proposed. The first calls for slowly adding the terminal al-
kynes, using a Pd0 catalyst in the presence a phase-transfer
reagent.[5] However for slow Sonogashira reactions, homo-
coupling is still not avoided by this method.[6] The second
method calls for additionally mixing hydrogen gas with the
inert gas to act as a reducing reagent.[6] However, in the
presence of a palladium catalyst, hydrogen adds to the
alkyne to form significant amounts of alkene and alkane
side products. For example, in one of our syntheses using
this method, a [previously unreported] 1,5-bisalkoxy-2,4-eth-
ylbenzene side product was isolated in 15% yield, (con-
chamber E, a reactant-addition passage C, and a solvent-ad-
dition passage B having a ground-glass joint connection A.
Solid reactant is added to chamber E through passage C,
which is then sealed (dashed line). Solvent and liquid reac-
tants (trimethylsilylacetylene [TMSA], Et3N) are then
added through passage A to chamber D, after which cham-
ber D is cooled to liquid nitrogen temperature, and the reac-
tion system is connected to high vacuum (about 10À5 Torr).
Solvent and liquid reactants are then subjected to three-to-
four freeze–thaw cycles to eliminate any dissolved oxygen,
following which passage B is sealed under vacuum. Finally,
solvent and liquid reactants are delivered to reaction cham-
ber E by vaper-phase solvent transfer.
The mechanism for catalytic oxidation of terminal acetyl-
ene is still undetermined; however, there is no question that
oxygen is involved as an oxidizing reagent.[6] Reaction yields
are high for both azulene diiodide[10] and dialkoxybenzene
diiodides (Table 1). For example, 1,3-ditrimethylsilylethynyl-
azulene and 1,3-diphenylethylnylazulene are obtained with
yields of 99 and 96%, respectively, which are much higher
than literature values (80–90%).[11a,b] It has also been repor-
ted[5a] that a lower mole percentage of catalyst results in
lower yields and longer reaction times and that under
normal reaction conditions the catalyst can decompose to
form a black-colored complex, either dissolved in solution
or precipitated out. In our reaction system, just a small
amount of catalyst can lead to completion of the reaction in
good yield, since no catalyst is consumed by the oxidation
during the reaction cycle. Thus a distinctive feature of our
system is that after complete reaction the solution retains a
light color due only to reaction product and residual catalyst
complex, with triethylamine hydroiodide crystallized
(Figure 2) on the bottom of chamber E.
The completion of reactions was conveniently monitored
by observing the formation of triethylamine hydroiodide
crystals during the reaction process. When no more crystal
growth was visible at the bottom of Chamber E, the reaction
solution was carefully moved to a new glass surface; the ap-
proximate completion of the reaction was indicated when
there were no more crystals depositing onto the new glass
surface. Approximate reaction times could be determined
with a precision of 8 h. The above monitoring method was
checked in one case by comparison with a more precise
method by sampling the reaction mixture in two-hour inter-
vals followed by TLC analyses (see the Supporting Informa-
tion for details). The reaction time in this case was deter-
mined with a precision of 2 h.
Our reaction system is well suited not only for syntheses
using iodoazulenes but also for those using dibromoazu-
lene.[12] The syntheses of 1-ethynylazulene and 1,3-diethynyl-
azulene from corresponding iodoazulenes have been report-
ed previously in 80–90% yield,[11a,b,d] as have the syntheses
of 2-ethynyl- and 6-ethynylazulene from 2-bromo- and 6-
bromoazulene, respectively.[1a,11e] However, it is well known
that it is difficult to obtain 1,3-diethynylazulene and its de-
rivatives from 1,3-dibromoazulene, since the bromo group is
a relatively poor leaving group, and the 1 and 3 positions
1
firmed by its H NMR spectrum; see Figure S4 in the Sup-
porting Information), whereas the desired product 1,5-bisal-
koxy-2,4-ethynylbenzene was obtained only as part of a
complicated mixture that also included 1,5-bisalkoxy 2-eth-
ylnyl-4-vinylbenzene, 1,5-bisalkoxy-2,4-divinylbenzene, and
other structurally similar components which were almost im-
possible to separate under conventional workup conditions.
Therefore Sonogashira coupling in a protecting atmosphere
containing reducing hydrogen gas is impractical for sp2–sp
coupling of electron-enriched aryl dihalides.
Because of the deficiencies of the above-mentioned two
methods, recent efforts have focused on the development of
copper-free Sonogoshira reactions that minimize the forma-
tion of diacetylene side products.[7] However, as a relatively
new and only partially understood methodology, the mecha-
nism of copper-free Sonogashira coupling is still under
debate. For example, it has been reported that the formation
of the (h2-RCꢀCH)Pd0 L2 intermediate, along with the
strong tendency for the Pd0 complex to undergo ligand ex-
change in some cases, leads to the termination of the cata-
lytic cycle prior to total conversion of the reactants.[8] To
date there has not been any systematic study of the copper-
free Sonogashira reaction applied to electron-enriched aryl
halide substrates. Therefore there remains a significant
demand for an ultra-clean, high-yield Sonogashira reaction.
The reaction system and conditions that we have devel-
oped and discussed herein address this demand. By provid-
ing a long-term, scrupulously oxygen-free and water-free en-
vironment, our novel two-chamber reaction system
(Figure 1) renders nearly ideal reaction conditions for suc-
cessfully completing even the slowest Sonogashira reactions
with high yield. More specifically, since most Sonogashira
reactions do not generate gaseous side products, and only
moderate heat is released from the neutralization of HX
with Et3N, we have been able to use a totally sealed, two-
chamber reaction system (Figure 1) to exclude oxygen. This
system contains a solvent degas chamber D, a reaction
Figure 1. The two-chamber reaction system.
2042
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 2041 – 2044
Chen, Xue-Yi
