On the mechanism of the reductive metallation of asymmetrically substituted silyl chlorides

Article abstract of DOI:10.1002/ejoc.200400622

An investigation of the stereochemical course of the reductive metallation of silyl chlorides with silicon-centred chirality has revealed two major events which are detrimental to stereoselection during silyl anion formation: (1) chloride-induced racemisation of silyl chlorides and (2) nonstereoselective formal dimerisation during metallation providing the corresponding disilane. In control experiments, the stereochemical course of these processes has been independently verified for the reductive metallation of the enantioenriched cyclic silyl chloride (SiS)-7a (R = H, er ≥ 88:12). A screening of several related derivatives of (SiS)-7a led to the sterically encumbered silyl chloride (SiR)-7c (R = iPr, er ≥ 94:6) which displays some unique features. This structural modification prevents racemisation by lithium chloride (T < -40 °C) as well as dimerisation (T < -100 °C) thus allowing for the first generation of an asymmetrically substituted silyl anion (SiS)-8c (er = 74:26) by reductive metallation of a silyl chloride with silicon-centred chirality. Moreover, the enantiospecificity of the preparation of (SiA)-7c by chlorination [(SiS)-9c → (SiR)-7c] and its reduction with aluminium hydrides [(SiR)-7c → (SiR)-9c] have been unambiguously determined by X-ray crystallography as retention (≥99%) and inversion (≥99%), respectively. ( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2005).

Full text of DOI:10.1002/ejoc.200400622

FULL PAPER
On the Mechanism of the Reductive Metallation of Asymmetrically Substituted
Silyl Chlorides
Martin Oestreich,*[a] Gertrud Auer,[a] and Manfred Keller[a]
Keywords: Silicon / Lithiation / Radicals / Enantioselectivity / Asymmetric synthesis
An investigation of the stereochemical course of the reduct-
ive metallation of silyl chlorides with silicon-centred chirality
has revealed two major events which are detrimental to ster-
eoselection during silyl anion formation: (1) chloride-induced
racemisation of silyl chlorides and (2) nonstereoselective
vents racemisation by lithium chloride (T Ͻ −40 °C) as well
as dimerisation (T Ͻ −100 °C) thus allowing for the first gen-
eration of an asymmetrically substituted silyl anion (SiS)-8c
(er = 74:26) by reductive metallation of a silyl chloride with
silicon-centred chirality. Moreover, the enantiospecificity of
formal dimerisation during metallation providing the corres- the preparation of (SiR)-7c by chlorination [(SiS)-9c Ǟ (SiR)-
ponding disilane. In control experiments, the stereochemical
course of these processes has been independently verified for
the reductive metallation of the enantioenriched cyclic silyl
chloride (SiS)-7a (R = H, er Ն 88:12). A screening of several
related derivatives of (SiS)-7a led to the sterically encum-
bered silyl chloride (SiR)-7c (R = iPr, er Ն 94:6) which dis-
7c] and its reduction with aluminium hydrides [(SiR)-7c Ǟ
(SiR)-9c] have been unambiguously determined by X-ray
crystallography as retention (Ն99%) and inversion (Ն99%),
respectively.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
plays some unique features. This structural modification pre- Germany, 2005)
Introduction
This long-standing problem was independently solved by
Kawakami[8] and Strohmann[9] by way of two conceptually
different strategies, namely (1) chemo-unselective
tinϪlithium exchange of chiral silyl stannane 1 (1 Ǟ 6,
Scheme 1)[8] and (2) reductive metallation of the chiral disil-
ane 3 using lithium metal (3 Ǟ 6, Scheme 1).[9] However,
these fundamental approaches are afflicted with an intrinsic
shortcoming which complicates synthetic applications since
(almost) equimolar amounts of another organometallic 2
and 4, respectively, are formed as by-products.
A straightforward methodology from a synthetic view for
the generation of stereochemically and chemically uniform
silyllithiums is therefore still not available. Herein, we wish
to report our efforts towards accessing enantioenriched
silyllithiums 6 from chiral silyl chlorides 5 by simple re-
ductive metallation using lithium metal (5 Ǟ 6, Scheme 1).
We anticipated that this approach would produce the silyl
anion 6 along with lithium chloride as the only by-product.
It should be noted that Kawakami has described the re-
ductive metallation of an acyclic enantioenriched silyl
chloride with silicon-centred chirality in a recent publi-
cation. The asymmetrically substituted silicon centre under-
went complete racemisation during metallation, providing
the silyl anion in racemic form.[8]
Silyl metal compounds have gained substantial import-
ance as silicon transfer reagents in synthetic organic, or-
ganometallic and polymer chemistry.[1Ϫ3] These organomet-
allics frequently originate from the requisite silyllithium
which in turn is easily accessible by reductive metallation of
the corresponding silyl chloride.[4] Whereas achiral silyl-
lithium compounds are well-established,[1] the chemistry of
asymmetrically substituted, tetracoordinate silyllithium
species has attracted relatively little attention. Nevertheless,
metallated silyl anions with silicon-centred chirality are at-
tractive organometallics with regards to their remarkably
high configurational stability,[5] potential synthetic
applications[1Ϫ3] and mechanistic properties.[6]
Until recently, the enantioselective preparation of asym-
metrically substituted silyl anions remained an open chal-
lenge in organosilicon chemistry. In his seminal work,
Sommer succeeded in the reductive cleavage of an enantio-
merically enriched disilane with lithium metal thus provid-
ing the silyl anion with partial racemisation.[7a] Similar ob-
servations were made by Corriu who accessed an optically
active silyl anion via a cobaltϪlithium exchange reac-
tion.[7b]
Although reductive lithiation of silyl chlorides is rou-
tinely employed for silyl anion generation, its mechanism
has not been elucidated so far.[1] Occasional reports have
indicated the subtle nature of the reductive metallation of
triorganosilyl chlorides R3SiCl[10] and amino-substituted
[a]
Institut für Organische Chemie und Biochemie, Albert-
Ludwigs-Universität Freiburg,
Albertstrasse 21, 79104 Freiburg im Breisgau, Germany
Fax: (internat.) ϩ 49-761-203-6100
E-mail: martin.oestreich@orgmail.chemie.uni-freiburg.de
184
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejoc.200400622
Eur. J. Org. Chem. 2005, 184Ϫ195

Reductive Metallation of Asymmetrically Substituted Silyl Chlorides
FULL PAPER
Scheme 1. Strategies for the enantioselective preparation of asymmetrically substituted silyl anions (R1 ϶ R2 ϶ R3)
silyl chlorides (Et2N)nPh3ϪnSiCl (n ϭ 1 and 2)[11] but with- (Scheme 2). Transfer of an electron into the σ*(SiϪCl) or-
out a detailed mechanistic understanding. Successful lithi- bital of A provides silyl radical B and a chloride anion (A
ation of R3SiCl usually requires at least one aryl group at Ǟ A· Ǟ B). Radical intermediate B with silicon-centred
the silicon.[1a] A number of arylalkylsilyl chlorides having chirality is either prone to another electron capture forming
the general formula Ph3ϪnRnSiCl (R ϭ Me or tBu, n ϭ 0, D (B Ǟ D) or to racemisation via the vertex mechanism[14]
1, and 2) undergo reductive metallation in good yields.[1,10a] (B Ǟ C Ǟ ent-B). Theoretical data predict a pyramidal
Conversely, treatment of trimethylsilyl chloride with lithium configuration for silyl radicals with high inversion bar-
metal cleanly furnishes hexamethyldisilane rather than the riers[15] indicating that the second electron transfer (B Ǟ D)
desired silyllithium.[12] The peralkylated disilane is not re- might occur prior to enantiomerisation (B Ǟ ent-B). Seri-
ductively cleaved by elemental lithium or cognate reagents ous experimental data suggest the configurational stability
(LiN,[13a] LiDBB[13b] or LDMAN[13c]). However, metall- of silyl radicals.[16] Therefore, the intermediacy of silyl rad-
ation of Ph3ϪnRnSiCl (n ϶ 3) is known to proceed through icals during metallation might not influence enantioselec-
disilanes which are subsequently cleaved under those reac- tion.
tion conditions.
These observations suggest that one or more aryl groups
intervene in the reductive metallation of an intermediate
disilane (or the silyl chloride itself) since an aryl group at
silicon could potentially function as an electron acceptor.
Nevertheless, we assume that the initial single electron
transfer occurs into the σ*(SiϪCl) orbital (ϵ LUMO). This
hypothesis is consistent with experimental evidence that tri-
alkylsilyl as well as arylalkylsilyl chlorides give the corre-
sponding disilanes when treated with an electropositive me-
tal. The fate of the initially formed radical anion [R3SiCl]·Ϫ
is unknown but decomposition into a silyl radical R3Si· and
a chloride anion ClϪ seems plausible. The silyl radical,
R3Si·, in turn is reduced by another equivalent of the elec-
tropositive metal providing a silyl anion R3SiϪ which for-
mally dimerises together with unchanged silyl chloride.
Alternatively, recombination of two molecules of R3Si·
might also lead to the disilane intermediate. The mechanism
of cleavage of the resultant disilane remains completely un-
certain.
Scheme 2. Vertex inversion of silyl radicals (R1 ϶ R2 ϶ R3)
We selected cyclic silyl chloride (SiS)-7a (Figure 1) as a
Based on these considerations, which involve the interme-
diacy of silyl radicals,[14] we have devised a working hypoth-
esis (omitting disilane formation) for the reductive metall-
ation of asymmetrically substituted silyl chlorides A model substrate for two reasons, namely (1) chiral silyl chlo-
Eur. J. Org. Chem. 2005, 184Ϫ195
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
185

M. Oestreich, G. Auer, M. Keller
FULL PAPER
rides for which practical syntheses are known[17,18] usually silyl chlorides by chloride anions.[20] Corriu[18a,18b] and,
incorporate an α-naphthyl group (α-Np) which is an excel- later, Bassindale[21] presented potential mechanistic ration-
lent electron acceptor.[8] (2) We reported a convergent and ales for these nucleophile-assisted racemisations based on
modular reaction sequence for a flexible preparation of kinetic[21b] and thermodynamic[21c] measurements.[22] How-
chiral silanes with a 1,2,3,4-tetrahydro-1-silanaphthalene ever, these useful investigations have not provided a com-
core.[19] This allows for straightforward access to (SiS)-7a pletely clear picture of the mechanism but have demon-
and the related silanes rac-7b and (SiR)-7c (Figure 1).
strated the complexity of this process.[22,23]
As shown for the silyl chlorides 7, we speculate that api-
cal nucleophilic attack at silicon by chloride generates the
achiral pentacoordinate silicon species 10 and equatorial at-
tack provides 10Ј (Scheme 4). Apart from reversible associ-
ative and dissociative processes, these pentacoordinate
anions with a trigonal-bipyramidal geometry are prone to
configurational isomerisation (10/10Ј). A recent study by
Lammertsma elegantly showed that cyclic silanes undergo
Berry pseudorotation![24]
Figure 1. 1-Aryl-1-chloro-1,2,3,4-tetrahydro-1-silanaphthalenes 7
Results and Discussion
First Generation Approach
We were pleased to find that treatment of (SiS)-7a (er ϭ
96:4) with lithium metal under standard reaction conditions
resulted in formation of the desired silyllithium (SiS)-8a in
an isolated yield of 80% after protolysis (Scheme 3). Deter-
mination of the enantiomeric ratio after acidic workup
[(SiS)-8a Ǟ (SiS)-9a] showed, however, that (SiS)-8a was
nearly racemic. The slightly inverted absolute configuration
of the isolated silane (SiS)-9a (er ϭ 45:55) was rather puz-
zling assuming that protolysis proceeds with retention of
stereochemistry.
Scheme 3. Reductive metallation of (SiS)-7a
These results prompted us to further investigate the reac-
tion pathway(s) of the reductive metallation with particular sation of chiral silyl chlorides 7
Scheme 4. Plausible mechanisms for the chloride-induced racemi-
focus on the stereochemical consequences of two aspects,
i.e. (1) does the lithium chloride ‘‘by-product’’ promote the
racemisation of (SiS)-7a and (2) does this silyl anion forma-
tion [(SiS)-7a Ǟ (SiS)-8a, Scheme 3] proceed through a di-
silane?
In order to examine the chloride-induced racemisation
of (SiS)-7a, we observed (SiS)-7a with or without lithium
chloride in THF (Table 1). After 1 h, treatment of the reac-
tion mixture with lithium aluminium hydride[19] stereospec-
ifically (Ն95% inversion) furnished the corresponding silane
(SiS)-9a in quantitative yield. This procedure allows for an
Chloride-Induced Racemisation
Early investigations by Price using [36Cl]-labelling tech- indirect yet reliable determination of the enantiomeric ratio
niques demonstrated a fast racemisation of optically active of silyl chloride (SiS)-7a by HPLC.
186
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 184Ϫ195

Reductive Metallation of Asymmetrically Substituted Silyl Chlorides
FULL PAPER
Table 1. Chloride-induced racemisation of (SiS)-7a
Entry[a]
er (SiS)-7a
Solvent
Additive[b]
T [°C]
er[c] (SiS)-9a
1
2
3
4
5
6
7
8
96:4
96:4
96:4
92:8
92:8
94:6
92:8
94:6
94:6
94:6
96:4
THF
THF
THF
THF
THF
THF
THF
THF
Ϫ
0
0
10:90
50:50
50:50
15:85
10:90
12:88
12:88
32:68
50:50
50:50
5:95
LiCl
LiCl
NaCl
KCl
LiF
LiBr
LiI
LiClO4
Li2CO3
LiCl
Ϫ78
Ϫ78
Ϫ78
Ϫ78
Ϫ78
Ϫ78
Ϫ78
Ϫ78
Ϫ100
9[d]
10
11
THF
THF
THF/Et2O/pentane[e]
[a]
Reactions were conducted with a substrate concentration of 0.15Ϫ0.25 m in degassed solvents and were kept at the indicated tempera-
ture for 1 h; reactions were terminated by addition of Et2O followed by solid LiAlH4 (Ն95% inversion).[19] [b] 0.5 Equiv. of additive in
[c]
[d]
the solvent prior to substrate addition. Measured by HPLC using a Daicel Chiralcel OD-H column (n-heptane at 20 °C).
LiClO4
[e]
led to almost complete decomposition of (SiS)-7a and only trace amounts of (SiS)-9a were isolated. 4:1:1 mixture (Trapp mixture).
Although it has been reported that silyl chlorides race- However, this metallation was sluggish and required stoi-
mise in THF,[25] racemisation of (SiS)-7a was marginal after chiometric amounts (100 mol %) of 4,4Ј-di-tert-butylbi-
1 h and only occurred after prolonged reaction times phenyl (DBB). Nevertheless, this modification clearly affec-
(Table 1, Entry 1).[17a] In accordance with a systematic ted the stereochemical outcome of the metallation of (SiS)-
study by Sommer,[17a] solvents such as Et2O and pentane 7a (Scheme 3) and provided almost racemic (SiR)-9a (er ϭ
also restrained racemisation and solvents with high dielec- 56:44) but with slight overall retention of stereochemistry.
tric constants such as acetonitrile led to rapid solvent- The dependence of the absolute configuration and enanti-
induced racemisation. However, all these solvents are un- omeric ratio of 9a on the reaction temperature and reducing
suitable for reductive lithiation reactions.
species (lithium metal versus LiDBB) suggests that an ad-
In contrast, addition of lithium chloride led to immediate ditional reaction pathway occurs.
racemisation at 0 °C and Ϫ78 °C (Table 1, Entries 2 and 3)
whereas racemisation was only marginal with sodium and
potassium chloride (Table 1, Entries 4 and 5). The degree
of racemisation was somewhat dependent on the lithium
halide but was not as clear cut as with lithium chloride
(Table 1, Entries 6Ϫ8). These observations indicate a com-
plex interplay of the inorganic salt (anion and cation), its
solubility in THF and solvent-separated ion pairs. Interest-
ingly, lithium perchlorate and carbonate led to complete ra-
cemisation (Table 1, Entries 9 and 10). If perchlorate and
carbonate anions are considered inert and noncoordinating,
the lithium counterion might also be involved in ionisation
processes liberating chloride which will in turn promote ra-
cemisation.[17a,22,26]
Formal Dimerisation: Disilane Formation
Disilanes have often been isolated quantitatively since the
reductive cleavage of the siliconϪsilicon bond is the rate-
determining step.[10a] We were able to isolate disilane 11a as
a mixture of all possible stereoisomers, meso-11a and
(R*,R*)-11a (er ϭ 61:39) in 52% yield when metallation of
(SiS)-7a (er ϭ 88:12) using lithium metal was terminated
after 1 h (Scheme 5). The presence of 11a has also been
verified for the Trapp mixture at Ϫ100 °C. Compound 11a
was even formed at temperatures as low as Ϫ120 °C.
Clearly, subsequent reductive cleavage of a mixture of
achiral meso-11a and slightly enantiomerically enriched
(R*,R*)-11a will only result in an almost racemic silyl-
lithium 8a.[28]
It is noteworthy that no racemisation occurred in THF/
Et2O/pentane (4:1:1) (Trapp mixture[27]) at Ϫ100 °C
(Table 1, Entry 11). We succeeded in performing the re-
While we were performing these studies, Kawakami de-
ductive metallation of (SiS)-7a using this solvent mixture. scribed the synthesis of a disilane (er ϭ 75:25) by nucleo-
Scheme 5. Reductive metallation of (SiS)-7a accompanied by the formation of disilanes 11
www.eurjoc.org
Eur. J. Org. Chem. 2005, 184Ϫ195
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
187

M. Oestreich, G. Auer, M. Keller
FULL PAPER
philic chloride displacement of highly enantioenriched with solid lithium metal. (3) Dimerisation is faster than
α-NpPhMeSiCl (er Ͼ 99:1) using α-NpPhMeSiLi (er ϭ chloride-induced racemisation (Scheme 6).
91:9).[29] This reaction appears to proceed with predomi-
nant stereoretention at the former lithium-bearing silicon
centre and with stereoinversion at the former electrophilic
silicon centre. However, this transformation displays only
moderate stereoselectivity, thereby confirming our findings.
In control experiments, precluding any diastereoselection,
we treated highly enantioenriched (SiS)-7a (er ϭ 96:4) with
the achiral silyl anion 12 under the reaction conditions of
reductive metallation (THF, Ϫ78 °C). Inverse as well as
normal addition furnished disilane (SiR)-13a with clean in-
version[8,25] of the configuration (Scheme 6). Notably, we
were able to demonstrate that the substitution reaction is
faster than the chloride-induced racemisation. In the case
of the inverse addition protocol, the enantiomeric ratio of
(SiR)-13a is dependent on the rate of addition of 12 which
is accompanied by equimolar amounts of lithium chloride.
Anion 12 is immediately consumed upon addition to silyl
chloride (SiS)-7a yielding disilane (SiR)-13a along with un-
changed (SiS)-7a and lithium chloride. If the addition of 12
is slow, chloride will partially racemise the unchanged (SiS)-
7a. If addition of 12 is rapid, the stereochemical identity of
(SiS)-7a is preserved. The normal addition protocol circum-
vents this side-reaction since (SiS)-7a and lithium chloride
do not coexist under these circumstances. This is reflected
Second Generation Approach
These orientating investigations clearly reveal that an en-
antioselective reductive metallation of an asymmetrically
substituted silyl chloride is impossible if the chiral silyl chlo-
ride is susceptible to chloride-induced racemisation and is
able to formally dimerise.
We became aware of an interesting observation reported
by Fleming[10a,10b] whereby reductive metallation of ArMe2-
SiCl (Ar ϭ Ph or p-Tolyl) is remarkably sensitive to the
presence of substituents at the aryl moiety. Whereas phenyl-
dimethylsilyl chloride is metallated providing the corre-
sponding silyl anion, p-tolyldimethylsilyl chloride is selec-
tively converted into its disilane which is not cleaved by lith-
ium.[10b] Moreover, tris(o-tolyl)silyl chloride gives the silyl-
lithium reagent without the formation of the disilane.[10c]
These findings are extraordinarily significant for our prob-
lem since steric hindrance around the silicon centre seems
to hamper disilane formation yet allows for the direct gen-
eration of the corresponding silyl anion.
Preparation of Sterically Hindered Silyl Chlorides
We decided to prepare these hindered silyl chlorides by
by an almost perfect inversion (99%) of the configuration at gradually increasing the steric bulk of the exocyclic aryl
silicon [(SiS)-7a with er ϭ 96:4 Ǟ (SiR)-13a with er ϭ 5:95]. substituent (7b and 7c, Figure 1). Their syntheses are based
We have drawn the following conclusions from these on the Barbier reaction of chirally modified dichlorosilanes
experiments. (1) Dimerisation during reductive metallation (Ϫ)-17 and dibromide 18 (Scheme 7).[19] The required trich-
can be expected to produce predominantly the meso-config- lorosilanes 15 were prepared from tetrachlorosilane and or-
ured (achiral) disilane. (2) The experimentally observed de- tho-alkylated Grignard reagents derived from the corre-
viation from this stereoselectivity (Scheme 5) is partially sponding aryl bromides 14. Acceptable yields of analytically
due to chloride-induced racemisation which is significantly pure 15 were only achieved using Et2O and not THF as the
faster than the slow heterogeneous reductive metallation solvent since THF undergoes ring-opening with tetrachlo-
Scheme 6. Nucleophilic displacement at (SiS)-7a using the achiral silyl anion 12
188
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 184Ϫ195

Reductive Metallation of Asymmetrically Substituted Silyl Chlorides
FULL PAPER
Scheme 7. Preparation of the sterically hindered silyl chlorides rac-7b and (SiR)-7c
rosilane. The dichlorosilanes (Ϫ)-17 were obtained by main-
taining 15 and (Ϫ)-menthol [(Ϫ)-16] in the presence of urea
in Et2O at room temperature. The Barbier reaction of (Ϫ)-
17 and 18 gave the cyclic silyl ethers (SiRS)-19 in good
yields. (SiRS)-19 were purified by flash chromatography on
silica gel and (SiRS)-19c was quantitatively separated into
its diastereomers. Subsequent stereospecific reduction un-
der forcing reaction conditions provided the silanes 9 in
good to excellent yields. The enantiomeric ratio of (SiS)-9c
(er ϭ 99:1) was determined by HPLC by comparison with
a racemic sample and agreed with the diastereomeric ratio
of (SiR)-19c (dr ϭ 99:1) which was determined by 1H NMR
spectroscopy. Chlorination of 9 using a saturated solution
of chlorine in CCl4 furnished the desired silyl chlorides 7 in
Figure 2. Molecular structure of (SiS)-9c
quantitative yields. The enantiomeric ratio of (SiR)-7c was
determined by HPLC after enantiospecific reduction
(Ն99% inversion) with lithium aluminium hydride.
ray crystallography (Figure 2 and Figure 3).[30] Highly en-
The absolute configurations of (SiS)-9c and (SiR)-7c and, antioenriched samples (er Ն 99:1) of silane (SiS)-9c and
therefore, the stereochemical course of the chlorination silyl chloride (SiR)-7c were both crystallised from pentane
[(SiS)-9c Ǟ (SiR)-7c] were unambiguously deduced by X- at Ϫ30 °C. Silyl chloride (SiR)-7c was prepared from the
Eur. J. Org. Chem. 2005, 184Ϫ195
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
189

M. Oestreich, G. Auer, M. Keller
FULL PAPER
of chloride[22] (from lithium chloride) or pseudorotational
processes[24] (Scheme 4). The latter processes can be ruled
out since complete racemisation occurred at Ϫ78 °C when
tetra(n-butyl)ammonium chloride (NBu4Cl) was used as the
chloride source (Table 2, Entry 9). It seems that chloride is
released from large aggregates of lithium chloride and
THF.[31] In contrast, tetra(n-butyl)ammonium chloride pro-
vides a ‘‘naked’’ chloride anion with a large noncoordinat-
ing tetraalkylammonium counterion. Thus, higher tempera-
tures are needed for nucleophilic attack of chloride derived
from lithium chloride.
The molecular structure of (SiR)-7c (Figure 3) obtained
from a single-crystal X-ray diffraction study illustrates the
increased steric hindrance around the silicon centre. Al-
though the exocyclic siliconϪcarbon bond freely rotates in
solution, this might account for the unprecedented configu-
rational stability of (SiR)-7c.
Figure 3. Molecular structure of (SiR)-7c
same batch of crystallised (SiS)-9c. These X-ray analyses
are the first conclusive evidence for the stereoretentive
nature of the chlorination and the stereoinvertive course of
the reduction at an asymmetrically substituted silicon
centre incorporated into a cyclic framework.
Enantiospecific Reductive Metallation
Treatment of silyl chloride (SiR)-7c with lithium in THF
(0 °C Ǟ Ϫ78 °C, 16 h) resulted in a complex mixture of
products after protolysis.[32] Silane 9c (10Ϫ35%) was iso-
lated along with minor amounts of the corresponding disil-
ane (C36H42Si2, M ϭ 530.89 g/mol) which was unambigu-
ously identified in the mass spectra (m/z ϭ 530 [Mϩ]) of
the crude reaction mixtures. Moreover, a major by-product
(20%) which could not be isolated in analytically pure form
indicated cleavage of an SiϪC(sp2) bond.
We were able to prevent disilane formation by performing
the metallation with addition of preformed LiDBB in THF
to a solution of (SiR)-7c in the Trapp mixture at Ϫ100 °C
and Ϫ120 °C, respectively [(SiR)-7c Ǟ (SiS)-9c, Scheme 8].
These were the first metallations which provided the silyl
anion (SiS)-8c with moderate stereoselectivity (er Յ 74:26)
with retention of the configuration at silicon. Unfortu-
nately, a complex mixture of by-products was formed which
we again believe to be the result of SiϪC(sp2) bond cleav-
age.
Chloride-Induced Racemisation
At first we performed two crucial experiments with (SiR)-
7c. As predicted, (SiR)-7c showed almost perfect configura-
tional stability when simply maintained in THF at room
temperature (Table 2, Entry 1) but unlike (SiS)-7a (R ϭ H)
no chloride-induced racemisation was observed for (SiR)-
7c (R ϭ iPr) in the presence of lithium chloride (Table 2,
Entries 2 and 3). In order to verify this remarkable result,
we investigated the racemisation of (SiR)-7c at several tem-
peratures (Table 2, Entries 4Ϫ7). Substantial racemisation
started to occur at temperatures above Ϫ60 °C whereas be-
low this temperature, (SiR)-7c remained almost untouched
by lithium chloride. Again, the corresponding bromide-
induced racemisation was less pronounced (Table 2, Entry 8).
Table 2. Chloride-induced racemisation of (SiR)-7c
Entry[a]
er (SiR)-7c
Additive[b]
T [°C]
er[c] (SiR)-9c
1
2
3
4
5
6
7
8
9
97:3
97:3
97:3
97:3
97:3
94:6
97:3
94:6
94:6
Ϫ
20
4:96
4:96
4:96
9:91
20:80
47:53
50:50
10:90
50:50
LiCl
LiCl
LiCl
LiCl
LiCl
LiCl
LiBr
NBu4Cl[d]
Ϫ100
Ϫ78
Ϫ60
Ϫ40
Ϫ20
0
Ϫ40
Ϫ78
[a]
Reactions were conducted with a substrate concentration of
0.15Ϫ0.25m in degassed THF and were kept at the indicated tem-
perature for 1 h; reactions were terminated by addition of Et2O
followed by solid LiAlH4 (Ն99% inversion).[19] [b] 0.5 Equiv. LiCl
in solvent prior to substrate addition. [c] Measured by HPLC using
a Daicel Chiralcel OD-H column (n-heptane at 10 °C). [d] Catalytic
amounts, since only trace amounts of NBu4Cl are soluble in THF.
Scheme 8. Reductive metallation of (SiR)-7c
Strohmann has demonstrated the stereoselective transme-
The additional ortho-substituent at the exocyclic aryl tallation of silyllithiums with magnesium bromide fur-
group in (SiR)-7c could hinder either nucleophilic attack nishing the corresponding silylmagnesiums which are less
190
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 184Ϫ195

Reductive Metallation of Asymmetrically Substituted Silyl Chlorides
FULL PAPER
prone to racemisation.[9] Our efforts of transmetallating
(SiS)-8c were most likely thwarted by remaining LiDBB.
reactions were performed in flame-dried glassware under a static
pressure of argon. Liquids were transferred with syringes or
double-ended needles and all solvents were dried and distilled prior
to use following standard procedures. Solvents (cyclohexane and
tert-butyl methyl ether) for extractions and chromatography were
distilled before use. Analytical thin-layer chromatography was per-
formed on silica gel SIL G-25 glass plates and flash chromatogra-
phy on silica gel (40Ϫ63 µm, 230Ϫ400 mesh, ASTM) using the
indicated solvents. All stationary phases were obtained from
Conclusion
In summary, we have conducted the first investigation of
the stereochemical course of the reductive metallation of
silyl chlorides with silicon-centred chirality. The present MachereyϪNagel, Germany. High vacuum distillations (10Ϫ5
mbar) were performed using standard glassware and an Edwards
turbo molecular pump. 1H and 13C NMR spectra were recorded in
CDCl3 on Bruker AM 400 and DRX 500 instruments. Analytical
HPLC analysis on a chiral stationary phase using a Daicel Chiral-
cel OD-H column (n-heptane as solvent) or Daicel Chiralcel OJ-R
column (MeCN:H2O mixtures as solvent) provided baseline resolu-
tion of enantiomers.
study has focussed on two out of several possible processes
which are detrimental to enantioselection, namely halide-
induced racemisation and formal dimerisation. Both reac-
tion pathways were independently verified for silyl chloride
(SiS)-7a. Interestingly, the stereoselectivity of the nucleo-
philic displacement of a SiϪCl bond is dependent on the
order and rate of addition of the nucleophile. Structural
modifications of the exocyclic aryl group in (SiS)-7a led to
the sterically hindered silyl chloride (SiR)-7c which displays
a remarkable peculiarity, i.e. (SiR)-7c is not racemised by
lithium chloride in THF at temperatures below Ϫ40 °C.
The absolute configuration of this highly configurationally
stable silyl chloride (SiR)-7c has been determined by X-ray
crystallography. Furthermore, the reductive metallation of
(SiR)-7c does not proceed through a disilane intermediate
at temperatures below Ϫ100 °C.
General Procedure for the Reductive Metallation of 7 with Lithium
Metal: Silyl chloride 7 was dissolved in THF (1 mL) and finely cut
lithium wire (large excess) was added. After sonication at 0 °C and
the appearance of a slightly red colour, the reaction mixture was
cooled to Ϫ78 °C. The resultant silyl anion 8 was separated from
the remaining lithium wire and H2SO4 in Et2O (2 mL) was added
followed by aqueous saturated NaHCO3 (10 mL). The organic
layer was separated and the aqueous phase extracted with tert-butyl
methyl ether (3 ϫ 30 mL), dried (MgSO4) and the volatiles were
evaporated under reduced pressure. Purification by flash chroma-
tography on silica gel (cyclohexane) provided silane 9. The enanti-
omeric ratio of 9a[19] and 9c was determined by HPLC.
However, metallation of (SiR)-7c (er ϭ 99:1) does not
result in highly enantioenriched silyllithium (SiS)-8c (er ϭ
74:26) even in the absence of chloride-induced racemisation
and formal disilane formation. The experimental finding
that treatment of (SiR)-7c with lithium or LiDBB leads to
General Procedure for the Reductive Metallation of 7 with LiDBB:
Silyl chloride 7 was dissolved in THF/Et2O/pentane, 4:1:1 (1 mL)
and a freshly prepared solution of LiDBB (2 equiv.) was added at
siliconϪcarbon bond cleavage, strongly indicates that the the indicated temperature. The reaction mixture was maintained at
this temperature for the desired time and H2SO4 in Et2O (2 mL)
was then added followed by saturated aqueous NaHCO3 (10 mL).
The organic layer was separated and the aqueous phase extracted
with tert-butyl methyl ether (3 ϫ 30 mL), dried (MgSO4) and the
volatiles were removed under reduced pressure. Purification by
flash chromatography on silica gel (cyclohexane) provided silane 9.
The enantiomeric ratio of 9a[19] and 9c was determined by HPLC.
electron transfer is not as selective as suggested by
Scheme 2. The use of enantioenriched, asymmetrically sub-
stituted silyl chlorides in reductive metallations indicates the
complexity of this process and demanding mechanistic
studies will be necessary for a more refined understanding.
In conclusion, by combining the restrained chloride-in-
duced racemisation and inhibiting formal dimerisation, we
accomplished the first enantioselective reductive metall-
ation of a chiral silyl chloride. This conflicts with previous
literature precedents.[8] Future work will be devoted to the
design of chiral, nonracemic bridgehead silyl chlorides.
Some achiral bi- and tricyclic derivatives have been success-
fully prepared by Sommer and Boudjouk.[33] Importantly,
investigations concerning nucleophilic displacements at
bridgehead silyl chlorides have indicated that stereoreten-
Preparation of Disilanes meso-11a and (R*,R*)-11a: To silyl chlo-
ride (SiS)-7a (166 mg, 0.641 mmol, er ϭ 88:12) in THF (1 mL)
was added lithium wire (large excess). After sonication at 0 °C until
a pale red colour appeared, the resultant reaction mixture was
maintained at Ϫ78 °C for 1 h. After addition of H2SO4 in Et2O
(3 mL), the reaction mixture was allowed to warm to ambient tem-
perature. Saturated aqueous NaHCO3 (10 mL) was added and the
aqueous layer extracted with tert-butyl methyl ether (3 ϫ 50 mL).
The combined organic phases were dried (MgSO4) and the volatiles
tion is favoured.[17a,33] This would remove the undesired in- were evaporated under reduced pressure. Flash chromatography on
silica gel (cyclohexane) provided 11a as a mixture of meso-11a and
(R*,R*)-11a (74.0 mg, 0.166 mmol, 52%, meso-11a: (R*,R*)-11a ϭ
45:55) as a white solid. HPLC (Daicel Chiralcel OD-H column,
column temperature 24 °C, solvent n-heptane/iPrOH, 199:1, flow
rate 0.8 mL·minϪ1): tR ϭ 12.54 min for meso-11a, tR ϭ 14.91 min
(minor enantiomer) and tR ϭ 19.55 (major enantiomer). M.p. 104
°C (cyclohexane). Rf (cyclohexane) ϭ 0.45. 1H NMR (CDCl3,
400 MHz): δ ϭ 1.18Ϫ1.32 (m, 4 H), 1.71Ϫ1.98 (m, 4 H), 2.53Ϫ2.62
(m, 2 H), 2.70Ϫ2.80 (m, 2 H), 7.04Ϫ7.13 (m, 4 H), 7.19Ϫ7.32 (m,
12 H), 7.38 (d, J ϭ 7.3 Hz, 1 H), 7.41 (dd, J ϭ 7.7, J ϭ 1.3 Hz, 1
H) ppm. 13C{1H} NMR (CDCl3, 100 MHz): δ ϭ 10.7, 11.5, 22.1,
22.5, 29.8, 35.3, 125.3, 125.4, 127.8, 127.9, 128.6, 128.8, 129.0,
fluence of the above racemisation processes. If these struc-
tural modifications enable an efficient reductive metall-
ation, the resultant asymmetrically substituted silyl anions
might serve as interesting chiral silyl transfer reagents after
transmetallation to copper.[2]
Experimental Section
General Remarks: Reagents obtained from commercial suppliers
were used without further purification unless otherwise noted. All
Eur. J. Org. Chem. 2005, 184Ϫ195
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
191

M. Oestreich, G. Auer, M. Keller
129.1, 129.2, 130.8, 131.3, 135.1, 135.2, 136.7, 136.9, 137.3, 137.4, (93 °C at 8 mbar) affording 15b (31.4 g, 139 mmol, 53%) as a
FULL PAPER
1
149.1, 149.4 ppm. IR (cuvette): ν˜ ϭ 3687 (m), 3026 (s), 2926 (w),
colourless liquid. H NMR (CDCl3, 400 MHz): δ ϭ 2.34 (s, 3 H),
2401 (m), 2341 (m), 1522 (m), 1476 (m), 1426 (m), 1336 (w), 1225 6.95Ϫ6.98 (m, 2 H), 7.13Ϫ7.17 (m, 1 H), 7.56 (d, J ϭ 7.3 Hz, 1 H)
(s), 1018 (m), 929 (s), 627 (m) cmϪ1. LRMS (EI): m/z ϭ 446 [Mϩ]. ppm. 13C{1H} NMR (CDCl3, 100 MHz): δ ϭ 22.7, 125.6, 129.6,
HRMS (EI): calcd. for C30H30Si2: 446.1886; found 446.1886.
131.3, 133.1, 134.8, 143.8 ppm. IR (cuvette): ν˜ ϭ 3854 (w), 3745
(w), 3058 (s), 2979 (s), 2931 (s), 2743 (w), 2671 (w), 2401 (m), 2361
(m), 1971 (m), 1935 (m), 1820 (m), 1700 (m), 1591 (s), 1567 (s),
1449 (s), 1385 (s), 1287 (s), 1216 (m), 1166 (m), 1133 (s), 1082 (s),
1033 (m), 987 (w), 929 (w), 772 (w) cmϪ1. LRMS (EI): m/z ϭ 224
[Mϩ]. C7H7Cl3Si (223.93): calcd. C 37.27, H 3.13; found C 37.29,
H 2.93.
Preparation of (SiR)-13a. Inverse Addition: Me2PhSiCl (76.2 mg,
0.446 mmol) in THF (2 mL) was treated with lithium wire (large
excess) at 0 °C and maintained at this temperature for 5 h. Silyl
chloride (SiS)-7a (92.4 mg, 0.357 mmol) was dissolved in THF
(1 mL), cooled to Ϫ78 °C and the freshly prepared Me2PhSiLi (12)
in THF was added via syringe (rapid addition: 2.00 mL·minϪ1
;
slow addition: 0.02 mL·minϪ1). The reaction mixture was main-
tained at Ϫ78 °C for 1 h.
Trichloro(2-isopropylphenyl)silane (15c): According to the prep-
aration of 15b, magnesium turnings (1.42 g, 58.4 mmol, 1.16 equiv.)
were flame-dried under vigorous stirring and Et2O (30 mL) was
added followed by dropwise addition of a solution of 2-isopropyl-
bromobenzene (14c) (10.0 g, 50.2 mmol) in Et2O (30 mL). The re-
sultant reaction mixture was heated to reflux for 18 h. The Grig-
nard reagent was then added dropwise to SiCl4 (7.00 mL,
60.2 mmol, 1.20 equiv.) in Et2O (30 mL) and maintained for 16 h
at 40 °C. After the usual workup, the crude product was distilled
(120 °C at 22 mbar) furnishing 15c (4.14 g, 16.3 mmol, 33%) as a
Normal Addition: Me2PhSiCl (19.8 mg, 0.116 mmol) in THF
(1 mL) was treated with lithium wire (large excess) at 0 °C and
maintained at this temperature for 5 h. The freshly prepared
Me2PhSiLi (12) was separated from the remaining lithium metal
and silyl chloride (SiS)-7a (24.3 mg, 0.0939 mmol) dissolved in
THF (1 mL) was added via syringe. Both reactions were terminated
by addition of aqueous saturated NH4Cl (10 mL). The aqueous
phase was separated and extracted with tert-butyl methyl ether (3
ϫ 50 mL). The combined organic phases were dried (MgSO4) and
the volatiles were evaporated under reduced pressure. Flash chro-
matography on silica gel (cyclohexane) provided (SiR)-13a
(77.9 mg, 0.217 mmol, 61%, er ϭ 8:92) (inverse addition) and
1
colourless liquid. H NMR (CDCl3, 400 MHz): δ ϭ 1.34 (d, J ϭ
6.8 Hz, 6 H), 3.47 (sept, J ϭ 6.8 Hz, 1 H), 7.29Ϫ7.88 (m, 4 H)
ppm. 13C{1H} NMR (CDCl3, 100 MHz): δ ϭ 24.2, 34.7, 125.9,
126.9, 133.4, 134.7, 155.5 ppm. IR (cuvette): ν˜ ϭ 3393 (m), 2969
(s), 2401 (m), 2253 (w), 1591 (m), 1540 (w), 1477 (s), 1435 (s), 1275
(m), 1213 (s), 1125 (s), 1069 (m), 1030 (w), 928 (w), 794 (m), 746
(w), 660 (w) cmϪ1. LRMS (EI): m/z ϭ 254 [Mϩ]. HRMS (EI):
calcd. for C9H11SiCl3: 251.9696; found 251.9690.
(20.4 mg, 0.0569 mmol, 61%, er ϭ 5:95) (normal addition) as
20
colourless oils. Rf (cyclohexane) ϭ 0.46. [α]2D0 ϭ Ϫ15.9, [α]
ϭ
578
20
20
20
Ϫ16.7, [α] ϭ Ϫ19.1, [α] ϭ Ϫ35.7, [α] ϭ Ϫ68.3 (c ϭ 12.6,
546
436
365
CHCl3). HPLC (Daicel Chiralcel OJ-R column, column tempera-
ture 20 °C, solvent MeCN:H2O ϭ 75:25, flow rate 0.5 mL·minϪ1):
(؊)-Dichloro-[(1R,2S,5R)-menthyloxy]-(2-methylphenyl)silane
tR ϭ 14.03 min for (SiR)-13a (major enantiomer), tR ϭ 15.67 min (17b): A three-necked 500-mL flask equipped with a pressure-
for (SiS)-13a (minor enantiomer). 1H NMR (CDCl3, 500 MHz):
equalising dropping funnel, a reflux condenser and a magnetic
δ ϭ 0.39 (s, 3 H), 0.45 (s, 3 H), 1.19 (t, J ϭ 6.2 Hz, 2 H), 1.78Ϫ1.95 stirrer bar was charged with 15b (31.4 g, 139 mmol) and urea
(m, 2 H), 2.57Ϫ2.63 (m, 1 H), 2.74Ϫ2.79 (m, 1 H), 7.10Ϫ7.17 (m,
2 H), 7.14Ϫ7.36 (m, 9 H), 7.38 (dd, J ϭ 7.2, J ϭ 1.2 Hz, 1 H),
(9.21 g, 153 mmol, 1.10 equiv.). At ambient temperature, a solution
of (Ϫ)-menthol [(Ϫ)-16] (21.7 g, 139 mmol, 1.00 equiv.) and Et2O
7.41Ϫ7.42 (m, 2 H) ppm. 13C{1H} NMR (CDCl3, 125 MHz): δ ϭ (200 mL) was added dropwise over a period of 1 h. The reaction
Ϫ3.4, Ϫ2.8, 10.6, 22.3, 35.4, 125.3, 127.8, 127.9, 128.6, 128.7, 128.8,
129.1, 131.2, 134.2, 134.9, 136.5, 137.6, 138.7, 149.2 ppm. IR (cu-
mixture was maintained at this temperature for 2 h followed by
filtration using Schlenk techniques. The filter cake was washed with
vette): ν˜ ϭ 3690 (m), 3155 (m), 3054 (m), 2924 (s), 2857 (s), 2360 Et2O (3 ϫ 50 mL) and the filtrate was concentrated in vacuo. High
(m), 2253 (s), 1793 (m), 1428 (s), 1382 (m), 1246 (m), 1104 (m), 937 vacuum distillation of the residue (135 °C at 6 ϫ 10Ϫ5 mbar) pro-
(m), 886 (m), 811 (w) cmϪ1. LRMS (EI): m/z ϭ 358 [Mϩ]. HRMS vided dichlorosilane (Ϫ)-17b (31.4 g, 91.1 mmol, 65%) as a colour-
less liquid. [α]2D0 ϭ Ϫ44.7, [α] ϭ Ϫ46.3, [α] ϭ Ϫ52.7, [α]
ϭ
20
20
20
(EI): calcd. for C23H26Si2: 358.1573; found 358.1577.
578
546
436
ϭ Ϫ134 (c ϭ 18.8, CHCl3). 1H NMR (CDCl3,
20
Ϫ87.8, [α]
365
Trichloro(2-methylphenyl)silane (15b): A three-necked 250-mL flask
equipped with a pressure-equalising dropping funnel, a reflux con-
denser and a magnetic stirrer bar was charged with magnesium
turnings (8.76 g, 360 mmol, 1.37 equiv.) and was subsequently
flame-dried (3 times) in vacuo. After filling with argon, Et2O
(60 mL) was added followed by dropwise addition of a solution
of 1-bromo-2-methylbenzene (14b) (45.0 g, 31.5 mL, 263 mmol) in
Et2O (30 mL). Ice-cooling was needed at the beginning of the Grig-
nard reaction in order to ensure a gentle reflux followed by heating
at reflux for 1 h. The Grignard reagent was decanted from the ex-
cess magnesium and transferred to a pressure-equalising dropping
funnel. A second three-necked flask (500 mL) equipped with the
above pressure-equalising dropping funnel, a reflux condenser and
a magnetic stirrer bar was charged with SiCl4 (45.0 mL, 395 mmol,
1.50 equiv.). The Grignard reagent was added dropwise and the
400 MHz): δ ϭ 0.78 (d, J ϭ 6.9 Hz, 3 H), 0.91 (d, J ϭ 7.3 Hz, 3
H), 0.92 (d, J ϭ 6.4 Hz, 3 H), 1.01 (dq, J ϭ 12.5, J ϭ 2.6 Hz, 2
H), 1.19 (dq, J ϭ 12.0, J ϭ 2.2 Hz, 1 H), 1.33 (dq, J ϭ 12.5, J ϭ
2.71 Hz, 1 H), 1.41Ϫ1.50 (m, 1 H), 1.63Ϫ1.68 (m, 2 H), 2.14Ϫ2.23
(m, 2 H), 2.60 (s, 3 H), 4.06 (dt, J ϭ 10.3, J ϭ 4.3 Hz, 1 H), 7.21
(d, J ϭ 8.6 Hz, 2 H), 7.39 (t, J ϭ 7.7 Hz, 1 H), 7.81 (d, J ϭ 7.3 Hz,
1 H) ppm. 13C{1H} NMR (CDCl3, 100 MHz): δ ϭ 15.8, 21.3, 22.2,
22.7, 22.9, 25.5, 31.8, 34.4, 44.2, 49.7, 76.5, 125.2, 130.4, 130.7,
132.1, 135.1, 144.9 ppm. IR (cuvette): ν˜ ϭ 3013 (m), 2959 (s), 2871
(s), 2401 (m), 1593 (s), 1455 (s), 1386 (m), 1371 (m), 1286 (m), 1214
(w), 1136 (w), 879 (m), 756 (w) cmϪ1. LRMS (CI/NH3): m/z ϭ 362
[[M ϩ NH4]ϩ]. C17H26Cl2OSi (345.38): calcd. C 59.12, H 7.59;
found C 58.96, H 7.53.
(؊)-Dichloro-(2-isopropylphenyl)-[(1R,2S,5R)-menthyloxy]silane
resultant reaction mixture was maintained for 2 h at 40 °C. The (17c): According to the procedure for (Ϫ)-17b, a solution of (Ϫ)-
reaction mixture was allowed to return to ambient temperature and
was filtered using Schlenk techniques. The filter cake was washed
menthol [(Ϫ)-16] (4.55 g, 29.1 mmol, 1.00 equiv.) and Et2O (20 mL)
was added dropwise to trichlorosilane 15c (7.34 g, 29.0 mmol) and
with Et2O (3 ϫ 20 mL) and the combined organic solvents were urea (1.91 g, 31.9 mmol, 1.10 equiv.) over a period of 1 h. The reac-
evaporated under reduced pressure. The crude product was distilled tion mixture was maintained at this temperature for 3 h followed
192
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 184Ϫ195

Reductive Metallation of Asymmetrically Substituted Silyl Chlorides
FULL PAPER
by filtration. The filter cake was washed with Et2O (3 ϫ 5 mL) and
the filtrate was concentrated in vacuo. High vacuum distillation of
the residue (160 °C at 6 ϫ 10Ϫ5 mbar) provided dichlorosilane (Ϫ)-
After a short period of heating at reflux, the magnesium was
treated with an equimolar mixture of dibromide 18 (1.26 g,
4.53 mmol) and (Ϫ)-17c (1.14 g, 4.53 mmol) in THF (10 mL).
Further heating at reflux was followed by the usual workup. Purifi-
cation and quantitative separation of the diastereomers by flash
17c (5.44 g, 14.6 mmol, 50%) as a colourless viscous liquid. [α]2D0
ϭ
20
20
20
20
Ϫ50.0, [α]
ϭ Ϫ51.2, [α]
ϭ Ϫ58.9, [α]
ϭ Ϫ95.8, [α]
ϭ
578
546
436 365
1
Ϫ144 (c ϭ 16.8, CHCl3). H NMR (CDCl3, 400 MHz): δ ϭ 0.79 chromatography on silica gel furnished (SiR)-19c (549 mg,
(d, J ϭ 6.8 Hz, 3 H), 0.81Ϫ0.87 (m, 1 H), 0.90 (d, J ϭ 6.8 Hz, 3 1.30 mmol, 29%) and (SiS)-19c (571 mg, 1.36 mmol, 30%) as vis-
H), 0.94 (d, J ϭ 6.4 Hz, 3 H), 1.21 (q, J ϭ 12.0 Hz, 1 H), 1.24Ϫ1.32 cous colourless oils. The diastereomeric ratios of 1:99 and 99:1,
(m, 1 H), 1.28 (d, J ϭ 1.7 Hz, 3 H), 1.30 (d, J ϭ 2.2 Hz, 3 H),
respectively, were determined from the 1H NMR spectra by inte-
1.31Ϫ1.36 (m, 1 H), 1.41Ϫ1.53 (m, 1 H), 1.64Ϫ1.72 (m, 2 H), gration of the baseline separated doublets at δ ϭ 0.35 and
2.15Ϫ2.26 (m, 2 H), 3.45 (sept, J ϭ 6.8 Hz, 1 H), 4.09 (ddd, J ϭ
0.61 ppm. Analytical data for (SiR)-19c: Rf (cyclohexane) ϭ 0.41.
10.3, J ϭ 4.3 Hz, 1 H), 7.23Ϫ7.83 (m, 4 H) ppm. 13C{1H} NMR
1H NMR (CDCl3, 500 MHz): δ ϭ 0.35 (d, J ϭ 6.9 Hz, 3 H), 0.81
(CDCl3, 100 MHz): δ ϭ 15.8, 21.3, 22.2, 22.9, 24.3, 24.4, 25.4, 31.7, (d, J ϭ 6.9 Hz, 3 H), 0.88 (d, J ϭ 6.4 Hz, 3 H), 0.94 (d, J ϭ 6.9 Hz,
34.2, 34.4, 44.2, 49.7, 76.4, 125.6, 126.2, 132.4, 135.1, 155.5 ppm. 3 H), 1.07Ϫ1.15 (m, 3 H), 1.22 (d, J ϭ 6.9 Hz, 3 H), 1.26Ϫ1.37
IR (cuvette): ν˜ ϭ 3691 (w), 3308 (m), 3156 (m), 2963 (s), 2929 (s), (m, 3 H), 1.48Ϫ1.67 (m, 3 H), 1.97Ϫ2.08 (m, 3 H), 2.19Ϫ2.26 (m,
2871 (s), 2361 (s), 2255 (s), 1794 (w), 1650 (vw), 1561 (w), 1460 1 H), 2.73Ϫ2.80 (m, 1 H), 2.87Ϫ2.93 (m, 1 H), 3.09 (sept, J ϭ
(m), 1384 (m), 1276 (vw), 1215 (w), 1126 (s), 1085 (m), 937 (m), 6.7 Hz, 1 H), 3.46 (ddd, J ϭ 10.3, J ؍ 4.3 Hz, 1 H), 7.10Ϫ7.19 (m,
874 (m), 780 (m), 726 (w), 676 (m), 637 (w) cmϪ1. LRMS (EI): 3 H), 7.24Ϫ7.30 (m, 2 H), 7.35 (t, J ϭ 7.3 Hz, 1 H), 7.45 (d, J ϭ
m/z ϭ 372 [Mϩ]. C19H30Cl2OSi (373.43): calcd. C 61.11, H 8.10;
found C 61.28, H 7.96.
7.3 Hz, 1 H), 7.55 (dd, J ϭ 6.0, J ϭ 0.6 Hz, 1 H) ppm. 13C{1H}
NMR (CDCl3, 125 MHz): δ ϭ 15.1, 15.7, 21.6, 22.5, 22.8, 23.2,
24.0, 24.7, 25.0, 31.7, 33.8, 34.6, 35.2, 45.7, 50.7, 73.1, 125.0, 125.2,
125.4, 128.5, 129.6, 130.0, 132.7, 134.7, 135.9, 136.2, 149.6,
155.6 ppm. IR (cuvette): ν˜ ϭ 3155 (w), 2927 (s), 2853 (s), 2341 (m),
2254 (m), 1793 (w), 1590 (w), 1560 (w), 1457 (m), 1384 (m), 1216
(w), 1081 (m), 972 (m), 904 (m), 776 (m) cmϪ1. LRMS (CI/NH3):
m/z ϭ 438 [[M ϩ NH4]ϩ]. C28H40OSi (420.70): calcd. C 79.94, H
10.04; found C 79.94, H 9.58. Analytical data for (SiS)-19c: Rf
(SiRS)-1-[(1R,2S,5R)-Menthyloxy]-(2-methylphenyl)-1,2,3,4-tetra-
hydro-1-silanaphthalene [(SiRS)-19b]: A 250-mL three-necked flask
equipped with a reflux condenser, a pressure-equalizing dropping
funnel, an argon inlet and a magnetic stirrer bar was charged with
magnesium turnings (6.60 g, 271 mmol, 9.00 equiv.). The flask was
flame dried with vigorous stirring. After filling with argon, THF
(50 mL) and 1,2-dibromoethane (1.10 mL, 11.2 mmol, 0.40 equiv.)
were added. The magnesium turnings were treated with an equimo-
lar mixture of dibromide 18 (8.40 g, 30.2 mmol) and (Ϫ)-17b
(10.4 g, 30.2 mmol) in THF (50 mL) at reflux. The reaction mixture
was maintained for a further 16 h at reflux and was then allowed
to return to ambient temperature. Diluting with n-pentane (50 mL)
resulted in additional precipitation of magnesium salts which were
filtered and the volatiles were removed under reduced pressure.
Flash chromatography on silica gel (cyclohexaneϪt-butyl methyl
ether, 99:1) furnished (SiRS)-19b (8.63 g, 22.0 mmol, 73%) as a
colourless oil. The diastereomeric ratio of 55:45 was determined
from the 1H NMR spectra by integration of the baseline separated
doublets at δ ϭ 0.33 and 0.61 ppm. Rf (cyclohexane) ϭ 0.43. 1H
NMR (CDCl3, 400 MHz): δ ϭ 0.33 (d, J ϭ 6.9 Hz, 1.65 H), 0.61
(d, J ϭ 6.9 Hz, 1.35 H), 0.76 (d, J ϭ 6.8 Hz, 1.65 H), 0.83 (d, J ϭ
7.3 Hz, 1.35 H), 0.89 (d, J ϭ 6.5 Hz, 1.35 H), 0.91 (d, J ϭ 7.3 Hz,
1.65 H), 1.00 (ddd, J ϭ 10.7, J ϭ 1.3 Hz, 1 H), 1.08Ϫ1.37 (m, 6
H), 1.53Ϫ1.63 (m, 3 H), 1.72Ϫ1.78 (m, 0.45 H), 1.97Ϫ2.04 (m, 2
H), 2.18Ϫ2.24 (m, 0.55 H), 2.32 (s, 1.35 H), 2.34 (s, 1.65 H),
2.75Ϫ2.83 (m, 1 H), 2.86Ϫ2.94 (m, 1 H), 3.49Ϫ3.65 (m, 1 H),
7.11Ϫ7.27 (m, 4 H), 7.28Ϫ7.33 (m, 2 H), 7.47Ϫ7.50 (m, 1 H),
7.55Ϫ7.61 (m, 1 H) ppm. 13C{1H} NMR (CDCl3, 100 MHz): δ ϭ
14.0, 14.6, 15.7, 15.8, 21.5, 22.3, 22.5, 22.8, 23.0, 23.1, 25.1, 25.3,
27.0, 29.2, 29.8, 31.5, 34.6, 35.4, 36.0, 45.1, 45.8, 50.5, 50.6, 73.2,
73.5, 124.6, 124.7, 125.4, 125.5, 125.7, 126.6, 127.2, 128.3, 128.4,
128.5, 128.6, 128.8, 129.7, 129.9, 132.3, 132.7, 135.6, 135.7, 135.9,
136.3, 144.1, 149.7, 149.9 ppm. IR (cuvette): ν˜ ϭ 3687 (m), 3010
(s), 2928 (s), 2870 (m), 2401 (s), 2361 (s), 1928 (w), 1590 (m), 1521
(m), 1554 (s), 1371 (m), 1207 (w), 1099 (w), 974 (w), 929 (m), 868
(w) cmϪ1. LRMS (CI/isobutene): m/z ϭ 449 [[M ϩ C4H9]ϩ].
C26H36OSi (392.65): calcd. C 79.53, H 8.89; found C 79.58, H 9.24.
1
(cyclohexane) ϭ 0.39. H NMR (CDCl3, 500 MHz): δ ϭ 0.61 (d,
J ϭ 6.9 Hz, 3 H), 0.75 (d, J ϭ 6.4 Hz, 3 H), 0.77Ϫ0.88 (m, 2 H),
0.91 (d, J ϭ 7.3 Hz, 3 H), 0.92 (d, J ϭ 6.4 Hz, 3 H), 0.95Ϫ1.10 (m,
2 H), 1.22 (d, J ϭ 6.5 Hz, 3 H), 1.24Ϫ1.32 (m, 2 H), 1.34Ϫ1.59
(m, 2 H), 1.72Ϫ1.77 (m, 1 H), 1.98Ϫ2.06 (m, 3 H), 2.29Ϫ2.36 (m,
1 H), 2.73Ϫ2.80 (m, 1 H), 2.85Ϫ2.92 (m, 1 H), 3.07 (sept, J ϭ
6.9 Hz, 1 H), 3.53 (ddd, J ϭ 9.9, J ϭ 4.7 Hz, 1 H), 7.11Ϫ7.17 (m,
3 H), 7.28 (dt, J ϭ 7.3, J ϭ 7.3 Hz, 2 H), 7.35 (dt, J ϭ 7.3, J ϭ
1.7 Hz, 1 H), 7.48 (dd, J ϭ 6.9, J ϭ 0.9 Hz, 1 H), 7.59 (dd, J ϭ
7.7, J ϭ 1.7 Hz, 1 H) ppm. 13C{1H} NMR (CDCl3, 125 MHz): δ ϭ
14.8, 15.8, 21.6, 22.3, 22.8, 23.2, 24.3, 24.5, 25.3, 25.4, 29.8, 31.7,
33.8, 34.5, 35.4, 45.2, 76.3, 125.0, 125.2, 128.7, 129.6, 130.1, 133.2,
134.8, 135.9, 136.4, 149.3, 155.6 ppm.
rac-1-(2-Methylphenyl)-1,2,3,4-tetrahydro-1-silanaphthalene
(rac-
9b): A 25-mL Schlenk flask equipped with a magnetic stirrer bar
and a reflux condenser was charged with (SiRS)-19b (388 mg,
0.988 mmol) and di-n-butyl ether (8 mL). DIBAL (3.00 mL,
3.00 mmol, 3.00 equiv., 1m in hexane) was added and the reaction
mixture was maintained at reflux for 16 h. The reaction mixture
was allowed to return to ambient temperature and acetone (15 mL)
was added followed by concentrated HCl (10 mL). After diluting
with H2O (20 mL), the organic layer was separated and the aque-
ous phase was extracted with tert-butyl methyl ether (3 ϫ 30 mL).
The combined organic layers were dried (MgSO4), filtered and the
volatiles were evaporated in vacuo. Purification by flash chroma-
tography on silica gel (cyclohexane) provided the cyclic silane rac-
9b (180 mg, 0.755 mmol, 77%) as a colourless liquid. The enanti-
omeric ratio of 55:45 was determined from HPLC by integration
of baseline separated peaks which were assigned as the enantiomers
by comparison with a racemic sample. Rf (cyclohexane) ϭ 0.45.
HPLC (Daicel Chiralcel OD-H column, column temperature 24 °C,
solvent n-heptane, flow rate 0.5 mL·minϪ1): tR ϭ 15.75 min (minor
enantiomer), tR ϭ 19.98 min (major enantiomer). 1H NMR
(CDCl3, 500 MHz): δ ϭ 1.10Ϫ1.17 (m, 1 H), 1.32Ϫ1.35 (m, 1 H),
2.00Ϫ2.07 (m, 2 H), 2.43 (s, 3 H), 2.85Ϫ2.87 (m, 2 H), 5.02 (t, J ϭ
(SiRS)-1-(2-Isopropylphenyl)-1-[(1R,2S,5R)-menthyloxy]-1,2,3,4-
tetrahydro-1-silanaphthalene [(SiRS)-19c]: By analogy with the pro-
cedure for (SiRS)-19b, magnesium turnings (0.99 g, 41 mmol, 9.0
equiv.) were flame dried under vigorous stirring. THF (10 mL) then
1,2-dibromoethane (0.17 mL, 1.67 mmol, 0.400 equiv.) were added. 2.8 Hz, 1 H), 7.12Ϫ7.41 (m, 8 H) ppm. 13C{1H} NMR (CDCl3,
Eur. J. Org. Chem. 2005, 184Ϫ195
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
193

M. Oestreich, G. Auer, M. Keller
FULL PAPER
125 MHz): δ ϭ 9.5, 22.4, 22.7, 35.1, 125.1, 125.6, 128.9, 129.5,
vette): ν˜ ϭ 3736 (w), 3677 (w), 3059 (m), 3010 (m), 2929 (s), 2863
129.6, 129.7, 130.0, 134.3, 136.2, 136.7, 144.2, 149.7 ppm. IR (cu-
(m), 2401 (m), 2361 (m), 1930 (w), 1824 (w), 1699 (w), 1590 (m),
vette): ν˜ ϭ 3687 (m), 3025 (s), 2927 (m), 2692 (m), 2401 (s), 2361 1560 (m), 1521 (w), 1473 (m), 1437 (m), 1337 (w), 1269 (w), 1202
(m), 2124 (s), 1889 (w), 1521 (s), 1474 (m), 1435 (m), 1227 (s), 1140 (m), 1131 (m), 1081 (m), 1029 (m), 974 (m), 916 (m), 799 (m), 690
(m), 1079 (w), 928 (s), 749 (w) cmϪ1. LRMS (EI): m/z ϭ 238 [Mϩ].
(m) cmϪ1. LRMS (EI): m/z ϭ 272 [Mϩ]. HRMS (EI): calcd. for
C16H18Si (238.40): calcd. C 80.61, H 7.61; found C 80.75, H 7.88.
C16H17ClSi: 272.0788; found 272.0793.
(SiS)-1-(2-Isopropylphenyl)-1,2,3,4-tetrahydro-1-silanaphthalene
[(SiS)-9c]: According to the preparation of rac-9b, a solution of
silyl ether (SiR)-19c (645 mg, 1.54 mmol, dr ϭ 1:99) and di-n-butyl
(SiR)-1-Chloro-(2-isopropylphenyl)-1,2,3,4-tetrahydro-1-silanaph-
thalene [(SiR)-7c]: According to the procedure of rac-7b, a satu-
rated solution of Cl2 in CCl4 (0.30 mL) was added to a solution of
ether (20 mL) was treated with DIBAL (4.70 mL, 4.70 mmol, 3.00 (SiS)-9c (40 mg, 0.15 mmol, er ϭ 1:99) in CCl4 (2 mL) at 0 °C. The
equiv., 1m in hexane) at reflux for 16 h. After the usual workup
with acetone (30 mL) and concentrated HCl (20 mL), the aqueous
phase was extracted with tert-butyl methyl ether (3 ϫ 60 mL). Puri-
fication by flash chromatography on silica gel (cyclohexane) pro-
vided cyclic silane (SiS)-9c (373 mg, 1.40 mmol, 91%) as a colour-
less liquid. The enantiomeric ratio of 1:99 was determined from
HPLC by integration of baseline separated peaks which were as-
reaction mixture was maintained for 2 min and a second portion
of Cl2 in CCl4 (0.10 mL) was then added. The solution was purged
with argon after 2 min. Evaporation of the solvent under high vac-
uum provided (SiR)-7c (44.7 mg, 0.148 mmol, 99%) as a colourless
liquid which solidified upon standing at Ϫ30 °C in n-pentane.
20
20
20
[α]2D0 ϭ ϩ24.5, [α]
ϭ ϩ25.3, [α]
ϭ ϩ29.6, [α]
ϭ ϩ54.1,
578
546
436
20
[α] ϭ ϩ91.1 (c ϭ 25.7, CHCl3). 1H NMR (CDCl3, 400 MHz):
365
signed as the enantiomers by comparison with a racemic sample. δ ϭ 1.00 (d, J ϭ 6.9 Hz, 3 H), 1.30 (d, J ϭ 6.9 Hz, 3 H), 1.38Ϫ1.46
20
20
Rf (cyclohexane) ϭ 0.45. [α]2D0 ϭ ϩ78.2, [α]
ϭ ϩ82.7, [α]
ϭ
(m, 1 H), 1.55Ϫ1.60 (m, 1 H), 2.07Ϫ2.20 (m, 2 H), 2.80Ϫ2.87 (m,
ϭ ϩ315 (c ϭ 11.0, CHCl3), HPLC 1 H), 2.95Ϫ3.00 (m, 1 H), 3.05 (sept, J ϭ 6.9 Hz, 1 H), 7.19Ϫ7.27
578
546
20
20
ϩ96.4, [α]
ϭ ϩ177, [α]
436
365
(Daicel Chiralcel OD-H column, column temperature 10 °C, sol-
(m, 3 H), 7.34Ϫ7.38 (m, 2 H), 7.47 (dt, J ϭ 7.7, J ϭ 1.7 Hz, 1 H),
7.53 (dd, J ϭ 7.3, J ϭ 1.3 Hz, 1 H), 7.70 (dd, J ϭ 7.3, J ϭ 1.3 Hz,
1 H) ppm. 13C{1H} NMR (CDCl3, 100 MHz): δ ϭ 17.3, 22.2, 24.0,
24.5, 34.7, 125.5, 125.8, 126.0, 128.9, 130.5, 131.2, 131.3, 131.6,
vent n-heptane, flow rate 0.4 mL·minϪ1): tR ϭ 16.43 min for (SiS)-
9c (major enantiomer), tR ϭ 19.98 min for (SiR)-9c (minor enanti-
omer). 1H NMR (CDCl3, 400 MHz): δ ϭ 1.05Ϫ1.13 (m, 1 H), 1.10
˜ ϭ 3027 (m), 2903
(d, J ϭ 6.9 Hz, 3 H), 1.29 (d, J ϭ 6.9 Hz, 3 H), 1.33Ϫ1.41 (m, 1 135.4, 136.3, 148.5, 155.4 ppm. IR (cuvette): ν
H), 1.99Ϫ2.05 (m, 2 H), 2.84Ϫ2.87 (m, 2 H), 3.11 (sept, J ϭ 6.9 Hz, (m), 2520 (m), 2360 (m), 2341 (s), 1793 (m), 1624 (m), 1540 (m),
1 H), 5.03 (t, J ϭ 3.4 Hz, 1 H), 7.10Ϫ7.40 (m, 8 H) ppm. 13C{1H} 1436 (m), 1216 (m), 1015 (m), 991 (m), 876 (w), 823 (m), 748 (m)
NMR (CDCl3, 100 MHz): δ ϭ 10.6, 22.6, 24.2, 24.6, 34.3, 35.1, cmϪ1. LRMS (EI): m/z ϭ 300 [Mϩ]. HRMS (EI): calcd. for
125.0, 125.4, 125.5, 128.9, 129.5, 130.3, 130.4, 133.3, 136.3, 136.8,
149.4, 155.5 ppm. IR (cuvette): ν˜ ϭ 3687 (m), 3010 (s), 2928 (s),
2401 (s), 2361 (s), 2123 (m), 1962 (w), 1928 (w), 1589 (m), 1522
(m), 1474 (m), 1435 (m), 1664 (m), 1224 (s), 1120 (w), 1076 (m),
1028 (w), 981 (w), 938 (m), 825 (w) cmϪ1. LRMS (CI/NH3): m/z ϭ
267 [[M ϩ H]ϩ]. C18H22Si (266.45): calcd. C 81.14, H 8.32; found
C 81.12, H 8.40.
C18H21ClSi: 300.1101; found 300.1108.
(SiS)-1-Chloro-(2-isopropylphenyl)-1,2,3,4-tetrahydro-1-silanaph-
thalene [(SiS)-7c]: According to the procedure for rac-7b, a satu-
rated solution of Cl2 in CCl4 (0.20 mL) was added to a solution of
(SiR)-9c (26.7 mg, 0.100 mmol, er ϭ 96:4) in CCl4 (1 mL) at 0 °C.
The reaction mixture was maintained for 2 min and a second por-
tion of Cl2 in CCl4 (0.05 mL) was then added. The solution was
purged with argon after 2 min. Evaporation of the solvent at re-
duced pressure provided (SiS)-7c (29.8 mg, 0.0990 mmol, 99%) as
a colourless liquid which solidified upon standing at Ϫ30 °C in n-
(SiR)-1-(2-Isopropylphenyl)-1,2,3,4-tetrahydro-1-silanaphthalene
[(SiR)-9c]: According to the preparation of (SiS)-9c, a solution of
silyl ether (SiS)-19c (793 mg, 1.89 mmol, dr ϭ 94:6) and di-n-butyl
ether (25 mL) was treated with DIBAL (5.70 mL, 5.70 mmol, 3.00
equiv., 1m in hexane) at reflux for 16 h. The usual workup and
purification by flash chromatography on silica gel (cyclohexane)
gave cyclic silane (SiR)-9c (389 mg, 1.46 mmol, 77%) as a colourless
liquid. The enantiomeric ratio of 94:6 was determined from HPLC
by integration of baseline separated peaks which were assigned as
20
20
20
pentane. [α]2D0 ϭ Ϫ20.9, [α]
ϭ Ϫ22.9, [α]
ϭ Ϫ25.2, [α]
ϭ
578
546
436
20
Ϫ46.2, [α] ϭ Ϫ72.9 (c ϭ 21.0, CHCl3).
365
Acknowledgments
The research was supported by the Deutsche Forschungsgemein-
schaft (Emmy Noether fellowship, Oe 249/2Ϫ2), the Fonds der
Chemischen Industrie and the Wissenschaftliche Gesellschaft, Frei-
burg im Breisgau. The authors thank Ilona Hauser for skilful tech-
nical assistance, and Gerd Fehrenbach and Dr. Richard Krieger
for performing the HPLC analyses. M. O. is indebted to Professor
Reinhard Brückner for his continuous support. Generous do-
nations of chemicals from Boehringer Ingelheim, Vienna, Austria,
and Haarmann & Reimer, Holzminden, Germany are gratefully
acknowledged.
the enantiomers by comparison with a racemic sample. [α]2D0
ϭ
20
20
20
20
Ϫ75.2, [α]
ϭ Ϫ78.8, [α]
ϭ Ϫ90.1, [α]
ϭ Ϫ166, [α]
436 365
ϭ
578
546
Ϫ291 (c ϭ 22.2, CHCl3).
rac-1-Chloro-(2-methylphenyl)-1,2,3,4-tetrahydro-1-silanaphthalene
(rac-7b): A saturated solution of Cl2 in CCl4 (1 mL) was added
dropwise to a vigorously stirred solution of rac-9b (138 mg,
0.580 mmol) in CCl4 (4 mL) at 0 °C until a permanent pale yellow
colour appeared. The reaction mixture was maintained for 2 min
and a second portion of Cl2 in CCl4 (0.60 mL) was added. The
yellow solution was purged with argon after 2 min. Evaporation of
the solvent under high vacuum provided rac-7b (157 mg,
0.577 mmol, 99%) as a yellowish oil which was used without further
purification. 1H NMR (CDCl3, 400 MHz): δ ϭ 1.40Ϫ1.57 (m, 2
H), 2.06Ϫ2.16 (m, 2 H), 2.40 (s, 3 H), 2.79Ϫ2.87 (m, 1 H),
2.92Ϫ2.99 (m, 1 H), 7.18Ϫ7.26 (m, 4 H), 7.36 (t, J ϭ 6.0 Hz, 2 H),
7.51 (d, J ϭ 7.3 Hz, 1 H), 7.68 (d, J ϭ 7.7 Hz, 1 H) ppm. 13C{1H}
NMR (CDCl3, 100 MHz): δ ϭ 16.4, 22.1, 23.0, 34.7, 125.2, 126.2,
129.0, 130.3, 130.6, 130.9, 135.3, 136.3, 134.9, 148.9 ppm. IR (cu-
[1]
For reviews on trivalent silyl anions see: [1a] K. Tamao, A. Ka-
[1b]
wachi, Adv. Organomet. Chem. 1995, 38, 1Ϫ58.
bert, W. J. Schulz, Jr. in The Chemistry of Organic Silicon Com-
J. B. Lam-
pounds, Part 2 (Eds.: S. Patai, Z. Rappoport), Wiley, Chichester,
[1c]
1989, p. 1007Ϫ1014.
J. Belzner, U. Dehnert in The Chemis-
try of Organic Silicon Compounds, Part 1 (Eds.: Z. Rappoport,
Y. Apeloig), Wiley, Chichester, 1998, p. 779Ϫ825.
For synthetic applications of silyl cuprates see:
[2]
[2a]
I. Fleming
in Organocopper Reagents (Ed.: R. J. K. Taylor), Oxford Aca-
194
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org Eur. J. Org. Chem. 2005, 184Ϫ195

Reductive Metallation of Asymmetrically Substituted Silyl Chlorides
FULL PAPER
E. Moreau in The Chemistry of Organic Silicon Compounds,
Part 1 (Eds.: S. Patai, Z. Rappoport), Wiley, Chichester, 1989,
p. 305Ϫ370.
[2b]
demic Press, New York, 1994, p. 257Ϫ292.
R. K. Dieter in
Modern Organocopper Chemistry (Ed.: N. Krause), Wiley-
VCH, Weinheim, 2002, p. 79Ϫ144.
For a general reference see: M. A. Brook, Silicon in Organic,
Organometallic and Polymer Chemistry, Wiley, New York, 2000.
[3]
[19]
M. Oestreich, U. K. Schmid, G. Auer, M. Keller, Synthesis
2003, 2725Ϫ2739.
[20] [20a]
[4] [4a]
H. Gilman, G. D. Lichtenwalter, J. Am. Chem. Soc. 1958,
M. W. Grant, R. H. Prince, Nature 1969, 222, 1163Ϫ1164.
80, 608Ϫ611. [4b] K. Tamao, A. Kawachi, Y. Ito, J. Am. Chem.
Soc. 1992, 114, 3989Ϫ3990. [4c] K. Tamao, A. Kawachi, Angew.
Chem. 1995, 107, 886Ϫ888; Angew. Chem. Int. Ed. Engl. 1995,
34, 818Ϫ820.
[20b] M. W. Grant, R. H. Prince, J. Chem. Soc. Chem. Commun.
[20c]
1968, 1076Ϫ1077.
M. W. Grant, R. H. Prince, J. Chem.
Soc. (A) 1969, 1138Ϫ1142.
[21] [21a]
A. R. Bassindale, J. C.-Y. Lau, P. G. Taylor, J. Organomet.
Chem. 1988, 341, 213Ϫ224. [21b] A. R. Bassindale, J. C.-Y. Lau,
P. G. Taylor, J. Organomet. Chem. 1995, 490, 75Ϫ82. [21c] A. R.
Bassindale, J. C.-Y. Lau, P. G. Taylor, J. Organomet. Chem.
1995, 499, 137Ϫ141.
A. R. Bassindale, S. J. Glynn, P. G. Taylor in The Chemistry
of Organic Silicon Compounds, Part 1 (Eds.: Z. Rappoport, Y.
Apeloig), Wiley, Chichester, 1998, vol. 2, p. 495Ϫ511.
For a halide-induced epimerization see: K. A. Trankler, J. Y.
Corey, N. P. Rath, J. Organomet. Chem. 2003, 686, 66Ϫ74.
E. P. A. Couzijn, M. Schakel, F. J. J. de Kanter, A. W. Ehlers,
M. Lutz, A. L. Spek, K. Lammertsma, Angew. Chem. 2004,
116, 3522Ϫ3524; Angew. Chem. Int. Ed. 2004, 43, 3440Ϫ3442.
K. Suzuki, Y. Kawakami, Organometallics 2003, 22,
2367Ϫ2369.
[5] [5a]
´
J. B. Lambert, M. Urdaneta-Perez, J. Am. Chem. Soc. 1978,
[5b]
100, 157Ϫ162.
1981, 37, 1105Ϫ1112.
J. Phys. Chem. 1988, 92, 33Ϫ36.
A. C. Hopkinson, M. H. Lien, Tetrahedron
[5c]
J. R. Damewood Jr., C. M. Hadad,
[5d]
M. Flock, C. Marschner,
Chem. Eur. J. 2002, 8, 1024Ϫ1030. [5e] A. Kawachi, H. Maeda,
K. Tamao, Organometallics 2002, 21, 1319Ϫ1321.
C. Strohmann, M. Bindl, V. C. Fraaß, J. Hörnig, Angew. Chem.
2004, 116, 1029Ϫ1032; Angew. Chem. Int. Ed. 2004, 43,
1011Ϫ1014.
[22]
[6]
[23]
[24]
[7] [7a]
L. H. Sommer, R. Mason, J. Am. Chem. Soc. 1965, 87,
[7b]
1619Ϫ1620.
E. Colomer, R. J. P. Corriu, J. Organomet.
Chem. 1977, 133, 159Ϫ168.
[8]
[9]
[25]
[26]
M. Omote, T. Tokita, Y. Shimizu, I. Imae, E. Shirakawa, Y.
Kawakami, J. Organomet. Chem. 2000, 611, 20Ϫ25.
C. Strohmann, J. Hörnig, D. Auer, Chem. Commun. 2002,
766Ϫ767.
The intermediacy of trisubstituted siliconium ions has been
ruled in the past. However, pentacoordinate siliconium ions
stabilised by two solvent molecules are probable intermediates:
[10] [10a]
I. Fleming, R. S. Roberts, S. C. Smith, J. Chem. Soc., Per-
kin Trans. 1 1998, 1209Ϫ1214. [10b] N. A. Rahman, I. Fleming,
L. H. Sommer, F. O. Stark, K. W. Michael, J. Am. Chem.
[26a]
[10c]
A. B. Zwicky, J. Chem. Res. (S) 1992, 292.
M. V. George,
Soc. 1964, 86, 5683Ϫ5684. [26b] R. Corriu, M. Leard, J. Masse,
[26c]
D. J. Peterson, H. Gilman, J. Am. Chem. Soc. 1960, 82,
Bull. Soc. Chim. Fr. 1968, 2555Ϫ2562.
R. J. P. Corriu, F.
403Ϫ406.
Larcher, G. Royo, J. Organomet. Chem. 1976, 104, 293Ϫ301.
G. Köbrich, H. Trapp, Chem. Ber. 1966, 99, 680Ϫ688.
It should be noted that the reductive cleavage of diastereomeric
disilanes could potentially proceed at different rates with one
of the two disilanes not being cleaved at all. Hence, it appears
almost impossible to draw any conclusion from a consideration
of the absolute configuration and enantiomeric ratio of the silyl
anions obtained. In particular, this has to be taken into ac-
count for low-yielding metallation reactions.
H.-S. Oh, I. Imae, Y. Kawakami, S. S. S. Raj, T. Yamane, J.
Organomet. Chem. 2003, 685, 35Ϫ43.
CCDC-249127 [for (SiS)-9c] and CCDC-249128 [for (SiR)-7c]
contain the supplementary crystallographic data for this paper.
These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data—request/cif.
[11]
[27]
[28]
C. Strohmann, O. Ulbrich, D. Auer, Eur. J. Inorg. Chem.
2001, 1013Ϫ1018.
[12] [12a] D. E. Seitz, L. Ferreira, Synth. Commun. 1979, 9, 451Ϫ456.
[12b]
P. Boudjouk, B. H. Han, Tetrahedron Lett. 1981, 22,
[12c]
3813Ϫ3814.
P. D. Lickiss, R. Lucas, J. Organomet. Chem.
[13] [13a]
[13b]
Lithium naphthalide.
[14b]
[29]
[30]
C. Chatgilialoglu, Organosilanes in Radical Chemistry, Wiley,
New York, 2004.
[15] [15a]
F. C. Cartledge, R. V. Piccione, Organometallics 1984, 3,
299Ϫ305. [15b] A. C. Hopkinson, C. F. Rodriquez, M. H. Lien,
[15c]
Can. J. Chem. 1990, 68, 1309Ϫ1316.
M. Guerra, J. Am.
Chem. Soc. 1993, 115, 11926Ϫ11929.
[16] [16a]
[31]
[32]
H. Sakurai, M. Murakami, M. Kumada, J. Am. Chem.
A. V. Yakimansky, A. H. E. Müller, M. Van Beylen, Macromol-
ecules 2000, 33, 5686Ϫ5692.
Metallations of silyl chloride rac-7b using Li or LiDBB were
unsuccessful due to partial proton abstraction in the benzylic
position of the o-tolyl substituent as shown by deuteriolysis.
[16b]
Soc. 1969, 91, 519Ϫ520.
A. G. Brook, J. M. Duff, J. Am.
[16c]
Chem. Soc. 1969, 91, 2118Ϫ2119.
H. Sakurai, M. Murak-
ami, Chem. Lett. 1972, 7Ϫ8.
[17] [17a]
L. H. Sommer, Stereochemistry, Mechanism and Silicon,
[17b]
[33] [33a]
McGraw-Hill, New York, 1965.
L. H. Sommer, Intra-Sci-
L. H. Sommer, O. F. Bennett, J. Am. Chem. Soc. 1957, 79,
[33b]
ence Chem. Rept. 1973, 7, 1Ϫ44.
1008Ϫ1009.
Soc. 1959, 81, 251Ϫ252.
Am. Chem. Soc. 1973, 95, 7700Ϫ7707.
Kapfer, R. F. Cunico, Organometallics 1983, 2, 336Ϫ343.
Received September 1, 2004
L. H. Sommer, O. F. Bennett, J. Am. Chem.
[18] [18a]
[33c]
´
R. J. P. Corriu, C. Guerin, J. J. E. Moreau in Topics in
Stereochemistry (Ed.: E. L. Eliel), Wiley, New York, 1984, vol.
G. D. Homer, L. H. Sommer, J.
[33d]
P. Boudjouk, C. A.
15, p. 43Ϫ198. [18b] R. J. P. Corriu, C. Guerin, Adv. Organomet.
´
[18c]
´
Chem. 1982, 20, 265Ϫ312.
R. J. P. Corriu, C. Guerin, J. J.
Eur. J. Org. Chem. 2005, 184Ϫ195
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
195