J. Schörgenhumer, M. Waser / Tetrahedron Letters 57 (2016) 1678–1680
1679
Based on this first hint, we tested a series of different commonly
employed hypervalent iodine reagents under different conditions
to evaluate the best-suited system for the formation of 7 (Table 1).
Initial attempts with reagent 6 revealed that n-BuLi and LiHMDS
are the only strong bases that allow for the formation of 7 (see
entries 1–4). Fortunately, in these two cases diyne 7 was obtained
in high yields of up to 90% (entries 1 and 2) when performing the
reaction in THF (starting from À40 °C for the addition of the base
and then stirring at room temperature for 2 h). Attempts to use
only half an equivalent of iodine-reagent 6 resulted in less than
50% yield after 2 h (entry 5). However, a prolonged reaction time
of 40 h gave 70% of product 7. This indicates that a mechanism
where half of the starting alkyne 4 will add to the iodonium
reagent first, giving an electrophilic alkyne-transfer reagent which
is then attacked by the remaining alkyne (which should be present
as the Li-acetylide) is possible. However, it is not clear why the
reaction is that much slower when using only 0.5 equiv of 6. Using
only 0.5 equiv of n-BuLi also resulted in less than 50% yield under
the standard conditions (entry 6). Unfortunately, we were not able
to use catalytic amounts of 6 in combination with an alternative
stoichiometric oxidant (entry 7). We next tested a variety of other
easily available hypervalent iodine reagents (entries 8–13) under
the optimum reaction conditions. Among those, only Dess–Martin
periodinane 9 and derivative 12 allowed for some product forma-
tion, whereas acyclic derivatives 10 and 11 gave no product at all.
Having identified suitable conditions for the high-yielding
homocoupling of 4, we employed other terminal alkynes in this
reaction. It was found that a variety of different aryl alkynes were
well tolerated under these conditions (giving products 13–17
shown in Scheme 2). In addition, aliphatic alkynes can also be
dimerized with this method. Hereby it is fair to mention that espe-
cially the presence of NH-carbamate groups resulted in a reduced
yield (see compound 24). Nevertheless, a variety of different termi-
nal alkynes could be successfully coupled.
Scheme 2. Application scope for the homocoupling of different alkynes in the
presence of benziodoxole reagent 6.
Scheme 3. Control experiment using preformed alkynebenziodoxole 25.
Finally, we were interested to obtain more insight into a possi-
ble mechanism of this reaction. As already noted during the opti-
mization of the reaction conditions, a possible path could be the
formation of the electrophilic alkyne-transfer reagent 25 first,
which is then reacting with a Li-acetylide (e.g., Li-4). We thus car-
ried out control experiments where we used preformed literature-
known alkynebenziodoxole 2512 under the optimized conditions
and reacted it with Li-4 (Scheme 3). Full conversion was observed
within 2 h giving dimer 7 in 90% yield, which strongly supports
that the herein reported reaction actually proceeds via the forma-
tion of compound 25 as the key-intermediate. This mechanism also
opens the door for future applications directed toward the forma-
tion of dissymmetrically substituted diynes.
Table 1
Identification of the best-suited iodine reagent and conditions for the dimerization of
4
Conclusion
The use of hypervalent iodine reagents allows for the efficient
transition metal-free dimerization of a variety of terminal alkynes.
Key to success in this transformation is the use of acetoxy-benzio-
doxole 6 as the activating agent, which clearly outperformed other
hypervalent iodine species. Based on control experiments that
were carried out using preformed hypervalent iodine-based
alkyne-transfer reagents we propose a mechanism where one
equivalent of the alkyne adds to benziodoxole 6 to give an elec-
trophilic alkynebenziodoxole derivative 25 first, which then reacts
with the Li-acetylide of the second equivalent of the alkyne. This
strategy should thus also provide a future option to obtain dissym-
metric 1,3-diynes.
Entry
[I]-reagent
Base
Yielda (%)
1
2
3
4
5
6
7
8
6 (1.1 equiv)
6 (1.1 equiv)
6 (1.1 equiv)
6 (1.1 equiv)
6 (0.5 equiv)
6 (1.1 equiv)
6 (0.1 equiv)d
1 (1.1 equiv)
8 (1.1 equiv)
9 (1.1 equiv)
10 (1.1 equiv)
11 (1.1 equiv)
12 (1.1 equiv)
LiHMDS (1.1 equiv)
n-BuLi (1.1 equiv)
t-BuOK (1.1 equiv)
NaH (1.1 equiv)
n-BuLi (1.1 equiv)
n-BuLi (0.5 equiv)
n-BuLi (1.1 equiv)
n-BuLi (1.1 equiv)
n-BuLi (1.1 equiv)
n-BuLi (1.1 equiv)
n-BuLi (1.1 equiv)
n-BuLi (1.1 equiv)
n-BuLi (1.1 equiv)
84
90
n.r.
n.r.
<50b (70)c
<50b
<10b
n.r.
9
n.r.
30
n.r.
n.r.
10
11
12
13
Experimental section
18
General procedure for the dimerization of terminal alkynes
The bold entry highlights the most active system identified in the screening.
a
Isolated yields.
NMR yield.
A 1.6 M solution of n-BuLi in hexanes (1.1 equiv) was added to a
solution of the corresponding terminal alkyne (1.0 equiv) in THF
(5 mL per mmol) at À40 °C. After stirring for 30 min, acetoxy-ben-
b
c
After 40 h reaction time.
Using additional stoichiometric oxidants like, e.g., Oxone, NaIO4 or Ca(ClO)2.
d