Angewandte
Communications
Chemie
catalytic alkynylation of terminal alkynes with alkynyl hyper-
valent iodine(III) reagents has not been reported to date.[16]
Initially, the cross-coupling of the aryl alkyne 1a with
phenyl-substituted ethynylbenziodoxole derivatives under
gold-catalyzed conditions was chosen to test the possibility
of our hypothesis (Table 1). The introduction of
observed, thus suggesting the vital effect of silver in the
transformation (entry 4). As expected, not adding Phen
resulted in poor yield and selectivity (entry 5). A control
experiment demonstrated the necessity of the gold catalyst
for the cross-coupling reaction (entry 6). The amount of Phen
also played an important role in this reaction, since decreasing
the amount of Phen to 0.1 equivalents afforded 3a in
a decreased yield with longer reaction time (entry 8). This
result is likely due to the competitive coordination of Phen to
silver to form a [Ag(Phen)2]+ complex,[19] which consumes
some of the Phen. Such results suggested that additional free
Phen was required, possibly as a ligand to complex with gold.
When the alkynylbenziodoxolone 5 was used instead of 2a,
only trace amounts of 3a were observed, thus indicating the
superior reactivity of 2a (entry 9). Specifically, we found Phen
and its derivatives L1 and L2 showed better reactivity than
bipyridine, presumably because of their significant backbone
rigidity compared to the nonfused analogues (entries 10–12).
Various gold(I) and gold(III) complexes were also examined
(entries 13–17). Among them, PicAuCl2 led to formation of
3a in 82% yield, and the commonly used PPh3AuNTf2
showed comparable activity to that of TA-Au.
With a set of optimized reaction conditions in hand, we
focused our attention on the preparative scope of this gold-
catalyzed C(sp)–C(sp) cross-coupling reaction (Table 2). The
scope with respect to alkynes was firstly investigated with 2a
as the coupling partner. To our delight, the present method
could be applied successfully to a wide variety of aryl-,
heteroaryl-, alkenyl-, and alkyl-substituted terminal alkynes.
Aryl alkynes, bearing either electron-withdrawing (p-F, p-Cl,
p-Br, p-CO2Et, p-CF3) or electron-donating [p-Me, p-MeO,
p-NHCbz and 3,4,5-tri(MeO)] groups coupled readily with 2a
to afford aryl,aryl-substituted butadiynes (3a–h and 3k) in
high yields (83–96%) within short reaction times. The
electronic nature of the aryl substituents had little influence
on the product yields. Of note is that halogen groups, which
are usually not tolerated in transition metal catalyzed
reactions, remained intact under the standard reaction con-
ditions, thus highlighting the orthogonal reactivity of gold
over palladium, copper, and nickel catalysis. Sterically
demanding o-Me- and o-CF3-, 2,4,6-tri(Me)-, and 1-naph-
thyl-substituted aryl alkynes reacted smoothly to give 3i–j
and 3l–m in 80–97% yields, thus indicating that the reaction
was not sensitive to steric effects of terminal alkynes.
Heteroaryl-substituted alkynes such as 2-thienyl- and
2-pyridyl-substituted ones also turned out to be effective
substrates, thus furnishing 3n,o in 73–77% yields. Addition-
ally, 3-en-1-ynes could be satisfactorily coupled in high yield,
as exemplified by 3p. Functional-group tolerance was also
observed with aliphatic alkynes. A large variety of alkyl
alkynes reacted well with 2a, thus providing the correspond-
ing alkyl,aryl-substituted butadiynes 3q–za in good to excel-
lent yields. Substrates containing nitrile (3v), ether (3w and
3x), and sulfamide (3y) groups were all suitable for this
reaction. In particular, propargyl acetate, which is known to
easily undergo 1,2-migration reactions in gold catalysis,[20]
could also be accommodated, thus furnishing 3z in 85%
yield. Intriguingly, free aliphatic alcohols (3za) do not
interfere, and illustrate the potential of our methodology in
Table 1: Optimization of the reaction conditions.[a]
Entry Deviation from standard condi-
tions
t [h] Yield [%][b] Yield [%][b]
of 3a
of 4a
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
none
0.5 98
1
46
7
0
10
0
0
4
0
8
without AgOTs and Phen
toluene instead of CH3CN, 508C 12
no AgOTs
no Phen
12
31
85
17
13
0
13
86
3
24
24
24
18
12
12
12
no TA-Au
NaOTs instead of AgOTs
0.1 equiv of Phen was used
5 instead of 2a
L1 instead of Phen
L2 instead of Phen
bpy instead of Phen
AuCl instead of TA-Au
AuCl3 instead of TA-Au
PicAuCl2 instead of TA-Au
PPh3AuCl instead of TA-Au
PPh3AuNTf2 instead of TA-Au
90
1.5 87
7
8
13
1
8
4
13
72
78
18
82
76
20
4
12
23
1
1.5 95
[a] 1a (0.20 mmol), 2a (0.20 mmol), TA-Au (0.01 mmol), AgOTs
(0.01 mmol), Phen (0.10 mmol) in CH3CN (0.1m) at room temperature
under argon. [b] Yields determined by 19F NMR spectroscopy using 1,3,5-
trifluorobenzene as an internal standard. Phen=1,10-phenanthroline,
Tf =trifluoromethanesulfonyl, Ts=4-toluenesulfonyl.
phenylethynyliodine(III) reagents to the gold-catalyzed
direct alkynylation reactions is highly desired, but more
challenging as only silyl-substituted ethynyliodine(III)
reagents, such as 1-[(triisopropylsilyl)ethynyl]-1,2-benzio-
doxol-3(1H)-one (TIPS-EBX), display good performance.[15b]
A thorough optimization study of the alkynyl source, gold
catalysts, and additives led us to discover that the use of
5 mol% of TA-Au,[17] 5 mol% AgOTs, and 0.5 equivalents of
Phen, with the bis(trifluoromethyl) benziodoxole 2a as the
alkynylation reagent enabled the reaction to go to completion
at room temperature within 30 minutes.[18] The desired
unsymmetrical diyne 3a was obtained with a high yield and
excellent heteroselectivity (entry 1). However, using TA-Au
alone as the catalyst led to a deleterious effect on the
selectivity (entry 2). Additionally, the coupling reaction
occurs faster in CH3CN than in toluene (entry 3). In the
absence of AgOTs, significant erosion in the yield was
2
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
These are not the final page numbers!