotonation
reaction of ethyl 2-methyl-3-oxobutanoate 4
4 and 1-deuterio-2-phenylethyne 2-D (1.2 equiv) was carried
out in the presence of In(OTf)3 (20 mol %) in C6D6 at 50 °C.
Conclusion
We reported in this paper the discovery and development of
the indium-catalyzed addition reaction of active methylene
compounds to alkynes. A number of examples provided in this
article illustrate the versatility of this new synthetic reaction.
The key to success is the choice of the indium metal and its
counteranion. The reaction may require a base but quite often
does not require any apparent base because the acetylene
substrate acts as a latent base by scavenging TfOH, generating
a π-Lewis acidic indium enolate, which is suitable for the
carboindation reaction. The resulting 2-alkenyl 1,3-dicarbonyl
compounds are useful synthetic intermediates that can be used
for further synthetic transformations and for the synthesis of
pharmaceutically important compounds.
1
The reaction was monitored in situ by H NMR. The product
5-D was obtained with 20% deuterium incorporation at cis-
position (Ha) and 50% deuterium incorporation at trans-position
(Hb) in the product 5-D. A considerable degree of H-D
exchange between ethyl 2-methyl-3-oxobutanoate 4 and 1-deu-
terio-2-phenylethyne 2-D also took place.
The reaction of the tetralone derivative 66 and 2-D was also
examined (eq 7). To speed up the reaction and suppress the
H-D exchange between 66 and 2-D as much as possible, we
Experimental Section
A Representative Procedure of the Alkenylation of 1,3-Ketoesters
(4.0-mmol Scale): Ethyl 2-Methyl-2-(1-oxoethyl)-3-phenyl-3-butenoate
9
5270 J. AM. CHEM. SOC. VOL. 129, NO. 16, 2007

Indium-Catalyzed 2-Alkenylation of 1,3-Dicarbonyl Compds
A R T I C L E S
Figure 2. Potential surface of the addition of indium enolate to acetylene.
Figure 3. TS models of addition of indium enolate to acetylene. (A) A theoretical TS model of addition of In(FSO3)2 enolate of acetylacetone to acetylene
(B3LYP/6-31G* and LANL2DZ for In). Carbon, gray; hydrogen, white; oxygen, red; fluorine, blue; sulfur, yellow; indium, purple. (B) The pseudo three-
dimensional representations of structure A. The substituent numbering follows the one in eq 1. Bond lengths and angles are in angstrom and degree, respectively.
(5). A CH3CN solution of In(OTf)3 (0.1 M, 20 µL, 0.05 mol %) was
introduced into a dried reaction vessel under an argon atmosphere. Then
CH3CN was removed under reduced pressure (130 Pa) at room
temperature for 1 h. Ethyl 2-methyl-3-oxobutanoate (4.0 mmol, 576
mL, 576 mg) and phenylacetylene (4.8 mmol, 527 µL, 490 mg, 1.2
equiv) were introduced into the reaction vessel. The reaction mixture
was stirred at 140 °C for 90 min and then cooled to room temperature.
Flash column chromatography (silica gel 10 g, eluent; hexane/Et2O )
95/5) of the reaction mixture gave the title compound in 99% yield
temperature. The reaction mixture was stirred at 100 °C for 24 h and
then quenched with a pH 6.88 phosphate buffer solution. The aqueous
layer was extracted with ether (3 times, 10 mL). The combined organic
layers were dried over Na2SO4. Removal of solvent and the following
flash column chromatography (silica gel 10 g, eluent; hexane/ethyl
acetate ) 95/5) gave the alkenylated compound in 97% yield (197 mg);
1H NMR (400 MHz, CDCl3) δ 1.98 (s, 6H), 5.23 (d, J ) 1.2 Hz, 1H),
5.90 (d, J ) 1.6 Hz, 1H), 7.29-7.44 (m, 5H, arom); 13C NMR (100
MHz, CDCl3) δ 23.5 (2C), 113.8, 118.3, 125.8 (2C), 128.3, 128.7 (2C),
139.6, 143.5, 191.2 (2C); IR (cm-1) 3083, 1598, 1574, 1493, 1391,
1243, 910.
1
(973 mg); H NMR (400 MHz, CDCl3) δ 1.20 (t, J ) 7.2 Hz, 3H),
1.52 (s, 3H), 2.29 (s, 3H), 4.18 (q, J ) 7.2 Hz, 2H), 5.28 (s, 1H), 5.44
(s, 1H) 7.18-7.31 (m, 5H, arom); 13C NMR (100 MHz, CDCl3) δ 13.8,
21.8, 27.2, 61.5, 65.8, 118.7, 127.5, 127.7 (2C), 128.0 (2C), 140.3,
147.9, 171.7, 205.2; IR (cm-1) 2987, 1710, 1239, 1092, 1021, 915.
Anal. Calcd for C15H18O3: C, 73.15; H, 7.37. Found: C, 73.32; H,
7.46.
A Representative Procedure of the Alkenylation of 1,3-Diketones
(1.0-mmol scale): 4-Hydroxy-3-(1-phenylethenyl)-3-penten-2-one
(3).1e,28 To phenylacetylene (5.0 mmol, 549 µL, 511 mg, 5.0 equiv)
n-BuLi (0.025 mmol, 1.51 M in hexane, 17 µL, 2.5 mol %) was added
at 0 °C. After 30 min, In(OTf)3 (0.025 mmol, 14.0 mg, 2.5 mol %)
and Et3N (0.025 mmol, 3.5 µL, 2.5 mg, 2.5 mol %) were added. Then
2,4-pentanedione (1.0 mmol, 103 µL, 100 mg) was added at room
Acknowledgment. We thank the Ministry of Education,
Culture, Sports, Science, and Technology (MEXT) of Japan for
financial support (KAKENHI Kiban-S to E.N.). This work was
also supported partly by a Grant-in-Aid for Scientific Research
on Priority Areas “Synergistic Effects for Creation of Functional
Molecules” from MEXT, Japan.
Supporting Information Available: Experimental procedure
and characterization data. This material is available free of
charge via the Internet at http://pubs.acs.org.
(28) Dybova, T. N.; Yurchenko, O. I.; Gritsai, N. V.; Buikliskii, V. D.;
Mel’nikova, E. D. Russ. J. Org. Chem. 2002, 38, 452-453.
JA0702014
9
J. AM. CHEM. SOC. VOL. 129, NO. 16, 2007 5271