investigation by the silaboration of (E,E)-5,7-dodecadiene
using 2 and different metal-ligand combinations.10 With Pt-
(0) and PPh3, PEt3, or P(OPh3) we observed little or no
conversion of the starting materials. Switching to Ni(0),
prepared by treatment of Ni(acac)2 with DIBALH, gave
immediate results. Although no reaction was observed using
PPh3 as the ligand, both PPh2Cy and PCy3 resulted in around
30% consumption of the starting silylborane whithin 24 h
at 80 °C. PEt3 turned out to give an almost complete
consumption of both starting materials, even though 2 equiv
of the diene were used. Instead of the expected 1:1 adduct,
two products, identified as 5-(dimethylphenylsilyl)-6-
dodecene (3) and 5-(4,4,5,5-tetramethyl-1,3,2-dioxaboro-
lane)-5,7-dodecadiene (4), were formed in a 1:1 ratio (crude
1H NMR). Allylsilane 3, the product of a formal hydrosily-
lation of the diene,11 was isolated as the essentially pure Z
isomer, whereas dienylborane 4 was obtained as a 5:1
mixture of the E,E and Z,E isomers (Scheme 1). The same
a suitable substrate as it is 1,4-disubstituted, symmetrical,
and readily available.14 When subjecting 5 to the same
reaction conditions as 1, the outcome was very similar
(Scheme 2). The silylborane was almost completely con-
Scheme 2. Addition of Silylborane 2 to 2,4-Hexadienea
a Isolated yields, based on starting silylborane 2.
sumed, forming 6 and 7 in good yields and in almost equal
amounts (crude 1H NMR). The allylsilane 7 was formed with
high Z selectivity,15 whereas 6 was obtained as a 1.25:1
mixture of Z,E and E,E isomers. It is notable that the
unsubstituted double bond has pure E configuration, although
the starting 2,4-hexadiene consisted of a mixture of isomers.
We did not observe any products from normal silaboration.
At 60 °C the reaction was severely retarded and no
conversion of the starting silylborane was observed (1H
NMR).
Scheme 1. Addition of Silylborane 2 to
(E,E)-5,7-Dodecadienea
When the sterically demanding (E,E)-2,2,7,7-tetramethyl-
3,5-octadiene was subjected to the standard conditions, no
product formation or disapperance of starting materials was
observed, and the starting diene could be recovered by
chromatography. We next turned to the 1,3- and 1-substituted
dienes 9 and 14. Compound 9 has previously been silaborated
using ligand-free conditions, giving two 1,4-addition products
(11 and its regioisomer) in a 2:1 ratio.5,16 Using Ni(0)/PEt3
we obtained a mixture of at least four different products:
allylsilane 10 (>97:3 Z/E ratio), 1,4-silaboration product 11
(Z stereochemistry), 1,2-silaboration product 12 (E stereo-
chemistry), and dienylborane 13.
The effect of ligand structure on the distribution of
procucts obtained from 9 was studied. As can be seen from
Table 1 the ratio between the two silaboration products 11
and 12 is heavily influenced by the ligand. Using the small
and electron-rich PPhMe2, we obtained the two products in
a 1:1.3 ratio, whereas the bulkier and more electron-deficient
PPh3 and PPh2Cy only gave 11. The ratio between allylsilane
10 and silaboration products 11 and 12 was less affected by
the choice of ligand. It varied between 1.16 (entry 2) and
0.71 (entry 1). Compounds 11-13 constituted the major
products from the reaction; although they account for most
of the silicon added, the fate of a large part of the boron is
unknown.17 A 1:1 correspondence between allylsilane and
dienylborane could thus not be verified.
a Isolated yields, based on starting silylborane 2.
products were obtained when Ni(COD)2 was employed in
place of in situ formed Ni(0).
This type of disproportionation does not seem to have been
previously observed in element-element additions to 1,3-
dienes. However, hydrogen transfer was shown to accompany
the Rh-catalyzed addition of dialkyl disulfides to 1,2-dienes,
resulting in 1:1 mixtures of (E)-alkylthio-1,3-dienes and (E)-
2-alkylthio-2-alkenes.12 In addition, during dimerization-
double stannation of 1,3-dienes, hydrostannation was ob-
served when bulky distannanes were employed, but no
observation of stannyldienes was reported.13
These intriguing results prompted us to investigate the
generality of this disproportionation reaction. 2,4-Hexadiene
(5), which previously resisted silaboration,5 was selected as
(9) Silylborane and diene (2 equiv) were reacted in the presence of Ni-
(acac)2 (5 mol %) and DIBALH (10 mol %) in toluene at 80 °C. See ref 5.
(10) General Procedure. Silylborane 2, diene (2 equiv), ligand (10 mol
%), and M(acac)2 (5 mol %) were dissolved in toluene. DIBALH (10 mol
%) was then added at -35 °C to reduce the metal. The reaction mixture
was heated at 80 °C (Ni) or 110 °C (Pt) for 24 h.
(11) For reviews on hydrosilylation of 1,3-dienes, see: (a) Pietruzka, J.
In Science of Synthesis; Fleming, I., Ed.; Georg Thieme Verlag: Stuttgart,
2002; Vol. 4, pp 172-176. (b) ComprehensiVe Handbook on Hydrosily-
lation; Marciniec, B., Ed.; Pergamon: Oxford, 1992; pp 110-114. (c)
Ojima, I. In The Chemistry of Organic Silicon Compounds; Patai S.,
Rappaport, Z., Eds.; Wiley Interscience: New York 1989; Vol. 2, pp 1493-
1499.
(14) 2,4-Hexadiene, tech., 90%, mixture of isomers. Remaining 10% is
positional double bond isomers.
(15) As the starting 2,4-hexadiene was only 90% pure the selectivity of
the reaction is most probably higher than 93:7. We were not able to separate
7 from the isomeric impurities.
(16) Suginome, M.; Ito, Y. J. Organomet. Chem. 2003, 680, 43-50.
(17) Entry 2: compounds 10 + 11 + 12 ) 89%, while 11 + 12 + 13
) 53%.
(12) Arisawa, M.; Suwa, A.; Fujimoto, K.; Yamaguchi, M. AdV. Synth.
Catal. 2003, 345, 560-563.
(13) Tsuji, Y.; Kakekhi, T. J. Chem. Soc., Chem. Commun. 1992, 1000-
1001.
2930
Org. Lett., Vol. 8, No. 14, 2006