Ethynylation
COMMUNICATION
[14,15]
À
nary centers with four different carbon substituents (Table 2,
entries 5–10). Both methyl and phenyl ketones (Table 2, en-
tries 5–10) with several different a-alkyl substituents, includ-
ing an allyl group, could be used, which provided access to
the versatile 1,5-enyne product 9 f (Table 2, entry 10). Cyano
and nitro esters (Table 2, entries 11–16) were also good sub-
strates for the reaction. Importantly, the alkynylation of
theses two classes of compounds has never been reported
before. The nitro substrates in particular are very sensitive
compounds and the mild conditions developed were crucial
to obtaining good yields.[12] A practical issue with the ethy-
nylation reaction is caused by the similar polarity of the
starting materials and products, which makes their separa-
tion by thin layer or column chromatography nearly impos-
sible. Consequently, complete conversion was required to
allow purification of the products. For slow reacting sub-
strates, a better conversion was achieved if reagent 2 was
added slowly at À788C by using a syringe pump.
ported conditions for the reduction of the N O bond.
If
Zn dust was used under more forcing conditions, allyl amine
16 was obtained in 57% yield.[16] Gratifyingly, we found that
À
selective reduction of the N O bond in 15 was possible by
using SmI2 in THF/tBuOH.[17] Purification of the free amine
was difficult, but quenching the reaction with trifluoroacetic
anhydride (TFAA) allowed the isolation of the correspond-
ing trifluoro amide (17) in good yield and purity. The ob-
tained protected alkynyl amino acids display interesting bio-
logical activities and few methods have been reported for
their synthesis.[18]
During optimization of the reaction, we speculated that
EBX (3) was the alkynylating agent. As it was not possible
to isolate this reagent, we decided to monitor its formation
by 1H and 13C NMR at low temperature. Treating TMS–
EBX (2) with TBAF at À788C led to immediate conversion
to a new compound, the spectrum of which was in full
agreement with the structure of EBX (3).[19] The H NMR
1
The obtained propargylic nitro and cyano products con-
taining an ester group are new structures that have never
before been synthesized. In particular, propargylic nitro
compounds with a free acetylene are a very rare class of
compounds and their properties have never been studied in
detail. Consequently, we decided to examine the synthetic
potential of product 13b more intensively (Scheme 2).
spectrum remained unchanged when the solution was
heated up to À208C. At this point, EBX (3) gradually de-
composed to generate several as yet unidentified products.
If a substrate was added to the EBX (3) solution, the only
signals observed belonged to EBX (3), 2-iodobenzoic acid,
the substrate and the ethynylation product; no other inter-
mediates were observed.[20]
In principle, two reaction pathways can be envisaged for
this reaction (Scheme 3): addition of the enolate to the
iodine atom followed by reductive elimination (pathway A)
Scheme 2. Scaled-up synthesis and functionalization of 13b. Reaction
conditions: a) 2 (1.3 equiv), TBAF (1.3 equiv), THF, À788C, 77%;
b) BnN3 (1 equiv), CuSO4 (5 mol%), sodium ascorbate (10 mol%),
tBuOH/H2O (1:1), 608C, 65%; c) Zn, HCl/AcOH (1 n), 08C, 57%;
d) Zn, NH4Cl, EtOH/H2O (1:1), 08C, 94%; e) SmI2, THF, tBuOH, then
TFAA, 67%.
Scheme 3. Possible mechanisms for the ethynylation reaction and labeling
experiment (Ar=phenyl-2-carboxylate).
or conjugate addition to the alkyne, followed by an elimina-
tion and 1,2-hydride shift (pathway B). For both pathways,
initial interaction with the carbonyl oxygen could also be en-
visaged. The use of 13C-labeled reagent 18[21] led to product
20, which is only consistent with the 1,2-hydride-shift path-
way. This mechanism has also been proposed in the case of
alkynyliodonium salts.[8b] Interestingly, the opposite result
was obtained in the case of metal-catalyzed alkynylation re-
actions that used TIPS–EBX (1).[10]
The use of benziodoxolone-based reagents for the ethyn-
AHCTUNGTREGyNNNU lation reaction allowed us to increase the scope and effi-
ciency of the reaction. Nevertheless, the obtained products
are racemic and an asymmetric method would be highly de-
The ethynylation of 12b proceeded with a yield of 77%
on a 4.7 mmol scale to give 13b. The Cu-catalyzed [3+2] cy-
cloaddition of 13b with BnN3 gave the corresponding tria-
zole (14) in 65% yield.[13] This constitutes the first example
of a [3+2] cycloaddition reaction of a propargyl nitro com-
pound. Reduction of the nitro group to the amine was at-
tempted next. To the best of our knowledge, there is only
one report of the reduction of a propargylic nitro compound
to the amine, which proceeds through the corresponding hy-
droxylamine.[14] Although reduction to hydroxylamine 15
with Zn dust worked well, it was not possible to use the re-
Chem. Eur. J. 2010, 16, 9457 – 9461
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9459