Mechanistic insights into the reaction of enantiomerically pure lithiosilanes and electrophiles: Understanding the differences between aryl and alkyl halides

Article abstract of DOI:10.1002/ejic.201000834

The reactivity of enantiomerically pure lithiosilanes against aliphatic and aromatic halo electrophiles has been investigated with the focus on product composition and enantiomeric ratios as a basis for a better understanding of ongoing mechanisms. Both parameters are strongly influenced by both the organic group and the corresponding halide of the used electrophile. Thus, high stereoselectivities and yields are only obtained if one distinct reaction mechanism dominates. In the case of aliphatic electrophiles, experimental and quantum chemical studies support the preference of an SN2 mechanism for chlorides resulting in retention of configuration on silicon, whereas bromides tend to react under inversion through a halide-lithium exchange (ate complex). For aromatic electrophiles these relationships change: chlorides and bromides both favor the formation of an ate complex over nucleophilic aromatic substitution, leading to inversion of configuration on silicon.

Full text of DOI:10.1002/ejic.201000834

FULL PAPER
DOI: 10.1002/ejic.201000834
Mechanistic Insights into the Reaction of Enantiomerically Pure Lithiosilanes
and Electrophiles: Understanding the Differences between Aryl and Alkyl
Halides
Christian Däschlein,[a] Simeon O. Bauer,[a] and Carsten Strohmann*[a]
Keywords: Silicon / Lithium / Silanes / Lithiosilanes / Substituent effects / Reaction mechanisms / Chirality
The reactivity of enantiomerically pure lithiosilanes against
aliphatic and aromatic halo electrophiles has been investi-
gated with the focus on product composition and enantio-
meric ratios as a basis for a better understanding of ongoing
mechanisms. Both parameters are strongly influenced by
both the organic group and the corresponding halide of the
used electrophile. Thus, high stereoselectivities and yields
are only obtained if one distinct reaction mechanism domi-
nates. In the case of aliphatic electrophiles, experimental and
quantum chemical studies support the preference of an SN2
mechanism for chlorides resulting in retention of configura-
tion on silicon, whereas bromides tend to react under inver-
sion through a halide–lithium exchange (ate complex). For
aromatic electrophiles these relationships change: chlorides
and bromides both favor the formation of an ate complex
over nucleophilic aromatic substitution, leading to inversion
of configuration on silicon.
cal intermediats or other substitution mechanisms are pos-
sible so that these reactions do not necessarily proceed un-
der preservation of the stereoinformation. Depending on
the subsiding mechanisms, stereochemically pure, enriched,
or racemic products may be obtained. For instance, our
group was able to show that the reaction of enantiomer-
ically pure lithiosilane (S)-2 with benzyl chloride or benzyl
bromide results in the formation of the (R)- and (S)-config-
uration, respectively, as the main stereoisomer in the formed
product, only depending on the halide used [Scheme 1; en-
antiomeric ratios are given as (R)/(S)].[10a]
Introduction
The reaction of organolithium reagents with electrophiles
is amongst the most important reactions for the synthesis
of a multitude of functionalized systems in almost all fields
of chemistry.[1] Although a variety of reactions have been
intensively investigated to date,[2] this method can generally
not be used for the synthesis of enantio- or dia-
stereomerically enriched and pure systems due to the rela-
tive high configurative lability of lithiated carbon cen-
ters.[3,4–6] In contrast, due to hybridization defects from the
third period on,[7,8] stereogenic lithiated silicon centers
(lithiosilanes)[9,10] have been shown to be suitable for the
introduction of stereoinformation and the synthesis of
stereochemically enriched and pure products.[11] During the
last decade, the reaction of highly enantiomerically enriched
lithiosilanes[12,13] with chlorosilanes and chlorogermanes
has proven to be an elegant method for stereo-
selective transformations.[12d,12e,14] Nevertheless, reactions
of lithiosilanes with functionalized halo electrophiles have
been of little interest and not well investigated due to low
yields and often undesired side-products (e.g., those ob-
tained through Si–Si or C–C coupling reactions due to hal-
ide–lithium exchange).[15] Furthermore, for the reaction of
lithiosilanes and organic electrophiles (or the formed halo-
silane/lithiumorganyl after halide–lithium exchange),
pathways by nucleophilic substitution but also by e.g. radi-
Scheme 1. Reaction of the enantiomerically pure lithiosilane (S)-2
with benzyl chloride and benzyl bromide: different mechanisms
lead to different main stereoisomers.[19,20] The er values refer to
(R)/(S), expect for (S)-2.
Thereby, the stereochemical probe at silicon could be
used to gain crucial insights into the reaction mechanisms.
Whereas the reaction with the chloride preferably proceeds
through an SN2 mechanism under retention of the absolute
configuration on silicon, based on the reactant (R)-1,[16] the
[a] Anorganische Chemie, Technische Universität Dortmund,
Otto-Hahn-Straße 6, 44227 Dortmund, Germany
Fax: +49-231-755-7063
E-mail: mail@carsten-strohmann.de
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201000834.
1454
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 1454–1465

Reaction of Enantiomerically Pure Lithiosilanes and Electrophiles
reaction with the bromide preferably proceeds through a der reduced pressure, the oily residue was dissolved in a
halide–lithium exchange (ate complex)[17,18] with inversion minimum amount of n-pentane and separated from all salts.
of configuration, based on the reactant (R)-1. Quantum The obtained product mixtures were then investigated by
chemical calculations have been used to confirm both pro- using NMR spectroscopy and GC/MS analysis without fur-
posed mechanisms.[19]
ther purification.[27]
Information concerning the obtained product composi-
tion as well as the enantiomeric ratio of the product and
therefore the dominating reaction mechanisms are of cru-
cial necessity for a wider application of enantiomerically
pure lithiosilanes and the synthesis of chiral silicon com-
pounds. The synthesis of chiral silicon compounds is re-
cently gaining importance, as silanes with a stereogenic sili-
con center can be used as auxiliaries for stereoselective syn-
theses, such as Mannich reactions,[21] cycloaddition reac-
tions,[22] or Friedel–Crafts alkylations.[23,24]
As part of our studies on functionalized lithiosilanes, we
present our latest results concerning the reactions of enan-
tiomerically pure lithiosilanes and halo electrophiles. The
systematic variation of the halide, as well as the organic
group, in combination with quantum chemical studies was
used to answer the three most important questions
(Scheme 2): (i) What products are formed and in what ratio
(product composition)?; (ii) is the stereoinformation re-
tained, inverted, or lost during the reaction process (enan-
tiomeric ratio)?; and (iii) what is the preferred reaction
mechanism?
Scheme 3. Synthesis of the functionalized silanes 5, 6, and 7; X =
Cl, Br, I; R = Me3SiCH2, Me3CCH2, PhSCH2.
These reactions were also carried out with rac-1 and the
corresponding chloro-organyl as electrophile to synthesize
the racemic compounds (reaction time: 6 h[28]). The crude
products were purified yielding silanes rac-5, rac-6, and rac-
7. The racemic silanes were used to perform NMR spectro-
scopic studies in the presence of (R)-mandelic acid (for 5
and 7) and (2R,3R)-di-O-benzoyltartaric acid (for 6). Thus,
a method for the separation of the different stereoisomers
and therefore a possible method to analyze the enantio-
meric ratios was developed, which was then utilized in the
reactions of the enantiomerically pure system.[29] In the fol-
lowing, we will first describe the results for each electrophile
(R–CH2–X; X = halide, R = functional group = SiMe3,
CMe3, SPh) before comparing the different trapping rea-
gents.
Scheme 2. Reaction of enantiomerically pure lithiosilanes with halo
electrophiles and the main questions concerning these transforma-
tions.
Results and Discussion
Reaction of Lithiosilane 2 with Aliphatic Halo Electrophiles
A) Reaction of 2 with Me3SiCH2X
For a systematic study of the reactivity of enantiomer-
The reaction of rac-2 and (chloromethyl)trimethylsilane
ically pure lithiosilanes and halo electrophiles, we first in- yielded rac-5 in good yield (86%) after distillation. Silane 8
vestigated the reaction of the enantiomerically pure lithio- was the only by-product obtained from the reaction and is
silane 2 with aliphatic electrophiles. Based on the trapping due to the reaction of 4 with Me3SiCH2Cl. In contrast, re-
reaction of 2 with benzyl chloride and benzyl bromide and action of the lithiosilane with the corresponding bromo-
its differing main stereoisomers,[10a] we have chosen the gen- and iodo-compounds yielded not only compounds 5 and 8,
erally related (halomethyl)trimethylsilanes, neopentyl ha- but also silanes 1 and 9 obtained by a Si–Si bond-coupling
lides, and (chloromethyl)phenyl sulfide as trapping reagents reaction. Table 1 summarizes the product distribution (the
(see below for more details on the choice of trapping rea- compounds were identified by GC/MS; their ratio was de-
gents). First, disilane (R)-1 was reacted at –78 °C in thf with termined by NMR spectroscopy).
elemental lithium to form enantiomerically pure (S)-2 after
Consequently, the trapping of 2 with (chloromethyl)tri-
a reaction time of 8 h.[12d,25] To monitor the completeness methylsilane is the only reaction to proceed without the for-
of the Si–Si bond cleavage, a small part of the solution of mation of multiple by-products. Noticeably, the disilane 10,
lithiosilane 2 was trapped with Me3SiCl and investigated a possible by-product formed through Si–Si coupling (Fig-
by NMR spectroscopy.[26] On complete formation of 2, the ure 1) was not identified in any case.
solution of the lithiosilane was divided into three parts and
The highly selective reaction of 2 and the chloride is also
each added to a twofold excess of the corresponding trap- confirmed by the stereochemical course of the reaction de-
ping reagent (Scheme 3). After removal of all volatiles un- termined by NMR spectroscopy of 5 in the presence of
Eur. J. Inorg. Chem. 2011, 1454–1465
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1455

C. Däschlein, S. O. Bauer, C. Strohmann
FULL PAPER
Table 1. Identified products and proportions [%] of the reaction of
Table 2. Identified products and proportions [%] of the reaction of
rac-2 with (halomethyl)trimethylsilanes.[a,b]
rac-2 with neopentyl halides.[a,b]
Identified products [%]
Trapping reagent
5
8
1
9
(Chloromethyl)trimethylsilane
(Bromomethyl)trimethylsilane
(Iodomethyl)trimethylsilane
50
36
43
50
18
28
n.i.
34
23
n.i.
12
6
Identified products [%]
Trapping reagent
6
11
1
9
12
13
[a] n.i.: not identified. [b] The given values are obtained from NMR
spectroscopic integration.[30]
Neopentyl chloride
Neopentyl bromide
Neopentyl iodide
36
32
29
25
9
traces
21
37
49
15
22
17
1.3
n.i.
2.6
1.7
n.i.
2.3
[a] n.i.: not identified. [b] The given values are obtained from NMR
spectroscopic integration.[30]
Significantly lower stereoselectivities for the reactions of
2 with the neopentyl halides – in comparison with the (hal-
omethyl)trimethylsilanes – could be determined by 1H
NMR spectroscopic studies in the presence of 1.5 equiv. of
(2R,3R)-di-O-benzoyltartaric acid in CDCl3. Under these
conditions, the resonance signals of the methyl groups at
the stereogenic silicon centers could be integrated sepa-
rately. Thus, the reaction of enantiomerically pure 2 with
Me3CCH2Cl yielded a ratio [(R)/(S)] of 55:45, Me3CCH2Br
yielded a ratio of 39:61, and Me3CCH2I of 45:55. Although
the formation of the opposite stereoisomer is again favored
for the chloride, all three trapping reagents strongly tend to
racemization of the stereogenic silicon center. As a conse-
quence, none of the reactions of 2 with neopentyl halides
proceed by the preference of one distinct mechanism.
Figure 1. Compound 10 – not identified from the reactions of rac-
2 with trimethylsilanes.
three equivalents of (R)-mandelic acid in CDCl3. Under
these conditions, the resonance signals of the methyl groups
of the SiMe3 group could be integrated separately in the 1H
NMR spectrum. Thereby, the reaction of enantiomerically
pure 2 with Me3SiCH2Cl yielded an enantiomeric ratio er –
(R)/(S) – of 97:3, whereas the reaction with Me3SiCH2Br
yielded a ratio of 22:78. The reaction of enantiomerically
pure 2 with Me3SiCH2I yielded a ratio of only 39:61. Thus,
the reaction with the chloride resulted in the formation of
the opposing stereochemical configuration to that of the
bromide and the iodide. This is a crucial observation and
hints at the viability of different reaction mechanisms for
the three halides. Scheme 4 summarizes the described re-
sults; all relevant 1H NMR spectra can be found in the Sup-
porting Information.
1
Scheme 5 summarizes the described results; all relevant H
NMR spectra can be found in the Supporting Information.
Scheme 5. Summary of the er in the reaction of enantiomerically
pure 2 with Me3CCH2X (X = Cl, Br, I). The er values refer to
(R)/(S).
Scheme 4. Summary of the er in the reaction of enantiomerically
pure 2 with Me3SiCH2X (X = Cl, Br, I). The er values refer to
(R)/(S).
C) Reaction of 2 with PhSCH2Cl
B) Reaction of 2 with Me3CCH2X
Finally, we were specifically interested in the influence of
Reaction of rac-2 and neopentyl chloride yielded rac-6 a sulfur atom in the α-position to the halide due to the
(39%) and several by-products. In addition to silane 11, high relevance of sulfur-containing compounds in various
which is formed due to the reaction of 4 with Me3CCH2Cl, reactions.[31] The reaction of rac-2 with (chloromethyl)-
disilanes 1 and 9 as well as silanes 12 and 13 were identified phenyl sulfide yielded rac-7 (28%) and the same by-prod-
in the reaction mixture. Almost all of these by-products ucts as detected for the reaction of 2 with neopentyl halides
were identified in the reaction of 2 with both Me3CCH2Br (silane 14 is formed instead of 11). Table 3 summarizes the
and Me3CCH2I; the amount of the 1 increases from the obtained product compositions. A high amount of com-
chloride to the iodide. Again, the Si–Si coupling product 10 pound 1 was formed in the reaction with the sulfide; the
(Figure 1) was not formed in any trapping reaction. Table 2 reason for this is currently under investigation. Although
summarizes the product distribution.
the yield is lower than in the case of the neopentyl chloride,
1456
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 1454–1465

Reaction of Enantiomerically Pure Lithiosilanes and Electrophiles
the reaction of enantiomerically pure 2 with PhSCH2Cl re- Table 4. Enantiomeric ratios obtained from the reactions of 2 with
R–CH2–X; X = halide, R = functional group = Me3Si, Me3C, PhS,
sulted in a considerably better er value [(R)/(S)] of 82:18
for compound 7, determined by the treatment of the crude
Ph[a] [er values based on (R)-1; (R)/(S)].
Trapping reagent
PhCH2X
er (chloride)
94:6[a]
97:3
82:18
55:45
er (bromide) er (iodide)
product with three equivalents of (R)-mandelic acid and
separate integration of the signals of the methyl group at
5:95[a]
22:78
n.d.
n.d.[b]
39:61
n.d.
the stereogenic silicon center. Scheme 6 summarizes the de- Me3SiCH2X
1
PhSCH2X
Me3CCH2X
[a] See ref.[10a] [b] n.d.: not determined.
scribed results; all relevant H NMR spectra can be found
39:61
45:55
in the Supporting Information.
Table 3. Identified products and proportions [%] of the reaction of
of rac-2 with (chloromethyl)phenyl sulfide.[a]
ii) Starting with the benzyl halides, there is a decrease of
stereoselectivity over the trimethylsilyl halides and (chlo-
romethyl)phenyl sulfide to the neopentyl halides.[32]
iii) High stereoselectivities – benzyl chloride, benzyl bro-
mide, (chloromethyl)trimethylsilane – hint to the strong
domination of one distinct reaction mechanism, further
confirmed by the lack of by-products in the GC/MS analy-
sis. In all other cases, additional, competing mechanisms
have to exist.
R–CH2–Cl and R–CH2–Br (R = Me3Si, Me3C, PhS, Ph)
Identified products [%]
Trapping reagent
7
14
1
9
12
13
In general, the preferred reaction mechanism should be
identical for chemically related trapping reagents. In 2004,
our group showed that for the reaction of enantiomerically
pure 2 and benzyl chloride, an SN2 mechanism on carbon,
resulting in retention of configuration on silicon based on
(R)-1, is preferred. In contrast, for the reaction of enantio-
merically pure 2 and benzyl bromide, a halide–lithium ex-
change via an ate complex leading to inversion of configu-
ration on silicon is preferred.[10a] Scheme 7 displays these
two main reaction mechanisms with the three newly investi-
gated trapping reagents already included (see Scheme 1 for
the formation of 2).[20] It is important for the mechanism
through the ate complexes, that both experimental[9,10a] and
quantum chemical calculations[33] support the selective re-
action of lithiumorganyls with chloro- and bromosilanes
(Chloromethyl)phenyl sulfide 25
13
40
19
1.4 1.6
[a] The given values are obtained from NMR spectroscopic integra-
tion.[30]
Scheme 6. Summary of the er in the reaction of enantiomerically
pure 2 with PhSCH2Cl. The er values refer to (R)/(S).
D) Comparison of the Different Trapping Reagents: New
Mechanistic Insights
The (halomethyl)trimethylsilanes, neopentyl halides, and from the back side under inversion at silicon. This assump-
(chloromethyl)phenyl sulfide have been chosen due to their tion is the basis for the following discussion.[34] It also has
similarity to the benzyl halides that have been previously to be emphasized that, due to the change of substituents on
investigated;[10a] all reagents are able to stabilize a negative silicon, the priorities of the CIP nomenclature change and
charge in the α-position to the corresponding substituent R thus the absolute configurations are inverted in some cases,
(R–CH2–X; X = halide, R = Me3Si, Me3C, PhS, Ph). As although no actual change of the stereoinformation takes
this property is differently pronounced in the investigated place.[20]
systems, we focused on the influence of this difference on
To investigate the mechanisms for the newly studied trap-
the obtained product composition and stereoselectivity. Re- ping reagents, we carried out quantum chemical calcula-
lated to this is the question of the relevant mechanisms for tions on model systems for the reaction of 2 with (chloro-
a better understanding of the resulting enantiomeric ratios, methyl)trimethylsilane and (bromomethyl)trimethylsilane,
and for further more controlled synthetic applications. (both for the ate complex and the SN2 mechanism). There-
Table 4 summarizes all experimentally determined er values fore, the geometries of the stationary points of H3Si–/Me3-
of the reaction of enantiomerically pure 2 with aliphatic SiCH2Cl and H3Si–/Me3SiCH2Br were optimized using
halo electrophiles; the reaction of enantiomerically pure 2 density functional theory with the hybrid B3LYP functional
with benzyl halides is included for completeness, see ref.[10a] and the 6-31+G(d) basis set. Since SN2 reactions are
Based on these results, the following conclusions can be strongly influenced by solvent effects, calculations including
drawn:
solvent models were performed. The stationary points were
i) In all cases, the reaction of enantiomerically pure 2 with energy optimized in thf solution (ε = 7.58 for thf) by means
the corresponding chloride results in the preferential forma- of the self-consistent reaction field (SCRF) method. The
tion of the opposite stereoisomer to the bromide and iodide, energies for the structures obtained were calculated with the
indicating different mechanisms.
polarizable conductor calculation model (CPCM).[16,35]
Eur. J. Inorg. Chem. 2011, 1454–1465
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1457

C. Däschlein, S. O. Bauer, C. Strohmann
FULL PAPER
the two possible transition states is even higher for Me3-
SiCH2Cl (ΔE = 34 kJmol–1) than for PhCH2Cl (ΔE =
13 kJmol–1), explaining the experimentally obtained,
slightly higher er for the silicon-containing trapping rea-
gent. For (bromomethyl)trimethylsilane these energy rela-
tionships are reversed and the bromide–lithium exchange
(ate-Br) is favored by 2 kJmol–1 over the substitution (SN2-
Br). Although this energetic difference is only very small,
this is in accordance with the experimentally obtained lower
er [(R)/(S) = 22:78]. Furthermore, this value is also signifi-
cantly smaller than the er obtained for the reaction of 2
and PhCH2Br [(R)/(S) = 5:95][10a] in which the ate complex
lies 10 kJmol–1 below the substitution (for energy values,
see Table 5).
Table 5. Relative energies Erel of the calculated stationary points
for the model systems H3Si–/PhCH2Cl, H3Si–/PhCH2Br, H3Si–/
Me3SiCH2Cl, and H3Si–/Me3SiCH2Br.[a,b]
Model system
Nucleophilic
Halide–lithium
substitution (SN2) exchange (ate complex)
[kJmol–1]
[kJmol–1]
Scheme 7. Preferred mechanisms for the reactions of lithiosilane
H3Si–/PhCH2Cl
0[c]
10[c]
0
13[c]
0[c]
34
(S)-2 with chloro (top) and bromo electrophiles (bottom).[20]
H3Si–/PhCH2Br
H3Si–/Me3SiCH2Cl
H3Si–/Me3SiCH2Br
2
0
Harmonic vibrational frequency analyses – to establish the
nature of the stationary points – were performed on the
same level, showing one imaginary frequency each. Figure 2
displays these calculated stationary points and their relative
energies.
[a] Values not relative to each other. [b] B3LYP/6-31+G(d), CPCM.
[c] See ref.[10a]
Thus, the DFT calculations of the model systems con-
firm all trends observed in the experiment. A SN2 mecha-
nism with retention of configuration on silicon is preferred
for the reaction of enantiomerically pure lithiosilanes with
chlorohalides, whereas a halide–lithium exchange under in-
version of configuration on silicon is preferred for the reac-
tion with bromides. Furthermore, an important conclusion
can be drawn based on the decreasing stereochemical purity
(simplified: er PhCH2X Ն Me3SiCH2X Ͼ PhSCH2X ϾϾ
Me3CCH2X, X = Cl, Br) for all investigated chlorides and
bromides; the higher the ability of the used organyl to stabi-
lize the negative charge of the respective transition state
(pentacoordinate for the SN2 mechanism, ate complex for
the halide–lithium exchange), the higher the obtained er
value of the main product. In the case of PhCH2Ph, this
negative charge is highly stabilized by the phenyl group,
whereas it is the silicon α-effect for Me3SiCH2X. This stabi-
lization is still pronounced (although smaller) for PhSCH2X
Figure 2. Calculated stationary points for the studied model sys- (negative hyperconjugation of sulfur), but almost no stabili-
tems H3Si–/Me3SiCH2Cl (left) and H3Si–/Me3SiCH2Br (right)
zation is present in the case of Me3CCH2X. The smaller
[B3LYP/6-31+G(d), CPCM]; in both cases, an ate complex (ate-X,
this stabilizing effect, the less pronounced the preferred
X = Cl, Br) and the SN2 mechanism (SN2-X, X = Cl, Br) were
mechanism and competing reaction mechanisms become
important.
optimized; Molekel plots.[36]
From these calculations, the following trends for the
model systems studied were obtained: in case of (chlo-
romethyl)trimethylsilane, the transition state ate-Cl for a
Iodides
The reaction of enenatiomerically pure 2 and aliphatic
chloride–lithium exchange lies 34 kJmol–1 higher than the iodides as trapping reagents provided worse results con-
alternative path through direct substitution of the halide sidering both the chemoselectivity and the enantiomeric ra-
(SN2-Cl). The preference of the SN2 reaction is in agree- tios (predominantly racemization in the investigated sys-
ment with our previous calculations for the reaction of (S)- tems). Thus, there is not one dominating reaction mecha-
2 and benzyl chloride.[10a] The energetic difference between nism for R–CH2–I (R = Me3Si, Me3C). This loss of stereo-
1458
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 1454–1465

Reaction of Enantiomerically Pure Lithiosilanes and Electrophiles
information was also observed in our group in previous in-
In addition to the above reactions of 2 with electrophiles,
vestigations and was attributed to the formation of silyl rad- we also wanted to investigate the reactivity of the enantio-
icals:[37] To elucidate a competing radical mechanism in merically pure lithiosilane 21 with halobenzenes to enable
these reactions, enenatiomerically pure 2 was treated with us to expand our studies to aromatic electrophiles. The
cyclopropylmethyl halides, which are commonly used in the questions about the preferred mechanism, the resulting
literature as radical scavengers.[38] These studies yielded – in products and their absolute configurations in the reaction
addition to products formed through polar mechanisms – of lithiosilanes and aromatic electrophiles are even less ex-
a high amount of products formed through radical mecha- plored than the described reactions of 2 with aliphatic trap-
nisms. Thus, the amount of the products formed by the po- ping reagents. To get a deeper insight into these reactions,
lar mechanism is strongly decreasing from the chloride to disilane (R)-20 was reacted at –50 °C in thf with elemental
the iodide, thus indicating the preference of radical mecha- lithium to form enenatiomerically pure 21 after a reaction
nisms for the reaction of iodo electrophiles (preference of a time of 6.5 h.[41] To determine complete cleavage of the S–
mechanism through the silyl radicals: R–I Ͼ R–Br Ͼ R– C bond, a portion of the solution of lithiosilane 21 was
Cl). This is also in agreement with comparable studies per- trapped with Me3SiSiMe2Cl and subjected to NMR spec-
formed by Krusic et al.[39] Under consideration of these re- troscopy.[42] On complete formation of 21, the solution of
sults, the low stereochemical purities in the reaction of (S)- the lithiosilane was divided into three parts and each added
2 and R–CH2–I can be understood, although, in principle, to a twofold excess of PhX (X = Cl, Br, I) (Scheme 9). After
a mechanisms via an ate complex, like that for the bro- removal of all volatiles under reduced pressure, the oily resi-
mides, should also be dominant for the iodides the forma- due was dissolved in a minimum amount of n-pentane and
tion of a silyl radical in the first step seems to get almost separated from all salts.
equal importance. Whereas the reaction through an ate
complex results in the formation of the product under inver-
sion of configuration on silicon, the reaction through a silyl
radical proceeds under retention.[40] Thus, no significant
stereochemical enrichment, but almost complete racemiza-
tion, is obtained in the corresponding products (slight pref-
erence of the ate complex).
Scheme 9. Reaction of (S)-21 with halobenzenes; X = Cl, Br, I. The
er values refer to (R)/(S).
The product mixtures were subject to investigation by
NMR spectroscopy and GC/MS analysis without further
purification.[27] Disilane 20 and biphenyl (by-product due
to the reaction of PhLi and PhX) could be identified in all
three reactions almost exclusively (only small amounts of
one additional compound with a molecular ion peak of m/z
= 367 could be identified). Thus, the first crucial observa-
tion is that the reaction of enantiomerically pure 21 with
PhCl, PhBr, and PhI is highly selective with regard to the
yield of the major product in each case. The determination
of the er was accomplished by using three equivalents of
(R)-mandelic acid (all relevant 1H NMR spectra can be
found in the Supporting Information).[12e,43] The trapping
reaction of (S)-21 with PhCl [(R)/(S) = 22:78] as well as
with PhBr (30:70) both resulted in the preference of the (S)-
enantiomer,[44] formed under inversion of configuration on
silicon; an almost full racemization was obtained for PhI
(56:44). Thus, contrary to the reactions of enantiomerically
Reaction of Lithiosilane 21 with Halobenzenes
In 2007, we reported the synthesis of the enantiomeri-
cally pure disilane (R)-20 through the reaction of enenatio-
merically pure 2 with chlorotrimethylsilane (Scheme 8). The
subsequent reaction of (R)-20 with elemental lithium af-
forded the selective cleavage of the Si–C bond to the phenyl
group forming the enantiomerically pure lithiosilane 21,
which could then be reacted with chlorosilanes. This led to
the first enantiomerically pure tetrasilane (R)-22 synthe-
sized under full preservation of the stereoinformation on
silicon.[12e]
pure
2
with aliphatic chloro and bromo electro-
philes, the trapping reaction of (S)-21 with the aromatic ha-
lides (PhCl and PhBr) obviously proceeds through the same
preferred reaction mechanism leading to the same main en-
antiomer.
How can these selectivities be understood? Contrary to
the reaction of lithiosilanes and aliphatic organyls, an SN2
mechanism is of no relevance for aromatic systems.
Furthermore, as the lithiosilane itself is reacting as a nu-
cleophile, these reactions cannot be discussed by electro-
philic aromatic substitution. The two most likely mecha-
Scheme 8. Synthesis of the enantiomerically pure lithiosilanes 2 and
21 through selective Si–Si and selective Si–C cleavage in the respec-
tive reactant and their trapping reactions with chlorosilanes. The
er values refer to (R)/(S).
Eur. J. Inorg. Chem. 2011, 1454–1465
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1459

C. Däschlein, S. O. Bauer, C. Strohmann
FULL PAPER
nisms are the reaction through an ate complex (proceeding
under inversion of configuration based on (R)-20; see also
Scheme 7) or through a Meisenheimer complex analogue
intermediate (proceeding under retention of configuration;
Scheme 10).
Figure 3. Calculated stationary points for the studied model sys-
tems H3Si–/PhCl (left) and H3Si–/PhBr (right) [B3LYP/6-31+G(d),
CPCM]; in both cases, an ate complex and a reaction via an inter-
mediate analogous to a Meisenheimer complex were optimized;
Molekel plots.[36]
case of PhCl as the aromatic system should, in general, sta-
bilize radicals (also for chloride). Consistent with the less-
preferred reaction of chlorides through a radical mecha-
nism, PhCl yielded a better er than PhBr. The er for PhI
[(R)/(S) = 56:44] can be explained as a result of the almost
equal importance of the ate complex and the silyl radical,
thus the product is obtained almost in racemic form. The
slightly higher amount of the (R)-enantiomer (retention)
could be the result of a small preference of the mechanism
via the silyl radical.
Scheme 10. Possible mechanisms of the reaction of lithiosilane 21
with aromatic electrophiles: via an ate complex under inversion of
configuration on silicon (top) and via an intermediate analogous
to a Meisenheimer complex under retention of configuration on
silicon (bottom).
For a better understanding of these reactions, we per-
formed quantum chemical calculations [B3LYP/6-31+G(d)]
on model systems for the reaction of 21 with PhCl and
Conclusion
We have presented the reaction of enantiomerically pure
PhBr (both the ate complex and Meisenheimer complex an- lithiosilanes with both aliphatic and aromatic halo electro-
alogue). The geometries of the stationary points of H3Si–/ philes to gain a better understanding of the reaction mecha-
PhCl and H3Si–/PhBr were optimized by using the same nisms involved. Based on the stereochemical probe encoded
procedure [SCRF, CPCM] as described above for H3Si–/ on silicon, we were able to systematically study the influ-
Me3SiCH2Cl and H3Si–/Me3SiCH2Br.[35] Figure 3 displays ences of the organic and halide groups of the trapping rea-
the calculated stationary points and their relative energies.
gents on product compositions and enantiomeric ratios (de-
In both cases, the ate complexes (PhCl-ate, PhBr-ate) pending on the dominating reaction mechanism, different
possess significantly lower energies than the alternative Me- stereoisomers were obtained). In all cases, high stereoselec-
isenheimer complex analogue intermediates (PhCl-Mei, tivities and product yields are only obtained if there is one
PhBr-Mei): ΔEchloride
=
27 kJmol–1, ΔEbromide
=
distinct, dominating reaction mechanism. In case of ali-
66 kJmol–1. Thus, although the energetic difference be- phatic trapping reagents (R–CH2–X; X = Cl, Br, I; R =
tween both possible mechanisms is even higher for the bro- Me3Si, Me3C, PhS, Ph), experimental and quantum chemi-
mide (this should result in a higher er value for PhBr com- cal studies support the preferred reaction of chlorides
pared to PhCl in the experiment), these calculations con- through an SN2 mechanism under retention of configura-
firm the preferred formation of the (S)-enantiomer via an tion on silicon, whereas bromides tend to react under inver-
ate complex under inversion of configuration. Nevertheless, sion on silicon through a halide–lithium exchange (ate com-
the smaller experimental er of BrPh might be explained by a plex). The corresponding iodides led to the almost complete
competing mechanism through a silyl radical, which would loss of stereoinformation due to the growing importance of
result in the formation of the opposite stereoisomer (de- silyl radicals. Thus, the stereochemical purity decreases as
scribed above, formation of radicals preferred R–I Ͼ R–Br follows: er(RCl) Ͼ er(RBr) ϾϾ er(RI). Another interesting
Ͼ R–Cl). The formation of a silyl radical could also explain observation could be made concerning the used organic
the incomplete preservation of the stereoinformation in the groups: The higher the ability of the used organyl to stabi-
1460
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 1454–1465

Reaction of Enantiomerically Pure Lithiosilanes and Electrophiles
Spectroscopic Data
lize the negative charge of the respective transition state
(pentacoordinate for the SN2 mechanism, ate complex for
the halide–lithium exchange), the higher the obtained er in
the main product. For aromatic trapping reagents (PhX, X
= Cl, Br, I), these relationships change: chlorides and bro-
mides both favor reaction through an ate complex over nu-
cleophilic aromatic substitution, leading to inversion of
configuration. Again, the formation of a silyl radical be-
comes more important for iodobenzene resulting in the al-
most complete racemization of the product. At the moment
we are trying to elucidate the influence of further functional
groups on the preferred reaction mechanism especially for
the reaction of enantiomerically pure lithiosilanes and aro-
matic halo electrophiles.
1A) rac-Methylphenyl(piperidinomethyl)[(trimethylsilyl)methyl]sil-
ane (rac-5): 1H NMR (500.1 MHz, CDCl3): δ = 0.14 [s, 9 H,
Si(CH3)3], 0.13–0.15 (AB system, not fully resolved, 2 H, SiCH2Si),
0.53 (s, 3 H, SiCH3), 1.33–1.40 (m, 2 H, NCCCH2), 1.55–1.61 (m,
4 H, NCCH2C), 2.23 (s, 2 H, SiCH2N), 2.35–2.44 (m, 4 H,
NCH2CC), 7.29–7.37 (m, 3 H, Hm and Hp), 7.67–7.71 (m, 2 H, Ho)
ppm. 13C{1H} NMR (75.5 MHz, CDCl3): δ = –1.84 (1 C,
NCSiCH3), 1.21 (1 C, SiCH2Si), 1.54 [3 C, Si(CH3)3], 24.3 (1 C,
NCCCH2), 26.7 (2 C, NCCH2C), 51.6 (1 C), (SiCH2N), 58.9 (2 C,
NCH2CC), 127.9 (2 C, Cm, C6H5Si), 129.0 (1 C, Cp, C6H5Si), 134.1
(2 C, Co, C6H5Si), 140.4 (1 C, Ci, C6H5Si) ppm. 29Si{1H} NMR
(59.6 MHz, CDCl3): δ = –6.62 (1 Si, NCSi), 0.67 [1 Si, SiCH2Si-
(CH3)3] ppm. GC/EI-MS: tR
=
8.16 min [80 °C (2 min)
–
–
10 °Cmin–1 – 280 °C (5 min)] m/z (%): 305 (7) [M+], 290 (10) [M+
CH3], 207 (6) [M+ – (H2C=NC5H10)], 98 (100) [(H2C=NC5H10)+].
C17H31NSi2 (305.61): calcd. C 66.8, H 10.2, N 4.58; found C 66.9,
H 10.2, N 5.29.
Experimental Section
1B) rac-Methylphenyl(neopentyl)(piperidinomethyl)silane (rac-6): 1H
NMR (400.1 MHz, CDCl3): δ = 0.40 (s, 3 H, SiCH3), 0.82 [s, 9 H,
C(CH3)3], 0.91–0.93 [m, 2 H, SiCH2C(CH3)3], 1.21–1.28 (m, 2 H,
NCCCH2), 1.38–1.45 (m, 4 H, NCCH2C), 2.02, 2.08 (AB system,
2JAB = 14.59 Hz, 2 H, SiCH2N), 2.13–2.24 (m, 4 H, NCH2CC),
7.19–7.27 (m, 3 H, ArH), 7.47–7.52 (m, 2 H, ArH) ppm. 13C{1H}
NMR (75.5 MHz, CDCl3): δ = –2.61 (1C, SiCH3), 19.7 [3 C,
SiCH2C(CH3)3], 23.7 (1 C, NCCCH2), 26.1 (2 C, NCCH2C), 31.1
[1 C, SiCH2C(CH3)3], 33.1 [1 C, C(CH3)3], 50.6 (1 C, SiCH2N),
58.5 (2 C, NCH2CC), 127.6 (2 C, Cm, C6H5Si), 128.6 (1 C, Cp,
C6H5Si), 133.9 (2 C, Co, C6H5Si), 127.9 (1 C, Ci, C6H5Si) ppm.
29Si{1H} NMR (59.6 MHz, CDCl3): δ = –8.39 (1 Si, NCSi) ppm.
GC/EI-MS: tR = 8.18 min [80 °C (2 min) – 10 °Cmin–1 – 280 °C
(5 min)] m/z (%): 289 (7) [M+], 218 (29) [M+ – CH2C(CH3)3], 98
(100) [(H2C=NC5H10)+]. C18H31NSi (289.54): calcd. C 74.7, H
10.8, N 4.84; found C 74.5, H 10.7, N 5.2.
General: All experiments were carried out under a dry, argon atmo-
sphere using standard Schlenk techniques. All solvents – excluding
NMR solvents – were dried with sodium and distilled prior to use.
NMR spectra were recorded on DRX-300, Avance-400, and AMX-
500 Bruker spectrometers at 22 °C. Assigning the signals was sup-
ported by additional DEPT-135, and C,H- and H,H-COSY experi-
ments. Elemental CHN analysis was performed on a Leco Elemen-
tal Analyser CHNS 932. GC/MS analysis were performed on a
ThermoQuest TRIO-1000 (EI = 70 eV); Column: Zebron, Capil-
lary GC Column, ZB-1.
Synthesis: All chlorosilanes were obtained from ABCR or Wacker
Chemie AG. Aliphatic and aromatic trapping reagents were pur-
chased from Acros Organics or Sigma–Aldrich. Elemental lithium
used for the synthesis of 2 and 21 was obtained from Chemetall.
(R)-1 was synthesized according to a literature procedure.[12d] All
spectroscopic data can be found in this reference.
1C) rac-Methylphenyl(piperidinomethyl)[(thiophenyl)methyl]silane
(rac-7): H NMR (300.1 MHz, CDCl3): δ = 0.50 (s, 3 H, SiCH3),
1
Synthesis of rac-5, rac-6, and rac-7: rac-1 was added to lithium (2
equiv.) in thf and cooled to 0 °C at the first occurrence of a color
change. After 6 h, the dark solution of lithiosilane 2 was added to
the respective chloro electrophile (2.2 equiv.) dissolved in thf
(15 mL). The solution was warmed to room temperature and all
volatiles were removed in vacuo. The residue was suspended in
NaOH (2 m, 20 mL) and extracted (5ϫ) with diethyl ether (20 mL).
The combined organic layers were extracted (5ϫ) with HCl (2 m,
15 mL) and afterwards set to pH 12 with KOH. Finally the aque-
ous layer was extracted (5ϫ) with diethyl ether (20 mL) and the
combined organic layers were dried with Na2SO4. After removal of
all volatile compounds in vacuo, the crude product was subjected
to kugelrohr distillation; pressure: 10–3 mbar; boiling point (b.p.).
Table 6 includes explicit information concerning the amounts of
involved reactants, volume of solvents, yields, and distillation tem-
peratures. The analytical data of the synthesized silanes are given
below.
1.25–1.37 (m, 2 H, NCCCH2), 1.44–1.55 (m, 4 H, NCCH2C), 2.25–
2.30 (m, 2 H, SiCH2N), 2.31–2.39 (m, 4 H, NCH2CC), 2.41–2.46
(m, 2 H, SCH2Si), 7.17–7.70 (m, 10 H, ArH) ppm. 13C{1H} NMR
(75.5 MHz, CDCl3): δ = –4.74 (1C, SiCH3), 16.4 (1 C, SCH2Si),
23.6 (1 C, NCCCH2), 26.1 (2 C, NCCH2C), 48.4 (1 C, SiCH2N),
58.5 (2 C, NCH2CC), 127.8 (2 C, Cm, C6H5Si), 128.6 (2 C, Cm,
C6H5S), 129.2 (1 C, Cp, C6H5Si), 129.6 (1 C, Cp, C6H5S), 134.0 (2
C, Co, C6H5Si), 134.6 (2 C, Co, C6H5S), 136.2 (1 C, Ci, C6H5Si),
140.1 (1 C, Ci, C6H5S) ppm. 29Si{1H} NMR (59.6 MHz, CDCl3):
δ = –7.31 (1 Si, NCSi) ppm. GC/EI-MS:tR = 9.18 min [80 °C
(2 min) – 10 °Cmin–1 – 280 °C (5 min)] m/z (%): 341 (3) [M+], 243
(1) [M+ – H2CNC5H10], 218 (35) [M+ – C6H5SCH2], 98 (100)
[(H2C=NC5H10)+].
The determination of the enantiomeric ratios was determined in
the presence of 3 equiv. of (R)-mandelic acid (for rac-5 and rac-
7) or 1.5 equiv. of (2R,3R)-di-O-benzoyltartaric acid (for rac-6) in
CDCl3 (500 μL).
1
2A) rac-5 with (R)-Mandelic Acid: H NMR (300.1 MHz, CDCl3):
Table 6. Synthesis of rac-5, rac-6, and rac-7: amounts of involved
reactants and volume of solvents as well as yields and distillation
temperatures.
δ = –0.17, –0.16 [s, 9 H each; Si(CH3)3, D1 and D2], 0.00, 0.07 (AB
system, JAB = 13.97 Hz, 4 H, SiCH2Si, D1 and D2), 0.39, 0.43 (s,
2
3 H each; SiCH3, D1 and D2), 0.91–1.13 (m, 2 H, NCCCH2, D1/
D2), 1.29–1.49 (m, 6 H, NCCH2CH2, D1/D2), 1.50–1.69 (m, 4 H,
NCCH2CH2, D1/D2), 2.09–2.35 (m, 4 H, NCH2CC, D1/D2), 2.40–
2.60 (AB system, not fully resolved, 4 H, SiCH2N), 2.95–3.15, 3.16–
3.35 (m, 2 H each; NCH2CC, D1/D2), 4.98 (s, 2 H, CHOH), 7.14–
7.48 (m, 20 H, aromat. H) ppm. The NH, OH signals were not
m (rac-1)
m (Li)
V (thf)
m (RCl)
Yield
B.p.
Product
[mg, mmol] [mg, mmol] [mL] [mg, mmol] [g, mmol, %] [°C]
rac-5
rac-6
rac-7
5.45, 13.1
4.85, 11.7
3.32, 8.00
182, 6.2
162, 3.3
111, 6
10
8
6
3.54, 8.9
2.74, 5.7
2.79, 7.6
3.41, 1.2, 86
1.31, .53, 39
0.81, .36, 28
161
155
190
Eur. J. Inorg. Chem. 2011, 1454–1465
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1461

C. Däschlein, S. O. Bauer, C. Strohmann
FULL PAPER
clearly localized. 13C{1H} NMR (75.5 MHz, CDCl3): δ = –3.89,
–3.85 (1 C each, NCSiCH3, D1 and D2), 0.12, 0.16 (1 C each,
SiCH2Si, D1 and D2), 0.81 [6 C, Si(CH3)3, D1 and D2], 20.9 (2 C,
NCCCH2, D1 and D2), 22.12, 22.15 (2 C each, NCCH2C, D1 and
D2), 49.1 (2 C, SiCH2N, D1 and D2), 55.2, 56.6 (2 C each,
NCH2CC, D1 and D2), 73.4 (2 C, CHOH, D1 and D2), 126.2 (4
C, Cm, C6H5CHOHCO2, D1 and D2), 126.9 (2 C, Cp,
C6H5CHOHCO2, D1 and D2), 127.7 (4 C, Co, C6H5CHOHCO2,
D1 and D2), 127.9 (4 C, Cm, C6H5Si, D1 and D2), 129.6 (2 C, Cp,
C6H5Si, D1 and D2), 133.3 (4 C, Co, C6H5Si, D1 and D2), 135.6
(2 C, Ci, C6H5Si, D1 and D2), 140.9 (2 C, Ci, C6H5CHOH, D1
and D2), 176.64 (2 C, COO, D1 and D2) ppm. 29Si{1H} NMR
(59.6 MHz, CDCl3): δ = –6.62, –6.60 (1 Si each, NCSi, D1 and
D2), 0.94 [2 Si, Si(CH3)3, D1 and D2] ppm.
C6H5CHOH, D1 and D2), 176.7 (2 C, COO, D1 and D2) ppm.
29Si{1H} NMR (99.4 MHz, CDCl3): δ = –7.93, (2 Si, NCSi, D1
and D2) ppm.
General Specification for the Reaction of Enantiomerically Pure 2
with (Halomethyl)trimethylsilanes, Neopentyl Halides and (Chloro-
methyl)phenyl Sulfide: (R)-1 was added to lithium (2 equiv.) in thf
and cooled to –78 °C at the first occurrence of color change. After
a reaction time of 8 h, the dark solution of lithiosilane 2 was sepa-
rated into three parts and added at –78 °C to the respective halo
electrophile (RCl, RBr, RI; 2.2 equiv.) in thf (2 mL). Afterwards,
the solution was warmed to room temperature and all volatiles
were removed in vacuo. The residue was suspended in a minimum
amount of n-pentane and separated from all salts. The mixtures of
the products were investigated with GC/MS and NMR spec-
2B) rac-6 with (2R,3R)-Di-O-benzoyltartaric Acid: 1H NMR troscopy without further purification. The following Table 7 in-
(500.1 MHz, CDCl3): δ = 0.49, 0.57 (s, 3 H each; SiCH3, D1 and cludes explicit information concerning the amounts of involved re-
D2), 0.66, 0.72 [s, 9 H each; C(CH3)3, D1 and D2], 0.76–0.89 (AB actants and volume of solvents.
system, not fully resolved, 4 H, SiCH2C, D1 and D2), 1.03–1.32
Table 7. Reaction of enantiomerically pure 2 with (halomethyl)tri-
methylsilanes, neopentyl halides and (chloromethyl)phenyl sulfide:
amounts of involved reactants and volume of solvents; n.i.: not
investigated.
(m, 6 H, NCCH2CH2, D1/D2), 1.33–1.71 (m, 6 H, NCCH2CH2,
D1/D2), 1.86–2.25 (m, 4 H, NCH2CC, D1/D2), 2.31, 2.35 (AB sys-
tem, JAB = 14.90 Hz, 2 H, SiCH2N, D1/D2), 2.58, 2.67 (AB sys-
2
tem, 2JAB = 14.90 Hz, 2 H, SiCH2N, D2/D1), 2.72–3.09, 3.10–3.54
(m, 2 H each; NCH2CC, D1/D2), 5.94 (s, 4 H, CHCO2 H), 7.03–
7.52 (m, 25 H, ArH), 8.02–8.18 (m, 5 H, ArH) ppm. The NH and
CO2H signals were not clearly localized. 13C{1H} NMR
(125.8 MHz, CDCl3): δ = –3.67, –3.66 (1 C each, NCSiCH3, D1
and D2), 19.79, 19.84 [3 C each, C(CH3)3, D1 and D2], 21.44, 21.46
(1 C each, NCCCH2, D1 and D2), 22.46, 22.50 (2 C each,
NCCH2C, D1 and D2), 29.87, 29.96 (1 C each, SiCH2C, D1 and
D2), 30.98, 31.04 [1 C each, C(CH3)3, D1 and D2], 47.7 (2 C,
SiCH2N, D1 and D2), 55.8, 58.0 (2 C each, NCH2CC, D1 and D2),
73.8 (4 C, CHCO2 H, D1 and D2), 127.9 (4 C, Cm, C6H5Si, D1
and D2), 128.1 (4 C, Cp, C6H5CO2C, D1 and D2), 129.51, 129.54
(1C each, Cp, C6H5Si, D1 and D2), 130.1 (8 C, Co, C6H5CO2C, D1
and D2), 130.7 (2 C, Ci, C6H5Si, D1 and D2), 132.3 (8 C, Cm,
C6H5CO2C, D1 and D2), 133.72, 133.75 (2 C each, Co, C6H5Si, D1
and D2), 136.46, 136.61 (2 C each, Ci, C6H5CO2C, D1 and D2),
165.7 (4 C, C6H5CO2C, D1 and D2), 171.4 (4 C, CO2 H, D1 and
D2) ppm. 29Si{1H} NMR (59.6 MHz, CDCl3): δ = –8.63, –8.61 (1
Si each, NCSi, D1 and D2) ppm.
m (R)-1
[mg,
mmol]
m (Li)
[mg,
mmol]
V
(thf)
[mL]
m (RCl)
[mg,
mmol]
m (RBr)
[mg,
mmol]
m (RI)
[mg,
mmol]
Me3SiCH2 1.00, 2.41 33.5, 4.82
Me3CCH2 1.07, 2.57 35.7, 5.14
4
4
2
217, .77
201, .89
304, .89
295, .77
285, .89
n.i.
378, .77
373, .89
n.i.
PhSCH2
0.36, 0.87 12.1, 1.74
The analytical data are in agreement with those given above.
Synthesis of rac-20 and (R)-20
The synthesis of rac-20 and (R)-20 used for the subsequent, selec-
tive cleavage of the Si–C bond has been carried out according to
the literature.[12e] All spectroscopic data can be found in this refer-
1
ence. Yet, as the H NMR spectroscopic data of 20 are of crucial
necessity for the determination of the enantiomeric ratio in the re-
action of enantiomerically pure 21 with halobenzenes, these data
are listed below.
1
rac-21: H NMR (300.1 MHz, C6D6): δ = 0.28 [s, 9 H, Si(CH3)3],
1
2C) rac-7 with (R)-Mandelic Acid: H NMR (500.1 MHz, CDCl3):
0.54 (s, 3 H, SiCH3), 1.30–1.40 (m, 2 H, NCCCH2), 1.50–1.62 (m,
4 H, NCCH2C), 2.35–2.45 (m, 4 H, NCH2CC), 2.29, 2.51 (AB
δ = 0.49, 0.50 (s, 3 H each; SiCH3, D1 and D2), 1.02–1.13 (m, 2
H, NCCCH2, D1/D2), 1.38–1.53 (m, 6 H, NCCH2CH2, D1/D2),
1.54–1.66 (m, 4 H, NCCH2C, D1/D2), 2.24–2.45 (m, 4 H,
NCH2CC, D1/D2), 2.25–2.52 (m, 4 H, SCH2Si, D1 and D2), 2.70,
2.73 (AB system,2JAB = 3.24 Hz, 2 H, SiCH2N, D1 and D2), 2.79,
2.82 (AB system,2JAB = 3.71 Hz, 2 H, SiCH2N, D2 and D1), 3.16–
3.25, 3.27–3.35 (m, 2 H each; NCH2CC, D1/D2), 5.03 (s, 2 H,
CHOH, D1 and D2), 7.20–7.53 (m, 30 H, ArH) ppm. The NH,
OH signals were not clearly localized. 13C{1H} NMR (125.8 MHz,
CDCl3): δ = –5.22, –5.18 (1 C each, NCSiCH3, D1 and D2), 15.1,
15.7 (1 C each, SiCH2S, D1 and D2), 21.18, 21.22 (1 C each,
NCCCH2, D1 and D2), 22.5 (4 C, NCCH2C, D1 and D2), 46.7,
46.8 (1 C each, SiCH2N, D1 and D2), 56.4 (1 C), 56.6 (2 C), 56.8
(1 C, NCH2CC, D1 and D2), 73.2 (2 C, CHOH, D1 and D2),
2
system, JAB = 14.4 Hz, 2 H, SiCH2N), 7.25–7.40 (m, 3 H, ArH),
7.45–7.55 (m, 2 H, ArH) ppm.
rac-21 with Three Equivalents of (R)-Mandelic Acid: 1H NMR
(300.1 MHz, C6D6): δ = 0.20 [s, 9 H, Si(CH3)3, D2], 0.21 [s, 9 H,
Si(CH3)3, D1], 0.47 (s, 3 H, SiCH3, D1), 0.53 (s, 3 H, SiCH3, D2),
0.90–1.15 (m, 2 H, NCCCH2, D1/D2), 1.30–1.80 (m, 10 H,
NCCH2CH2, D1 and D2), 2.10–2.50 (m, 4 H, NCH2CC, D1 and
D2), 2.40, 3.02 (AB system, 2JAB = 14.6 Hz, 4 H, SiCH2N, D1 and
D2), 3.05–3.60 (m, 4 H, NCH2CC, D1 and D2), 5.48 (s, 2 H,
CHOH, D1 and D2), 7.20–7.50 (m, 20 H, ArH) ppm. The OH and
NH signals were not clearly localized.
Reaction of Enantiomerically Pure 21 with Halobenzenes: (R)-20
126.6 (4 C, Cm, C6H5CHOHCO2, D1 and D2), 127.8 (2 C, Cp, (0.25 g, 0.86 mmol) was added to a suspension of lithium (11.9 mg,
C6H5CHOHCO2, D1 and D2), 128.2 (4 C, Co, C6H5CHOHCO2, 1.72 mmol) in thf (2 mL) and cooled to –50 °C at the first occur-
D1 and D2), 128.35 (4 C, Cm, C6H5Si, D1 and D2), 128.44 (4 C,
Cm, C6H5S, D1 and D2), 128.84, 128.85 (1 C each, Cp, C6H5Si, D1
and D2), 130.35, 130.37 (2 C each, Co, C6H5Si, D1 and D2),
132.96, 132.97 (1 C each, Ci, C6H5Si, D1 and D2), 133.9 (2 C, Cp,
C6H5S, D1 and D2), 134.4 (4 C, Co, C6H5S, D1 and D2), 138.27,
rence of color change. After 6.5 h, the dark solution of lithiosilane
21 was separated into three parts and added at –78 °C to i) chloro-
benzene (70.8 mg, 0.63 mmol), ii) bromobenzene (98.7 mg,
0.63 mmol) and iii) iodobenzene (128 mg,0.63 mmol) in of thf
(2 mL). The solution was warmed to room temperature and all
138.28 (1 C each, Ci, C6H5S, D1 and D2), 139.6 (2 C, Ci, volatiles were removed in vacuo. The residue was suspended in a
1462 Eur. J. Inorg. Chem. 2011, 1454–1465
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Reaction of Enantiomerically Pure Lithiosilanes and Electrophiles
119, 8429–8432; Angew. Chem. Int. Ed. 2007, 46, 8281–8284;
b) S. T. Kerrick, P. Beak, J. Am. Chem. Soc. 1991, 113, 9709–
9710; c) D. Hoppe, F. Hintze, P. Tebben, Angew. Chem. 1990,
102, 1457–1459; Angew. Chem. Int. Ed. Engl. 1990, 29, 1422–
1424; d) A. Hameau, F. Guyon, M. Knorr, C. Däschlein, C.
Strohmann, N. Avarvari, Dalton Trans. 2008, 4866; e) M. J.
McGrath, J. L. Bilke, P. O’Brien, Chem. Commun. 2006, 2607–
2609; f) J.-C. Kizirian, Chem. Rev. 2008, 108, 140–205; g) A.
Basu, S. Thayumanavan, Angew. Chem. 2002, 114, 740–763;
Angew. Chem. Int. Ed. 2002, 41, 716–738.
There are three general methods for the stabilization of a lithi-
ated carbon center: i) introduction of a strongly electronegative
substituent, ii) introduction of a coordinating side arm, and
iii) introduction of an organosulfur substituent in the α-posi-
tion to the negative charge; for a current example dealing with
these three effects, see: H. Ott, C. Däschlein, D. Leusser, D.
Schildbach, T. Seibel, D. Stalke, C. Strohmann, J. Am. Chem.
Soc. 2008, 130, 11901–11911.
minimum amount of n-pentane and separated from all salts. The
obtained product mixtures were investigated with GC/MS and
NMR spectroscopy without further purification. The analytical
data comply with those given above.
Determination of the Product Composition: The product composi-
tions of all described reactions have been determined through a
combination of GC/MS and NMR spectroscopic analysis. The GC/
MS measurements were used to determine the various products,
whereas product ratios were determined by the integration of rel-
[3]
1
evant groups in the H NMR spectra.
Computational Details: All calculations were done without sym-
metry restrictions. Starting coordinates were obtained with Chem3-
DUltra 10.0. Optimization and additional harmonic vibrational
frequency analyses (to establish the nature of stationary points on
the potential energy surface) were performed with the software
package Gaussian 03 (rev. E.01) on the B3LYP/6-31+G(d) level.[35]
The stationary points were energy optimized in thf solution (ε =
7.58 for thf) by means of the self-consistent reaction field (SCRF)
method. The energies for the structures obtained were calculated
with the polarizable conductor calculation model (CPCM). The
calculated stationary points exhibited exactly one imaginary fre-
quency in the direction of the expected reaction coordinate in each
case. The total (SCF) and zero-point energies (ZPE) of all systems
can be found in Table 8. All coordinates of the investigated systems
are available in the Supporting Information.
[4]
[5]
For some examples concerning the introduction of a strongly
electronegative substituent for the stabilization of a lithiated
carbon center in reactions with halo electrophiles, see: a) W. C.
Still, C. Sreekumar, J. Am. Chem. Soc. 1980, 102, 1201–1202;
b) H. A. Bent, Chem. Rev. 1961, 61, 275–311.
For some examples concerning the introduction of a coordinat-
ing side arm for the stabilization of a lithiated carbon center
in reactions with halo electrophiles, see: a) G. W. Klumpp, Recl.
Trav. Chim. Pays-Bas 1986, 105, 1–21; b) T. H. Chan, P. J. Pel-
lon, J. Am. Chem. Soc. 1989, 111, 8737–8738; c) T. H. Chan,
S. Lamonthe, Tetrahedron Lett. 1991, 32, 1847–1850; d) C.
Strohmann, K. Lehmen, K. Wild, D. Schildbach, Organometal-
lics 2002, 21, 3079–3081; e) C. Strohmann, B. C. Abele, K. Leh-
men, F. Villafane, L. Sierra, S. Martín-Barrios, D. Schildbach,
J. Organomet. Chem. 2002, 661, 149–158; f) C. Strohmann,
B. C. Abele, K. Lehmen, D. Schildbach, Angew. Chem. 2005,
117, 3196-3199; Angew. Chem. Int. Ed. 2005, 44, 3136–3139; g)
C. Däschlein, V. H. Gessner, C. Strohmann, Chem. Eur. J. 2010,
16, 4048–4062.
For some examples concerning the introduction of an organo-
sulfur substituent to the α-position to the negative charge for
the stabilization of a lithiated carbon center in reactions with
halo electrophiles, see: a) R. K. Dress, T. Rölle, R. W.
Hoffmann, Chem. Ber. 1995, 128, 673–677; b) A. E. Reed,
P. v. R. Schleyer, J. Am. Chem. Soc. 1990, 112, 1434–1445; c) U.
Salzner, P. v. R. Schleyer, J. Am. Chem. Soc. 1993, 115, 10231–
10236; d) J. A. Dobado, H. Martínez-Garzía, J. M. Molina,
M. R. Sundberg, J. Am. Chem. Soc. 1998, 120, 8461–8471; e)
T. Stefan, R. Janoschek, J. Mol. Model. 2000, 6, 282–288.
W. Kutzelnigg, Angew. Chem. 1984, 96, 262–286; Angew. Chem.
Int. Ed. Engl. 1984, 23, 272–295.
For studies on silyl anions, see: a) J. B. Lambert, M. Urdaneta-
Perez, H.-N. Sun, J. Chem. Soc., Chem. Commun. 1976, 806–
807; b) J. B. Lambert, M. Urdaneta-Perez, J. Am. Chem. Soc.
1978, 100, 157–162; c) M. Flock, C. Marschner, Chem. Eur. J.
2002, 8, 1024–1030; d) M. Flock, C. Marschner, Chem. Eur. J.
2005, 11, 4635–4642.
For some selected pioneer works on functionalized lithiosil-
anes, see: a) L. H. Sommer, J. E. Lyons, H. Fujimoto, J. Am.
Chem. Soc. 1969, 91, 7051–7061; b) E. Colomer, R. J. P. Corriu,
J. Chem. Soc., Chem. Commun. 1976, 176–177.
For selected works on lithiosilanes, see: a) C. Strohmann, M.
Bindl, V. C. Vraaß, J. Hörnig, Angew. Chem. 2004, 116, 1029–
1032; Angew. Chem. Int. Ed. 2004, 43, 1011–1014; b) D. Auer,
M. Kaupp, C. Strohmann, Organometallics 2004, 23, 3647–
3655; c) M. Nanjo, M. Masayuki, Y. Ushida, Y. Awamura, K.
Mochida, Tetrahedron Lett. 2005, 46, 8945–8947; d) D. Bravo-
Zhivotovskii, I. Ruderfer, S. Melamed, M. Botoshansky, B. Tu-
manskii, Y. Apeloig, Angew. Chem. 2005, 117, 749–753; Angew.
Chem. Int. Ed. 2005, 44, 739–743; e) T. I. Kückmann, F.
Dornhaus, M. Bolte, H.-W. Lerner, M. C. Holthausen, M.
Wagner, Eur. J. Inorg. Chem. 2007, 1989–2003; f) D. Scheschke-
Table 8. Total (SCF) and zero-point energies (ZPE) of the calcu-
lated model systems.
Optimized system
SCF [Hartree]
ZPE [Hartree]
SN2-Cl
ate-Cl
SN2-Br
ate-Br
TS-PhCl-ate
TS-PhCl-Mei
TS-PhBr-ate
TS-PhBr-Mei
–1082.168985
–1082.155393
–3193.696841
–3193.697703
–983.197581
–983.187544
–3094.745311
–3094.719164
–1082.095785
–1082.083454
–3193.623446
–3193.625500
–983.087307
–983.077039
–3094.634502
–3094.609352
[6]
Supporting Information (see footnote on the first page of this arti-
cle): Extracts of relevant NMR spectra, computational data.
Acknowledgments
[7]
[8]
We are grateful to the Deutsche Forschungsgemeinschaft (DFG)
for financial support. Furthermore we acknowledge Wacker
Chemie AG and Chemetall GmbH for providing chemicals. C. D.
thanks the Studienstiftung des deutschen Volkes for a doctoral
scholarship. S. O. B. thanks the Fonds der chemischen Industrie for
an undergraduate scholarship.
[9]
[1] For some general articles and reviews concerning organo-
lithium compounds, see: a) D. Hoppe, T. Hense, Angew. Chem.
1997, 109, 2376; Angew. Chem. Int. Ed. Engl. 1997, 36, 2282–
2316; b) P. Beak, A. Basu, D. J. Gallagher, Y. S. Park, S. Thayu-
manavan, Acc. Chem. Res. 1996, 29, 552–556; c) M. C. Whisler,
S. MacNeil, V. Snieckus, P. Beak, Angew. Chem. 2004, 116,
2256; Angew. Chem. Int. Ed. 2004, 43, 2206–2225; d) T. Stey,
D. Stalke, in: The Chemistry of Organolithium Compounds
(Eds.: Z. Rappoport, I. Marek), Wiley, Chichester, U.K., 2004,
vol. 1, pp. 47–120; e) V. H. Gessner, C. Däschlein, C.
Strohmann, Chem. Eur. J. 2009, 15, 3320–3334.
[10]
[2] For some of the countless articles including such trapping reac-
tions, see: a) C. Strohmann, V. H. Gessner, Angew. Chem. 2007,
Eur. J. Inorg. Chem. 2011, 1454–1465
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1463

C. Däschlein, S. O. Bauer, C. Strohmann
FULL PAPER
witz, Angew. Chem. 2004, 116, 3025–3028; Angew. Chem. Int.
Ed. 2004, 43, 2965–2967; g) H.-W. Lerner, Coord. Chem. Rev.
2005, 249, 781–798; h) H.-W. Lerner, S. Scholz, M. Bolte, M.
Wagner, Z. Anorg. Allg. Chem. 2004, 630, 443–451.
For some of our studies on lithiosilanes, see: a) C. Strohmann,
C. Däschlein, D. Auer, J. Am. Chem. Soc. 2006, 128, 704–705;
b) C. Strohmann, C. Däschlein, Chem. Commun. 2008, 2791–
2793; c) C. Däschlein, V. H. Gessner, C. Strohmann, Acta Crys-
menclature. This change of the priority of the CIP nomencla-
ture has to be kept in mind in all of these transformations.
G. T. Notte, J. L. Leighton, J. Am. Chem. Soc. 2008, 130, 6676–
6677.
S. Shirakawa, P. J. Lombardi, J. L. Leighton, J. Am. Chem. Soc.
2005, 127, 9974–9975.
[21]
[22]
[23]
[24]
[11]
[12]
S. Shirakawa, R. Berger, J. L. Leighton, J. Am. Chem. Soc.
2005, 127, 2858–2859.
tallogr., Sect.
Strohmann, Eur. J. Inorg. Chem. 2009, 43–52.
E
2008, 64, o1950; d) C. Däschlein, C.
For some other examples concerning the application of Si-chi-
ral silanes, see: a) B. H. Lipshutz, N. Tanaka, B. R. Taft, C.-T.
Lee, Org. Lett. 2006, 8, 1963–1966; b) B. Zhu, J. S. Panek, Eur.
J. Org. Chem. 2001, 1701–1714; c) S. Rendler, G. Auer, M. Oes-
treich, Angew. Chem. 2005, 117, 7793–7797; Angew. Chem. Int.
Ed. 2005, 44, 7620–7624; d) S. E. Denmark, G. L. Beutner, J.
Am. Chem. Soc. 2003, 125, 7800–7801; e) P. J. Stang, A. E. Le-
arned, J. Org. Chem. 1989, 54, 1779–1981.
The low temperature (–78 °C) in the synthesis of lithiosilane
(S)-2 is necessary to prevent the racemization process, which
occurs at temperatures above –30 °C; for a detailed description,
see: D. Auer, J. Hörnig, C. Strohmann, in: Organosilicon Chem-
istry V: From Molecules to Materials (Eds.: N. Auner, J. Weis),
Wiley-VCH, Weinheim, Germany, 2003, pp. 167.
To date, six enantiomerically enriched or enantiomerically pure
lithiosilanes can be found in the literature: a) L. H. Sommer, R.
Mason, J. Am. Chem. Soc. 1965, 87, 1619–1620; b) E. Colomer,
R. J. P. Corriu, J. Organomet. Chem. 1977, 133, 159–168; c) M.
Omote, T. Tokita, Y. Shimizu, I. Imae, E. Shirakawa, Y. Ka-
wakami, J. Organomet. Chem. 2000, 611, 20–25; d) C.
Strohmann, J. Hörnig, D. Auer, Chem. Commun. 2002, 766–
767; e) C. Strohmann, C. Däschlein, M. Kellert, D. Auer, An-
gew. Chem. 2007, 119, 4864–4866; Angew. Chem. Int. Ed. 2007,
46, 4780–4782; f) M. Oestreich, G. Auer, M. Keller, Eur. J. Org.
Chem. 2005, 184–195.
Recently, we have reported new details about the reactivity of
disilanes and elemental lithium for the formation of lithioisil-
anes; this manuscript also includes the synthesis of the seventh
enantiomerically pure lithiosilane, although not isolated in
pure form; for details see: C. Däschlein, C. Strohmann, Dalton
Trans. 2010, 39, 2062–2069.
[25]
[13]
[26]
[27]
The reaction of 2 and Me3SiCl is known to proceed selectively
under full preservation of the stereoinformation (retention)
yielding the corresponding disilane. All relevant NMR spectro-
scopic data can be found in ref.[12e]
.
In all cases, the product composition has been determined by
1
GC/MS analysis whereas H NMR spectroscopic studies have
[14]
[15]
For another recent example, see: C. Strohmann, C. Däschlein,
Organometallics 2008, 27, 2499–2504.
been used to determine the ratios between identified products.
In case of the racemic disilane rac-1, the reaction time can be
decreased due to the higher reaction temperature.
[28]
[29]
For some examples, see: a) A. Sekiguchi, M. Nanjo, C. Cabuto,
H. Sakurai, Organometallics 1995, 14, 2630–2632; b) I. Flem-
ing, R. S. Robert, S. C. Smith, J. Chem. Soc. Perkin Trans. 1
1998, 1209–1214.
For selected examples concerning an SN2 mechanism in the
reaction with other nucleophiles: Y. Fang, Y. Gao, P. Ryberg,
J. Eriksson, M. Kolodziejska-Huben, A. Dybala-Defratyka, S.
Madhavan, R. Danielsson, P. Paneth, O. Matsson, K. C. Wes-
taway, Chem. Eur. J. 2003, 9, 2696–2709, and references cited
therein.
Quantum chemical studies of Boche et al. on methyl ate-com-
plexes suggested that ate complexes are pivotal intermediates
in halogen–lithium exchange reactions with alkyllithium com-
pounds: G. Boche, M. Schimeczek, J. Cioslowski, P. Piskorz,
Eur. J. Org. Chem. 1998, 1851–1860.
For selected examples concerning discussions of the mecha-
nism of halide–lithium exchange reactions, see: a) R. W.
Hoffmann, M. Brönstrup, M. Müller, Org. Lett. 2003, 5, 313–
316; b) M. Müller, M. Brönstrup, O. Knopff, V. Schulze, R. W.
Hoffmann, Organometallics 2003, 22, 2931–2937; c) V. P. W.
Böhm, V. Schulze, M. Brönstrup, M. Müller, R. W. Hoffmann,
Organometallics 2003, 22, 2925–2930; d) K. B. Wiberg, S.
Sklenak, W. F. Bailey, J. Org. Chem. 2000, 65, 2014–2021; e) P.
Beak, D. J. Allen, J. Am. Chem. Soc. 1992, 114, 3420–3425; f)
W. F. Bailey, J. J. Patricia, J. Organomet. Chem. 1988, 352, 1–
46; g) H. Gilman, G. D. Lichtenwalter, J. Am. Chem. Soc. 1958,
80, 608–611; h) K.-W. Lee, J. S. Filippo Jr, Organometallics
1983, 2, 906–908; i) J. S. Filippo Jr, J. Silbermann, J. Am. Chem.
Soc. 1982, 104, 2831–2836, and references cited therein.
The absolute configuration of lithiosilane (S)-2 is not experi-
mentally confirmed. Yet, although no enantiomerically pure
lithiosilane could be isolated as single crystal for the determi-
nation of the absolute configuration until today, it is in general
supposed, that the Si–Si bond cleavage proceeds under reten-
tion of configuration, e.g., see: K. Tamao, A. Kawachi, Adv.
Organomet. Chem. 1995, 38, 1–58.
Treatment of the corresponding racemic silane with (R)-man-
delic or (2R,3R)-di-O-benzoyltartaric acid results in the forma-
tion of diastereomeric salts, which can be differentiated in the
1H NMR spectroscopic studies. In all cases, the signal of at
least one group of these diastereomers (e.g., one of the methyl
groups) was separated fully in the spectrum and therefore inde-
pendent integration of the different diastereomers was possible.
These diastereomeric ratios comply with enantiomeric ratios.
The formed compounds have been identified by GC/MS spec-
troscopy, their ratio has been determined by NMR spec-
troscopy.
For some selected examples, see: a) C. Strohmann, S. Lüdtke,
E. Wack, Chem. Ber. 1996, 129, 799–805; b) C. Strohmann, S.
Lüdtke, O. Ulbrich, Organometallics 2000, 19, 4223–4227.
There is one slight exception to this trend: as displayed in
Table 4, Me3SiCH2Cl has a slightly higher er value than
PhCH2Cl.
Currently, we are performing quantum chemical calculations
on the reactivity of lithiumorganyls and cloro-/bromosilanes;
all results show that the negatively charged carbanionic center
of the carbon attacks selectively under inversion of configura-
tion at silicon; publication in preparation.
For some recent theoretical studies on hypervalent (pentacoor-
dinate)silicon atoms in substitution reactions, see: a)
S. C. A. H. Pierrefixe, C. F. Guerra, F. M. Bickelhaupt, Chem.
Eur. J. 2008, 14, 819–828; b) M. A. van Bochove, F. M. Bickel-
haupt, Eur. J. Org. Chem. 2008, 649–654; c) A. P. Bento, F. M.
Bickelhaupt, J. Org. Chem. 2007, 72, 2201–2207.
[16]
[30]
[31]
[32]
[33]
[17]
[18]
[34]
[35]
[19]
M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, J. A. Montgomery Jr, T. Vreven,
K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tom-
asi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega,
G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota,
R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratch-
ian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gom-
perts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C.
Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A.
Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dap-
[20]
Important remark: Although the synthesis of the lithiosilane
starting with disilane (R)-1 is supposed to proceed under reten-
tion of the stereoinformation, lithiosilane (S)-2 has the (S)-con-
figuration due to the change of the priority of the CIP no-
1464
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 1454–1465

Reaction of Enantiomerically Pure Lithiosilanes and Electrophiles
prich, A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, [40]
A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q.
Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov,
G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin,
Contrary to some textbook opinions, silyl radicals should –
like “silyl anions” – have a stable configuration at least at low
temperatures. These low temperatures were present in the per-
formed reactions.
D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayak-
kara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen,
M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian 03, rev. E.01,
Gaussian, Inc., Wallingford CT, 2004.
[41]
The higher reaction temperature (–50 °C) in the synthesis of
lithiosilane 21 (in comparison with lithiosilane 2) is necessary
due to the decreased reactivity of the Si–C bond against ele-
mental lithium in (R)-20.
The reaction of 21 and Me3SiSiMe2Cl is known to proceed
selectively under full preservation of the stereoinformation (re-
tention) yielding the corresponding tetrasilane, see Scheme 8.
[36] P. Flükinger, H. P. Lüthi, S. Portmann, J. Weber, Molekel 4.3,
Swiss Center for Scientific Computing, Manno, Switzerland,
2000–2002.
[37] J. Hörnig, Dissertation, University of Würzburg, 2004.
[38] For examples, see: a) T. Linker, M. Schmittel, Radikale und
Radikalionen in der Organischen Synthese, Wiley-VCH,
Weinheim, Germany, 1998, pp. 129–133; b) P. Dowd, W.
Zhang, Chem. Rev. 1993, 93, 2091–2115; c) A. Effio, D. Griller,
K. U. Ingold, A. L. J. Beckwith, A. K. Serelis, J. Am. Chem.
Soc. 1980, 102, 1734–1736.
[42]
[43] Under these conditions, the resonances of the methyl groups
of the SiMe moiety could be integrated separately in the 1H
NMR spectra.
[44] The absolute configuration of 20 can be determined through
the corresponding hydrochloride, see ref.[12e]
Received: August 3, 2010
Published Online: February 18, 2011
[39] P. J. Krusic, P. J. Fagan, J. S. Filippo Jr, J. Am. Chem. Soc. 1977,
99, 250–252.
Eur. J. Inorg. Chem. 2011, 1454–1465
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1465