Gold(I)-catalyzed diastereoselective hydroacylation of terminal alkynes with glyoxals

Article abstract of DOI:10.1002/anie.201307685

The reaction of an α-ketoaldehyde and a terminal alkyne in the presence of piperidine and a catalytic amount of AuCl delivers 1,2-dicarbonyl-3-enes, products of the formal hydroacylation of the triple bond. The scope of the method is broad; different aryl substituents on the dicarbonyl unit and on the alkyne are well tolerated. The products can be transformed selectively into vinylquinoxalines. Mechanistic studies, including isotope-labeling experiments, indicate that after an initial A³-type conversion to propargylic amines, a subsequent base-mediated alkyne-to-allene isomerization and a hydrolysis of the enamine substructure during the workup deliver the formal hydroacylation products. The simple AuCl catalyst and piperidine convert terminal alkynes and α-ketoaldehydes into 1,2-dicarbonyl-3-enes, the products of a formal hydroacylation of the triple bond, with excellent regio- and diastereoselectivity. There is no evidence for classical A³ coupling products, which could be expected from such a gold-catalyzed reaction of an aldehyde, an amine, and a terminal alkyne. Copyright

Full text of DOI:10.1002/anie.201307685

.
Angewandte
Communications
DOI: 10.1002/anie.201307685
Homogeneous Gold Catalysis
Hot Paper
Gold(I)-Catalyzed Diastereoselective Hydroacylation of Terminal
Alkynes with Glyoxals**
Shuai Shi, Tao Wang, Vanessa Weingand, Matthias Rudolph, and A. Stephen K. Hashmi*
Abstract: The reaction of an a-ketoaldehyde and a terminal
alkyne in the presence of piperidine and a catalytic amount of
AuCl delivers 1,2-dicarbonyl-3-enes, products of the formal
hydroacylation of the triple bond. The scope of the method is
broad; different aryl substituents on the dicarbonyl unit and on
the alkyne are well tolerated. The products can be transformed
selectively into vinylquinoxalines. Mechanistic studies, includ-
ing isotope-labeling experiments, indicate that after an initial
A3-type conversion to propargylic amines, a subsequent base-
mediated alkyne-to-allene isomerization and a hydrolysis of
the enamine substructure during the workup deliver the formal
Scheme 1. Initial observation.
hydroacylation products.
ketone 5a with an exclusively E-configurated double bond
I
n the last years, transition-metal-catalyzed Mannich-type
was obtained as the major product, but only in moderate
yields (Scheme 1). Encouraged by this unusual result and the
fact that alkenyl-1,2-diketones, valuable potential building
blocks, are barely documented,[5] we decided to investigate
this unexpected transformation in detail.
three-component coupling reactions of aldehydes, amines and
3
À
terminal alkynes (A couplings) through C H bond activation
have become an established, convenient, and efficient method
for the synthesis of propargylamines.[1] Our group has recently
reported an efficient gold-catalyzed oxidation which allowed
us to make various glyoxals through the oxidation of terminal
alkynes.[2] The application of glyoxals for the construction of
organic scaffolds, especially for the generation of hetero-
cycles, is an often used strategy.[3] Since a wide range of
aldehydes can be applied in the A3 coupling, we assumed that
glyoxals might also be considered as starting materials, which
should open up new applications for this well-known building
block. A transformation into different oxopropargyl amines
might deliver useful precursors for the synthesis of highly
functionalized furans.[4]
We focused on optimizing the reaction conditions using
phenylglyoxal and phenylacetylene as starting materials.
Increasing the reaction temperature from room temperature
to 508C strongly accelerated the reaction rate (Table 1,
entry 2). Therefore all of the catalysts were screened at
elevated temperatures. Among the employed gold(III) salts,
dichloro(2-pyridinecarboxylato)gold[6] (Table 1, entry 6)
turned out to be the best catalyst while other complexes
containing chloride ligands delivered only poor results.
Interestingly, gold(I) chloride gave even better results than
the gold(III) salts, while cationic gold(I) sources (activated by
a silver(I) salt) delivered poor results (Table 1, entries 8–10).
Other transition metals showed low or no conversions
(Table 1, entries 11–13); hence we used AuCl for our further
optimizations.
As the next optimization step we performed a solvent
screening (Table 1, entries 14–20). None of the other solvents
was superior to toluene. Polar solvents (Table 1, entries 14
and 15) usually led to faster conversion but lower yields. We
further explored the effect of different bases on the reaction
(Table 1, entries 21–26). Piperidine gave the best results.
Control experiments (Table 1, entries 27 and 28) showed that
both a gold catalyst and a base are necessary for this reaction.
Attempts to lower the catalyst loadings were unsuccessful and
yields dropped significantly (Table 1, entries 29 and 30). In
addition, only minor conversion of the starting material
resulted when a substoichiometric amount of piperidine was
present (Table 1, entry 31). Furthermore, oxygen must be
excluded from the reaction system as otherwise a-ketoamides
will be produced in quantitative yields (Scheme 2).[7]
To test our hypothesis, we initially attempted to synthesize
polysubstituted furans by reacting phenylglyoxal with phenyl-
acetylene and piperidine in the presence of AuBr3. To our
surprise, only traces of the expected furan were obtained.
Instead, under our reaction conditions the alkenyl-1,2-di-
[*] M. Sc. S. Shi, M. Sc. T. Wang, V. Weingand, Dr. M. Rudolph,
Prof. Dr. A. S. K. Hashmi
Organisch-Chemisches Institut
Ruprecht-Karls-Universitꢀt Heidelberg
Im Neuenheimer Feld 270, 69120 Heidelberg (Germany)
E-mail: hashmi@hashmi.de
Homepage: http://www.hashmi.de
Prof. Dr. A. S. K. Hashmi
Chemistry Department, Faculty of Science
King Abdulaziz University
Jeddah 21589 (Saudi Arabia)
[**] S.S. and T.W. are grateful to the CSC (Chinese Scholarship Council)
for fellowships. Gold salts were generously donated by Umicore AG
& Co. KG.
Next we explored the scope of this cascade reaction by
using glyoxals 1 and terminal alkynes 2 under the optimized
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201307685.
1148
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1148 –1151

Angewandte
Chemie
Table 1: Optimization of the catalytic hydroacylation reaction.
examined. Phenylacetylenes bearing either electron-donating
or electron-withdrawing groups afforded the corresponding
products in good to moderate yields (5b–j). However, 5k was
obtained in only 29% yield most probably due to the strong
electron-withdrawing effect of the nitro group. Biphenyl,
naphthyl, and thiophenyl moieties were also tolerated in this
reaction (5l–n). Attempts to prepare bis-hydroacylated
products in a bidirectional process failed. Instead, only
mono-hydroacylation products were obtained in moderate
yields (5o,p). In addition to the aromatic systems, an aliphatic
alkyne could also be transformed (5q), albeit in low yields.
Finally, two other glyoxal derivatives were tested. Both of
them turned out to be suitable starting materials leading to
the desired products in moderate yields (5r,s). Most of these
reactions had a clear preference for the E diastereomer; an
olefinic coupling constant of 16.2–16.4 Hz clearly demon-
strated that. Only in a few cases were small amounts (7.5–
10%) of the Z product detected.
Entry Catalyst[a]
Solvent R1R2NH
T [8C] t [h] Yield[b]
1
2
3
4
5
AuBr3
AuBr3
AuCl3
toluene piperidine
toluene piperidine
toluene piperidine
RT
50
50
50
50
48 66%
16 60%
16 45%
16 38%
16 trace
NaAuCl4·2H2O toluene piperidine
IPrAuCl3
toluene piperidine
6
toluene piperidine
50
16 63%
7
8
AuCl
toluene piperidine
toluene piperidine
50
50
16 71%
16 trace
Ph3PAuCl/
AgNTf2
IPrAuCl/
AgNTf2
9
toluene piperidine
50
16 trace
The products prepared by this procedure are valuable
precursors for the synthesis of heterocycles. For example, 5a
could readily be reacted with o-phenylenediamine to afford
the interesting vinyl-substituted quinoxaline 6a in good yield
(Scheme 4).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
AuBr3/3AgOTf toluene piperidine
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
16 trace
16 trace
16 19%
16 trace
AgOTf
CuCl
PtCl2
AuCl
AuCl
AuCl
AuCl
AuCl
AuCl
AuCl
AuCl
AuCl
AuCl
AuCl
AuCl
AuCl
–
toluene piperidine
toluene piperidine
toluene piperidine
CHCl3
piperidine
3
5
34%
15%
As the next step we investigated the reaction mechanism.
1
CH3CN piperidine
CH2Cl2 piperidine
When the crude product was directly analyzed by H NMR
16 47%
16 53%
16 55%
16 trace
16 trace
18 16%
18 NR
16 32%
16 41%
16 25%
16 42%
16 NR
spectroscopy, surprisingly, product 5a was not detected. Only
the spectrum of a complex mixture was obtained. GC/MS
analysis showed that the mixture contained one main product
with a molecular weight corresponding to the A3 coupling
product 7a. The final product 5a became the main component
after silica gel was added to this crude product (Scheme 5).
To gain further insight, the a-deuterated phenylglyoxal
was prepared by oxidation of deuterated phenylacetylene.[2]
Its conversion under standard conditions showed no deute-
rium incorporation into the final product, which excludes
a 1,3-hydrogen shift[8] as an elementary step of the reaction
mechanism. Our next labeling experiment was conducted
with D2O as a deuterium source and product 5a’ was
obtained. Both of the double-bond positions bear deuterium
labels (46% and 63% incorporation). In another labeling
experiment, the crude intermediate, which was obtained
directly by evaporating toluene from the reaction mixture,
was added to an acetone/H218O mixture which resulted in the
18O-labeled product 5a’’. [D11]Piperidine was then used as
a base but no deuterium was incorporated into product 5a,
which excludes a possible 1,5-hydride shift from the piper-
idine onto the activated alkyne[9] (Scheme 6).
THF
piperidine
benzene piperidine
hexane piperidine
Et2O
piperidine
toluene Et2NH
toluene Et3N
toluene morpholine
toluene pyrrolidine
toluene (nBu)2NH
toluene hexamethyleneimine 50
toluene piperidine
toluene
50
50
50
50
50
AuCl
AuCl[c]
AuCl[d]
AuCl
–
16 NR
toluene piperidine
toluene piperidine
toluene piperidine[e]
16 44%
16 31%
16 trace
[a] 10 mol% catalyst. [b] Yield of isolated product. [c] 3 mol% catalyst.
[d] 1 mol% catalyst. [e] 0.2 equiv piperidine. IPr=1,3-bis(2,6-diisopropyl-
phenyl)imidazol-2-ylidene, Tf=triflate, NR=no reaction.
On the basis of these observations we propose the
following mechanism (Scheme 7): Initial gold-catalyzed A3
coupling results in carbonyl propargylamine 6a. Based on the
acceptor in a-position (which is not present in the known A3
couplings), the acidity of the propargylic proton is enhanced
which enables isomerization[10] (assisted by the piperidine) to
give the conjugated allenylamine intermediate 6a’. Hydrol-
ysis of the enamine substructure on silica gel then leads to the
final product 5a; the thermodynamically more stable E
diastereomer is formed preferentially.
Scheme 2. Gold-catalyzed oxidative coupling of phenylglyoxal and
piperidine.
conditions (2.0 equiv of piperidine, 10 mol% AuCl, toluene,
N2 atmosphere, 508C). The results are summarized in
Scheme 3. A wide range of alkenyl diketones could readily
be obtained by following this procedure. The reactions of
phenylglyoxal 1a with various terminal alkynes were first
In conclusion, we have developed an efficient cascade
reaction for the preparation of alkenyl-1,2-diketones from
Angew. Chem. Int. Ed. 2014, 53, 1148 –1151
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1149

.
Angewandte
Communications
Scheme 3. Synthesis of alkenyl 1,2-diketones from different glyoxals and alkynes. Reaction conditions: 1 (0.3 mmol), 2 (0.6 mmol), piperidine (3a,
0.6 mmol), AuCl (7 mg, 10 mol%), toluene, N2 atmosphere, 508C, yield of isolated product.
Scheme 4. Synthesis of vinylquinoxaline derivative 6a from 5a.
Scheme 7. Reaction mechanism in accordance with the labeling stud-
Scheme 5. Conversion of intermediate 7a into the final product.
ies.
1150
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1148 –1151

Angewandte
Chemie
2746; g) V. A. Peshkov, O. P. Pereshivko, E. V.
Van der Eycken, Chem. Soc. Rev. 2012, 41, 3790 –
3807.
[2] S. Shi, T. Wang, W. Yang, M. Rudolph, A. S. K.
Hashmi, Chem. Eur. J. 2013, 19, 6576 – 6580.
[3] a) H. Jiang, H. Huang, H. Cao, C. Qi, Org. Lett.
2010, 12, 5561 – 5563; b) O. Shuichi, T. Eiji, O.
Kazuei, K. Tomoko, T. Yukari, I. Yoshio, J. Org.
Chem. 1999, 64, 8660 – 8867; c) S. D. Jadhav, N. D.
Kokare, S. D. Jadhav, J. Heterocycl. Chem. 2009, 5,
1461 – 1464; d) B. Khalili, P. Jajarmi, B. Eftekhari-
Sis, M. M. Hashemi, J. Org. Chem. 2008, 73, 2090 –
2095; e) S. Onitsukaa, H. Nishinob, K. Kurosawac,
Tetrahedron Lett. 2000, 17, 3149 – 3152; f) W. Xue,
Q. Li, Y. Zhu, J. Wang, A. Wu, Chem. Commun.
2012, 48, 3485 – 3487.
[4] During our investigation, one gold-catalyzed three-
component coupling reaction for the synthesis of
aminofurans was published, see: J. Li, L. Liu, D.
Ding, J. Sun, Y. Ji, J. Dong, Org. Lett. 2013, 15,
2884 – 2887.
[5] a) S. Ahmad, J. Iqbal, J. Chem. Soc. Chem.
Commun. 1987, 692 – 693; b) M. S. Yusubov, V. D.
Filimonov, V. P. Vasilyeva, K.-W. Chi, Synthesis
1995, 1234 – 1236; c) A. R. Katritzky, Z. Wang, H.
Lang, D. Feng, J. Org. Chem. 1997, 62, 4125 – 4130;
d) T.-C. Hu, P.-I. Hsiao, T.-H. Wang, Y.-W. Yang, C.-
Y. Chang, Y.-H. Wu, W.-C. Sun, M.-S. Yu, C.-Y. Lee, Y.-H. Lo,
Dalton Trans. 2011, 40, 12663 – 12666.
Scheme 6. Labeling experiments.
readily available staring materials. The obtained alkenyl-1,2-
diketones can serve as useful building blocks for the synthesis
of heterocycles, which was demonstrated by the formation of
a vinylquinoxaline derivative. Further research on the scope
and mechanism of this new gold-catalyzed hydroacylation is
ongoing in our laboratories.
[6] A. S. K. Hashmi, J. P. Weyrauch, M. Rudolph, E. Kurpejovic,
Angew. Chem. 2004, 116, 6707 – 6709; Angew. Chem. Int. Ed.
2004, 43, 6545 – 6547.
[7] For the transition-metal-catalyzed synthesis of amides from
aldehydes and amines, see: a) T. Naota, S.-I. Murahashi, Synlett
1991, 693 – 694; b) G.-L. Li, K. K.-Y. Kung, M.-K. Wong, Chem.
Commun. 2012, 48, 4112 – 4114; c) J. A. Thomson, L. L. Schafer,
Dalton Trans. 2012, 41, 7897 – 7904.
Experimental Section
An N2-protected Schlenk tube was charged with anhydrous
toluene (2 mL), glyoxal derivative 1 (0.3 mmol, 1.0 equiv),
alkyne 2 (0.6 mmol, 2.0 equiv), piperidine 3a (0.6 mmol,
2.0 equiv), and AuCl (7 mg, 0.1 equiv) in sequence. This
solution was stirred at 508C for 16 h under an N2 atmosphere.
After complete conversion, as monitored by TLC analysis, the
solvent was evaporated and the residue was purified by
column chromatography on silica gel.
[8] For this type of transformation from propargyl alcohols to
chalcone derivatives, see: a) T. Baba, H. Kizuka, H. Handa, Y.
Ono, Appl. Catal. A 2000, 194 – 195, 203 – 211; b) T. Ishikawa, T.
Mizuta, K. Hagiwara, T. Aikawa, T. Kudo, S. Saito, J. Org. Chem.
2003, 68, 3702 – 3705; c) R. Erenler, J.-F. Biellmann, Tetrahedron
Lett. 2005, 46, 5683 – 5685; d) M. J. Albaladejo, F. Alonso, M.
Yus, Chem. Eur. J. 2013, 19, 5242 – 5245.
[9] For selected publications of transition-metal-catalyzed 1,5-
hydride shifts, see: a) J. Kuang, S. Ma, J. Am. Chem. Soc. 2010,
132, 1786 – 1787; b) M. Melchionna, M. Nieger, J. Helaja, Chem.
Eur. J. 2010, 16, 8262 – 8267; M. Saifuddin, P. K. Agarwal, B.
Kundu, J. Org. Chem. 2011, 76, 10122 – 10128.
Received: September 1, 2013
Published online: December 11, 2013
[10] a) A. S. K. Hashmi in Modern Allene Chemistry (Eds.: A. S. K.
Hashmi, N. Krause), Wiley-VCH, Weinheim, 2004, pp. 3 – 50;
the process is strongly related to the formation of allenyl ketones
from propargyl ketones, see: b) A. S. K. Hashmi, Angew. Chem.
1995, 107, 1749 – 1751; Angew. Chem. Int. Ed. Engl. 1995, 34,
1581 – 1583; c) A. S. K. Hashmi, T. L. Ruppert, T. Knçfel, J. W.
Bats, J. Org. Chem. 1997, 62, 7295 – 7304; d) A. S. K. Hashmi,
J. W. Bats, J.-H. Choi, L. Schwarz, Tetrahedron Lett. 1998, 39,
7491 – 7494; e) A. S. K. Hashmi, J.-H. Choi, J. W. Bats, J. Prakt.
Chem. 1999, 341, 342-357; f) A. S. K. Hashmi, L. Schwarz, J.-H.
Choi, T. M. Frost, Angew. Chem. 2000, 112, 2382 – 2385; Angew.
Chem. Int. Ed. Engl. 2000, 39, 2285 – 2288.
Keywords: 1,2-dicarbonyl compounds · alkynes · gold ·
Michael acceptors · quinoxaline
.
[1] For selected examples of A3 coupling, see: a) A. B. Dyatkin,
R. A. Rivero, Tetrahedron Lett. 1998, 39, 3647 – 3650; b) C.-J. Li,
C. Wei, Chem. Commun. 2002, 268 – 269; c) C. M. Wei, C.-J. Li, J.
Am. Chem. Soc. 2003, 125, 9584 – 9585; d) C. Wei, Z. Li, C.-J. Li,
Org. Lett. 2003, 5, 4473 – 4475; e) B. Yan, Y. Liu, Org. Lett. 2007,
9, 4323 – 4326; f) N. Chernyak, V. Gevorgyan, Angew. Chem.
2010, 122, 2803 – 2806; Angew. Chem. Int. Ed. 2010, 49, 2743 –
Angew. Chem. Int. Ed. 2014, 53, 1148 –1151
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1151