base (see the Supporting Information). (2) Silicate pathways
for silane deprotection (both with and without added
Cu+)9a,d,e,14 are invalidated by the almost quantitative bisaryl-
ethyne product yields observed when reactions were per-
formed with substoichiometric (40 mol % relative to 7)
amounts of added water (Table 1). In addition, yields are
depressed when no water was purposefully added, suggesting
the involvement of adventitious moisture.15 (3) Bisarylethyne
yield suppression was observed under “protonless” reaction
conditions (mixtures of 7, 9, DBU, benzene, PdCl2(PPh3)2,
CuI, and no added trimethylsilylethynylene or water; see the
Supporting Information). Moreover, the addition of catalytic
HI16 to anhydrous reactions such as those listed in Table 1
show good yields of bisarylethyne 8 (see the Supporting
Information). Taken together, these three results indicate the
amidine base is acting as a proton shuttle. In support of this
determination, the literature reinforces the invocation of a
DBU salt in the organic reaction mixture,17 and DBU has
been used catalytically in the nucleophilic addition of
acyldiazomethanes to aldehydes and imines.18 Therefore, we
speculate that after proceeding through the commonly
accepted cross-coupling chemistry, the silane-protected aryl-
ethynylene converges with Cu+ and a water/DBU salt,
resulting in protodesilylation to yield the terminal ethynylene.
Consequently, the aryl-substituted terminal ethynylene is
resubmitted to the cross-coupling cycle, generating the
bisarylethynyl product after a second pass.
Table 2. Synthesis of Symmetrical Bisarylethynylenes from
Aryl Halides and Aryl Triflatesa
Using the optimized protocol detailed above (Scheme 1;
Table 1, entry 1), several symmetrical bisarylethynylenes
were prepared (Table 2). Similar procedures yielding such
symmetrical products have been reported; however, these
employ the in situ elimination of acetone under basic
conditions9b,c or the use of acetylene gas.19 Aryl iodides,
triflates, and bromides are all reactive and give rise to the
symmetrical bisarylethynylenes. Of particular note is the
substituent effect observed for both iodides and bromides.
Iodides with electron-withdrawing groups (acetyl, methoxy-
carbonyl, and trifluoromethyl) give excellent yields of their
respective bisarylethynes at room temperature. In contrast,
iodides with electron-donating groups such as methoxy and
a All reactions were carried out in the absence of light for 18 h with a
0.2 M solution of aryl halide/triflate in benzene using 6 mol % PdCl2(PPh3)2,
10 mol % CuI, 6.0 equiv of DBU, 40 mol % water, and 0.5 equiv of
trimethylsilylethynylene (all relative to 1 equiv of aryl halide/triflate).
b Isolated yield based on average of two runs. c From the precursor iodide.
d From the precursor bromide. e From the precursor triflate.
methyl require elevated temperatures (60 °C) to proceed in
good yield. The same applies to the more sterically hindered
o-iodohalide substrates; in comparison to reactions run at
60 °C, reduced yields were observed at room temperature.
Triflates follow this pattern as well. Aryl iodides 2-chloro-
5-iodopyridine, 1-iodonaphthalene, and 2-iodothiophene gave
high yields of their respective symmetrical bisarylethynylene
analogues at room temperature employing the standard
protocol. Additionally, room-temperature iodides are easily
scaled up into the 2 g range (see the Supporting Information).
In general, aryl bromides required elevated temperatures (80
°C) for complete conversion.20
(13) We ascribe this observation to the known comparison of the reaction
rates of DBU (faster) and DBN (slower) in elimination reactions. Ency-
clopedia of Reagents for Organic Synthesis; Paquette, L. A., Ed.; John Wiley
& Sons: New York, 1995.
(14) (a) Correia, R.; DeShong, P. J. Org. Chem. 2001, 66, 7159. (b)
Manoso, A. S.; DeShong, P. J. Org. Chem. 2001, 66, 7449 and references
therein.
(15) Without intentional addition of water, we always isolated 8 as the
major product, although rigorously purified solvents and reagents were used
as well as careful syringe transfer techniques. For example, we observed
72% yield of 8 (by GC-MS analysis) using a 0.20 M benzene solution of
1 equiv of 1-bromo-3-iodobenzene (7), 6.0 equiv of DBU, 0.5 equiv of
trimethylsilylethynylene, 6.0 mol % PdCl2(PPh3)2, and 10 mol % CuI.
(16) Grieco, P. A.; Markworth, C. J. Tetrahedron Lett. 1999, 40, 665.
(17) Values of pKa in water for triethylamine (the most common base
used for Sonogashira couplings) have been measured at 10.8. In organic
solvents, this value is known to decrease (in DMSO, pKa ) 9.00). Bordwell,
F. G. Acc. Chem. Res. 1988, 21, 456. With DBU, on the other hand, the
exact opposite is observed. In water, the pKa of DBU has been measured
at 11.0. In DMSO, this value rises to 12.0, but a recent investigation of
this amidine in acetonitrile has produced a pKa of 24. Kalijurand, I.; Rodima,
T.; Leito, I.; Koppel, I. A.; Schwesinger, R. J. Org. Chem. 2000, 65, 6202.
(18) Jiang, N.; Wang, J. Tetrahedron Lett. 2002, 43, 1285.
Unsymmetrical bisarylethynylenes were also prepared
using a one-pot base toggle protocol (Scheme 2; Table 3).
Addition of benzene to a flask containing 6 mol % PdCl2-
(19) Li, C.-J.; Chen, D.-L.; Costello, C. W. Org. Process Res. DeV. 1997,
1, 325.
(20) 4-Bromobenzonitrile gave only a trace (<1%) of the bisarylethyne
product, and complete recovery of starting material was observed for
4-bromo-N,N-dimethylaniline and 4-bromobiphenyl. Other unsuccessful
attempts suggest that a detrimental single-electron-transfer pathway may
take precedence over cross-coupling in certain substrates. Two iodides we
investigated (1-iodo-4-nitrobenzene and 1-iodo-3,5-dinitrobenzene) and two
bromides (4-bromobenzophenone and 5-bromopyrimidine) gave no discern-
ible products when subjected to our reaction conditions.
Org. Lett., Vol. 4, No. 19, 2002
3201