Organic Letters
Letter
a
Watson and co-workers found that the ligand effect could
control the stereoselectivity of palladium-catalyzed carbosily-
lation of internal symmetrical alkynes with trisubstituted silyl
iodides and primary alkyl zinc iodides. While the use of
(3,5-tBu2C6H3)3P (DrewPhos) led to syn selectivity,
(3,5-tBu2C6H3)2P(tBu) (JessePhos) gave rise to anti selectivity
(Scheme 1d).4c Similarly, syn-selective carbosilylation of
internal alkynes has been achieved via palladium-/copper-
cocatalyzed tandem silylboration/Suzuki reactions with a new
silane donor nucleophile (PhMe2SiBPin) and aryl bromide
electrophiles using the PCy3 ligand, wherein the PhMe2Si-[Cu]
intermediates are first formed and then directly added syn
across the CC bonds (Scheme 1e).5a In contrast, when a
similar Pd/Cu cooperative catalysis is used, the presence of
PPh3 enables the anti-selective silylallylation of electron-poor
internal alkynes with silyl boronic esters and allylcarbonates,
whereas omission of the ligand led to the syn-selective version
(Scheme 1f).5b Iron-catalyzed anti-selective silylalkylation of
terminal aliphatic alkynes with silyl boronic esters and alkyl
iodides occurred through direct addition of the in situ
generated R3Si-[Fe] intermediates across the CC bond
(Scheme 1g).5c Nevertheless, all existing approaches have been
limited to alkyne silyl-carbofucntionalization and only two
papers to date allowed the formation of functionalized anti-
alkenes from electron-deficient internal alkynes and terminal
aliphatic alkynes. A precise regioselective sequence control
between the silylation and carbofunctionalization processes to
realize carbosilylation of alkynes, especially including non-
symmetric terminal and internal alkynes, still remains
unprecedented.
We hypothesized that if more highly reactive carbon donors
were used as the electrophiles and the nucleophilicity of the in
situ generated PhMe2Si-[Cu] intermediates were tuned, the
palladium-catalyzed addition of the more highly reactive
carbon donors across the CC bond would take precedence
over the direct addition of the in situ generated PhMe2Si-[Cu]
intermediates, thus enabling exquisite control over both
regioselectivity and stereoselectivity to create new methods.
Herein, we report a new, general dual palladium-/copper-
catalyzed intermolecular trans-allenyl silylation of terminal
alkynes with propargyl acetates and PhMe2SiBpin to produce
(E)-silyl enallenes (Scheme 1h). The method allows propargyl
acetates as highly reactive allene precursors6 to enable
allenylation with alkynes prior to silylation and provides a
conceptually novel alkyne carbosilylation route to forge new,
valuable, functionalized vinyl silane scaffolds7 with excellent
regio-/stereoselectivity and functional group tolerance.
Table 1. Screening of Optimal Reaction Conditions
entry
variation from the standard conditions
none
yield (%)
1
72
0
trace
60
59
2
3
4
5
6
7
8
9
without Pd(PPh3)2Cl2, CuF2, or Na2CO3
PdCl2 instead of Pd(PPh3)2Cl2
PdCl2/PPh3 instead of Pd(PPh3)2Cl2
Pd(dppp)Cl2 instead of Pd(PPh3)2Cl2
Pd(dppf)Cl2 instead of Pd(PPh3)2Cl2
Pd(PPh3)4 instead of Pd(PPh3)2Cl2
Cu(OAc)2 instead of CuF2
CuCl instead of CuF2
10
trace
<5
40
25
41
10
11
CuI instead of CuF2
CuCl2 instead of CuF2
CuCl2 instead of CuF2
none
b
12
c
69
67
13
a
Standard reaction conditions unless specified otherwise: 1a (0.4
mmol), 2a (0.2 mmol), Me2(Ph)SiBPin (0.4 mmol), Pd(PPh3)2Cl2
(10 mol %), CuF2 (20 mol %), Na2CO3 (0.6 mmol), DMF (2 mL),
argon, 90 °C, 12 h. Only the E isomer was obtained, which was
determined by a 1H NMR and/or GC-MS analysis of the crude
b
c
product. NaF (40 mol %). 2a (1 mmol), DMF (4 mL), 20 h.
CuCl2, exhibited reactivity, but they were less efficient than
CuF2 (entries 8−11). Interestingly, using CuCl2 combined
with NaF gave results identical with those of CuF2 (entry 12),
suggesting that fluoride ions may promote Si−B bond
cleavage.7,8 Screening the effect of the CuF2 amount, bases
(such as Na2CO3, NaHCO3, K2CO3, and NEt3) and
temperatures (entries 1 and 13−18 in Table S1 in the
Supporting Information) proved that the reaction with 20 mol
% of CuF2 and Na2CO3 at 90 °C was the best option (entry 1).
Intriguingly, the conditions were applicable to a 1 mmol scale
of 2a, giving 3aa in 67% yield (entry 13).
With the optimized conditions in hand, we next explored the
reliable leaving groups in propargyl acetates 2 for the alkyne
allenylsilylation protocol (Scheme 2). A wide range of acyl
Scheme 2. Optimization of the Reliable Leaving Groups in
the Propargyl Acetates 2
Initial investigations on the allenylsilylation of 1-ethynyl-4-
fluorobenzene (1a) with 2-methyl-4-phenyl-but-3-yn-2-yl
acetate (2a) and Me2(Ph)SiBPin were performed (Table 1
and Table S1 in the Supporting Information). Allenylsilylation
of alkyne 1a with propargyl acetate 2a, Me2PhSiBPin (3a), 10
mol % of Pd(PPh3)2Cl2, 20 mol % of CuF2, and Na2CO3 in
DMF was executed smoothly, producing the desired product
(E)-3aa with a 72% yield (entry 1). It should be noted that
these parameters, including Pd(PPh3)2Cl2, CuF2, Na2CO3, and
phosphorus ligands (such as Ph3P, dppp, and dppf), are all
equally crucial for the success of the reaction, since omission of
each of them led to no isolation of 3aa (entries 2 and 3). While
using PdCl2 (entry 3) or Pd(PPh3)4 (entry 7) delivered a trace
amount of 3aa, other Pd catalysts, including PdCl2/PPh3,
Pd(dppp)Cl2, and Pd(dppf)Cl2, were highly reactive (entries
4−6). Other Cu catalysts, such as Cu(OAc)2, CuCl, CuI, and
groups, including Boc (2c), PhCO (2d), 4-MeC6H4CO (2e),
and 4-CF3C6H4CO (2f), were suitable to furnish 3aa, although
they had lower reactivity. However, using either bulky BuCO
(2b) or CF3SO2 (2g) led to no conversion of 2. Likewise, 2-
methyl-4-phenylbut-3-yn-2-ol (2h) was inert.
Encouraged by these results, we investigated the substrate
scope (Scheme 3). This alkyne allenylsilylation was applicable
to various substituents, such as F, Cl, Br, CF3, MeO, Ph, and
t
6554
Org. Lett. 2021, 23, 6553−6557