Silyl-cupration of a propargylsilane as a test for the reversibility of a metallo-metallation

Article abstract of DOI:10.1016/S0022-328X(00)00838-X

Silyl-cupration of propargylsilanes is normal in giving mainly the product with a silyl group at the terminus and the copper atom in the centre. There is no evidence of the copper atom and the originally-resident silyl group undergoing a retro silyl-cupration to give an allenylsilane.

Full text of DOI:10.1016/S0022-328X(00)00838-X

Journal of Organometallic Chemistry 624 (2001) 69–72
www.elsevier.nl/locate/jorganchem
Silyl-cupration of a propargylsilane as a test for the reversibility of
a metallo-metallation
Maria Angeles Cubillo de Dios, Ian Fleming *, Wibke Friedhoff, Philip D.W. Woode
Department of Chemistry, Lensfield Road, Cambridge CB2 1EW, UK
Received 21 August 2000; accepted 21 September 2000
Abstract
Silyl-cupration of propargylsilanes is normal in giving mainly the product with a silyl group at the terminus and the copper
atom in the centre. There is no evidence of the copper atom and the originally-resident silyl group undergoing a retro
silyl-cupration to give an allenylsilane. © 2001 Elsevier Science B.V. All rights reserved.
Keywords: Metallo-metallation; Silyl-cupration; Alkynes; Allenylsilane; Propargylsilane
1
. Introduction
almost all other electrophiles, including quite weak ones
like epoxides, reacted cleanly with the isomer 2. In this
paper we report our efforts to examine a system that
might be expected to show the reverse step 3“1+al-
lene, if it is a feasible pathway (Scheme 1).
We already knew that silyl-cupration of terminal
acetylenes was highly regioselective, placing the silyl
group cleanly at the terminus and the copper atom on
the inside [2]. We reasoned that if we were to carry out
this reaction on the propargylsilane 6, we would obtain
the intermediate 7 (Scheme 2). This intermediate has
critically the same feature as the intermediate 3, namely
a silyl group at the tetrahedral terminus and a copper
atom on the central carbon atom (7, lasso). If it were to
We showed some years ago that the silyl-cuprate
reagent 1 reacted with allene to give the intermediate 2,
in which the silyl group was attached to the central
atom and the copper to the terminus [1], as shown by
its reaction with almost all electrophiles, including chlo-
rine, to give vinylsilanes 4. There were however two
extraordinary exceptions. When the electrophile was
bromine, the product was a mixture of the vinylsilane 4
(
E=Br) and the allylsilane 5 (E=Br), and when the
electrophile was iodine the product was exclusively the
allylsilane 5 (E=I). The simplest explanation for these
results was that the intermediates 2 and 3 were in rapid
equilibrium, and that the silyl-cupration, an example of
a metallo-metallation, was easily reversible, as some
other metallo-metallations are. Our efforts to prove this
at that time were unsuccessful. In particular, although
we were able to reassemble separately cuprates with the
stoichiometry of each of the intermediates 2 and 3,
starting with the iodides 4 (E=I) and 5 (E=I), they
did not react in the same way as the intermediate(s)
produced directly from the silylcuprate and allene. Even
if reversibility is the explanation, it still remains a
puzzle why iodine, surely a powerful electrophile,
would select the regioisomer 3 to react with, when
*
Corresponding author.
E-mail address: if10000@cam.ac.uk (I. Fleming).
Scheme 1.
0
022-328X/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S0022-328X(00)00838-X

7
0
M. Angeles Cubillo de Dios et al. / Journal of Organometallic Chemistry 624 (2001) 69–72
Scheme 4.
reaction mixture at −78°C with aqueous ammonium
chloride solution, cleanly the allylsilane 12. However,
when we allowed the reaction mixture to warm to room
temperature before quenching, the allylsilane 12 was
accompanied by its regioisomer 10 (Scheme 4). This
isomer might have been produced by either of two
pathways. One is the simple reversal of the first step,
and readdition to the acetylene group with the opposite
regiochemistry, the other is that it might have been the
reaction we were looking for, a retro metallo-metalla-
tion in the direction 7“8, followed by readdition,
although we would have to accept, since there was no
sign of the vinylsilane 11, that the protonation step had
been more highly regioselective than it had been when
we had made the intermediate 9 deliberately. The re-
versibility in the reaction with the acetylene was not
something that had been noticed before, and indeed a
simple acetylene, oct-1-yne, gave only the terminal
vinylsilane no matter what temperature the mixture was
quenched at. This made it seem a little more possible
that the product 10 was not produced by the first
pathway. Accordingly, we carried out an experiment
that would distinguish between the two pathways.
Silyl-cupration of propargyltrimethylsilane 13 simi-
larly gave only the allylsilane 14 when the mixture was
quenched at −78°C, and a mixture of the allylsilane 14
and its regioisomer 15 when it was quenched at room
temperature. The critical detail is that the product 15
has the trimethylsilyl group in the allylic position and
the phenyldimethylsilyl group on the central atom, as
shown by protodesilylation 15“16, which cleanly re-
moved the trimethylsilyl group. If the allene had been
formed, the by-product analogous to 15 would have
had the silyl groups interchanged. This product was
therefore the result of reversibility in the addition to the
acetylene, and the allene was not being formed (Scheme
5).
Scheme 2.
Scheme 3.
undergo a retro metallo-metallation it would give the
allene 8 and the silylcuprate, and these might then
combine in the opposite sense to give the regioisomer 9.
2
. Results and discussion
First we checked what would happen if the allene 8
and the silylcuprate were to react with each other. We
have just suggested that it would give the intermediate
, and this is indeed what happened — the product
from an aqueous work-up was a mixture of the two
vinylsilanes 10 and 11, in ratios that varied with the
temperature of the quench, but both were always
present (Scheme 3).
We then carried out the critical silyl-cupration of the
propargylsilane 6, and obtained, after quenching the
9
In conclusion, we have shown once again that the
unusual regioselectivity when iodine is used to quench
the product from the silyl-cupration of allene is proba-
bly not explained simply by the reversibility of the

M. Angeles Cubillo de Dios et al. / Journal of Organometallic Chemistry 624 (2001) 69–72
71
silyl-cupration step. It remains something of a mystery,
to which we now add the mystery that whereas the
silyl-cupration of terminal acetylenes is highly regiose-
lective at any temperature between −78°C and room
temperature, the corresponding reaction with propar-
gylsilanes is evidently not.
and chlorodimethyl(phenyl)silane (8 ml) added slowly
by syringe. The mixture was stirred overnight at r.t. A
saturated solution of ammonium chloride was added at
0°C, followed by diethyl ether. The ether layer was
dried (MgSO ), evaporated off, and the residue chro-
4
matographed [SiO , light petroleum (bp 40–60°C)] to
2
give propargyldimethyl(phenyl)silane [3] (8.0 g, 82%);
l (250 MHz; CDCl ) 7.65–7.55 (2 H, m, Ph), 7.46–
H
3
7
.32 (3 H, m, Ph), 1.87 (1 H, t, J 2.9, CꢀCH), 1.74 (2 H,
3. Experimental
d, J 2.9, CH Si) and 0.42 (6 H, s, SiMe ) and al-
2
2
lenyldimethyl(phenyl)silane [3] (1.2 g, 12%); l (250
H
NMR spectra were recorded on Bruker AM400 and
MHz; CDCl ) 7.6–7.5 (2 H, m, Ph), 7.45–7.15 (3 H, m,
3
AC250 instruments with chemical shifts measured rela-
tive to chloroform l7.25. Coupling constants are ex-
pressed in Hertz. The IR spectrum was recorded on a
Perkin–Elmer FT-IR 1620 machine using sodium chlo-
ride plates. Column chromatography was carried out
using Merck Kieselgel 60 (230–400 mesh ASTM) and
TLC using glass plates coated to 1 mm with Kieselgel
0 PF254. Tetrahydrofuran (THF) and ether were
freshly distilled from lithium aluminium hydride under
argon.
Ph), 5.06 (1 H, t, J 7, CꢁCHSi), 4.40 (2 H, d, J 7, ꢁCH2)
and 0.40 (6 H, s, SiMe2).
3.2. Silyl-cuprations (general procedure)
Typically, the alkyne or allene (1 mmol) in THF (3
ml) was injected into a stirred solution of the silyl-
cuprate reagent [4] (1.1 mmol) in THF under argon at
−78°C and the mixture stirred for 1 h. The mixture
was quenched at −78°C with saturated aqueous am-
monium chloride solution (10 ml) and allowed to warm
to r.t. After 2 h the solution was extracted with diethyl
ether (3×10 ml). The combined organic layers were
washed with distilled water (3×10 ml), and brine (2×
6
3
.1. Preparation of propargyldimethyl(phenyl)silane (6)
and allenyldimethyl(phenyl)silane (8)
Magnesium turnings (1.2 g, 41 mmol) were covered
with dry ether (5 ml) under argon, and mercury(II)
chloride (0.02 g) was added followed by propargyl
bromide (0.5 g, 4.2 mmol) in a single portion. After 1.3
h, the reaction started (the mixture became dark and
started to boil). The flask was cooled with water at
room temperature (r.t.) and propargyl bromide (4.5 g,
1
0 ml), dried (MgSO ), and evaporated under reduced
4
pressure. Chromatography [SiO , light petroleum (bp
2
3
0–40°C)] gave the products.
3.3. Silyl-cupration of the allenylsilane 8
4
6 mmol) in dry ether (35 ml) was added via cannula at
Silyl-cupration of allenyldimethyl(phenyl)silane (0.2
g, 1.15 mmol) gave a 1:1 mixture (0.163 g, 46%) of
2,3-bis[dimethyl(phenyl)silyl]-1-propene (10), showing
a rate to maintain the internal temperature at r.t. The
mixture was then stirred at r.t. for 12 h, cooled to 0°C
1
the signals ( H-NMR) identical to those of the pure
sample
prepared
below,
and
E-1,2-bis[-
dimethyl(phenyl)silyl]-1-propene (11) with diagnostic
signals at l (250 MHz; CDCl ) 6.31 (1 H, q, J 1,
H
3
ꢁ
4
CHSi), 1.84 (3 H, d, J 1, ꢁCMe) and 0.45–0.20 (12 H,
×s, 2×SiMe Ph).
2
3.4. E-1,3-Bis[dimethyl(phenyl)silyl]-1-propene (12)
Silyl-cupration of propargyldimethyl(phenyl)silane
(
0.2 g, 1.15 mmol) gave the silane 12 (0.25 g, 69%);
l (250 MHz, CDCl ) 7.58–7.26 (10 H, m, 2×Ph), 6.08
H
3
(
1 H, dt, J 18.4. and 7.76, CH ꢁCH Si), 5.56 (1 H, dt,
A
B
J 18.4 and 1.2, CH ꢁCH Si), 1.91 (2 H, dd, J 7.76 and
1
A
B
.2, CH Si) and 0.33 (12 H, s, 2×SiMe Ph); l (100
2 2 C
MHz, CDCl ) 145.3+, 139.6+, 138.5−, 133.9+,
3
1
2
ꢁ
33.7+, 129.1+, 128.8+, 127.8+, 127.7+, 126.6+,
7.8−, −2.22+ and −3.34+ (+ꢁCH or CH , −
3
+
C or CH ); m/z (EI) 310 (29%, M ), 295 (13, M−
2
Scheme 5.
Me), 175 (8, M−SiMe Ph), 160 (74, M−Me−
2

7
2
M. Angeles Cubillo de Dios et al. / Journal of Organometallic Chemistry 624 (2001) 69–72
+
SiMe Ph) and 135 (100) SiMe Ph) (Found: M ,
3.8. 2-Dimethyl(phenyl)silylpropene (16)
2
2
3
10.1572. C H Si requires M, 310.1573).
19 26 2
The mixture of silyl-cupration products 14 and 15
was stirred with trifluoroacetic acid (three equivalents)
in dichloromethane (2 ml) at r.t. for 3 h, conditions that
completely protodesilylated the isomer 14. The mixture
was washed with sodium carbonate solution (2 N, 10
3
.5. Silyl-cupration of the propargylsilane 6 with
warming to room temperature before quenching
ml), dried (MgSO ) and evaporated under reduced pres-
4
The experiment was repeated, but the mixture was
allowed to warm to r.t. and stirred for 1 h before being
quenched with aqueous saturated ammonium chloride
solution. The standard work-up gave the silanes 12
sure to give only the vinylsilane 16; l (250 MHz;
H
CDCl ) 7.67–7.48 (2 H, m, Ph), 7.43–7.28 (3 H, m,
3
Ph), 5.69 (1 H, q, J 1.3, CH H ꢁCSi), 5.35 (1H, q, J
A
B
1
.3, CH H ꢁCSi), 1.84 (3 H, t, J 1.3, CMe) and 0.42 (6
A
B
(
(
48%) and 2,3-bis[dimethyl(phenyl)silyl]-1-propene (10)
32%); l (400 MHz; CDCl ) 7.57–7.25 (10 H, m, Ph),
1
H, s, SiMe Ph) identical ( H-NMR) with the known
2
H
3
compound [1].
5
ꢁ
4
1
1
.53 (1 H, d, J 2.6, ꢁCH CH ), 5.34 (1 H, d, J 2.6,
A
B
CH CH ), 1.86 (2 H, s, CH ) and 0.33–0.20 (12 H,
A
B
2
3.9. E-1-Dimethyl(phenyl)silyloctene
×s, 2×SiMe Ph); l (100 MHz, CDCl ) 146.8−,
2
C
3
39.4−, 138.3−, 134.1+, 133.7+, 129.7+, 129.5+,
28.1+, 127.9+, 125.7−, 23.85−, −2.64+ and
1
-Octyne (0.11 ml, 0.8 mmol) was injected over a
stirred solution of the silyl-cuprate reagent (0.8 mmol)
under argon at −78°C and the mixture stirred for 1 h.
The mixture was allowed to warm to r.t. and stirred for
−
3.02+ (+ꢁCH or CH , −ꢁC or CH ).
3
2
1
h before being quenched with aqueous saturated
ammonium chloride solution (10 ml). The standard
3
1
.6. E-1-[Dimethyl(phenyl)silyl]-3-(trimethylsilyl)-
-propene (14)
aqueous work-up gave only the 1-silyloctene; R [light
f
petroleum (bp 30–40°C)] 0.51; l (250 MHz; CDCl )
H
3
7
.55–7.25 (5 H, m, Ph); 6.11 (1 H, dt, J 18 and 6,
CHꢁCHSi), 5.74 (1 H, dt, J 18 and 1.6, CHꢁCHSi),
2.15 (2 H, m, CH CHꢁCHSi), 1.43–1.20 (8 H, m,
×CH ), 0.88 (3 H, t, J 7, CH Me) and 0.3 (6 H, s,
Silyl-cupration of propargyltrimethylsilane (0.07 ml,
0
3
1
1
.5 mmol) gave the silane 14; R [light petroleum (b.p.
f
1
−
2
0–40°C)] 0.47; wmax(film)/cm
1950 (Ph), 1877 (Ph),
4
2
2
813 (Ph), 1603 (CꢁC), 1426 (SiMe ), 1250 (SiMe ) and
2
3
SiMe ), closely similar to the lower homologue (E)-1-
2
113 (SiPh); l (400 MHz; CDCl ) 7.58–7.48 (2 H, m,
H
3
dimethyl(phenyl)silylhexene [2] and the higher homo-
logue (E)-1-dimethyl(phenyl)silyltridecene [7].
Ph), 7.43–7.26 (3 H, m, Ph), 6.16 (1 H, dt, J 18.4 and
.8, CH ꢁCH Si), 5.56 (1 H, dt, J 18.4 and 1.2,
7
A
B
CH ꢁCH Si), 1.69 (2 H, dd, J 7.8 and 1.2, CH Si), 0.33
A
B
2
(
6 H, s, SiCMe Ph) and 0.03 (9 H, s, SiMe ); l (100
2 3 C
Acknowledgements
MHz; CDCl ) 146+, 139.8−, 133.9+, 128.8+,
3
1
27.8+, 125.5+, 28.8−, −2.1+ and −3.9+ (+
We thank Ajay Mandal, Amit Mandal, Kah Ling
Pang and Matthew Russell for supervising this under-
graduate project, and the ERASMUS scheme for a
maintenance grant (W. Friedhoff).
1
ꢁ
CH or CH , −ꢁC or CH ), identical ( H-NMR) with
3
2
the known compound [5].
3
.7. Silyl-cupration of the propargylsilane 13 with warm-
References
ing to room temperature before quenching
[
1] I. Fleming, M. Rowley, P. Cuadrado, A.M. Gonzalez-Nogal, F.J.
Pulido, Tetrahedron 45 (1989) 413.
The experiment was repeated, but the mixture was
allowed to warm to r.t. and stirred for 1 h before being
quenched with aqueous saturated ammonium chloride
solution. After 2 h, the solution was worked up as
before to give a mixture (3:1) of the silanes 14 and 15;
l (400 MHz; CDCl ) showed the signals of the allylsi-
[2] I. Fleming, T.W. Newton, F. Roessler, J. Chem. Soc. Perkin Trans.
(1981) 2527.
[
1
3] (a) A.K. Aliev, S.F. Karaev, K.G. Gasanov, Nauch. Tr. Azerb.
In-ta Nefti i Khimii (1979) 59; Chem. Abstr. 93 (1980) 204727x. (b)
M. Kira, T. Taki, H. Sakurai, J. Org. Chem. 54 (1989) 5647.
H
3
[4] I. Fleming, in: L.A. Paquette (Ed.), Encyclopedia of Reagents for
Organic Synthesis, Wiley, Chichester, 1995.
[
[
lane 14 together with the following diagnostic signals
for 2-dimethyl(phenyl)silyl-3-trimethylsilyl-1-propene
15) [6]: l (250 MHz; CDCl ) 5.55 (1 H, br d, J 2.8,
5] I. Fleming, J.A. Langley, J. Chem. Soc. Perkin Trans. 1 (1981) 1421.
6] Y. Morizawa, H. Oda, K. Oshima, H. Nozaki, Tetrahedron Lett.
(
H
3
2
5 (1984) 1163.
ꢁ
CH H ), 5.31 (1 H,br d, J 2.8, ꢁCH CH ), 1.64 (2 H,
A
B
A
B
[7] I. Fleming, N.J. Lawrence, J. Chem. Soc. Perkin Trans. 1 (1998)
2679.
d, J 1 CH Si) and −0.1 (9 H, s, SiMe ).
2
3
.