Formation of organosilicon compounds 117: C-brominated 1,3,5-trisilacyclohexanes and their reactions with BuLi

Article abstract of DOI:10.1139/v00-035

The photobromination of 1,1,3,3,5,5-hexamethyl-1,3,5-trisilacyclohexane (1) almost exclusively attacks CH₂ groups and results in 2,2-dibromo-trisilacyclohexane (2) as well as 2,2,4,4-tetrabromo-trisilacyclohexane (3) in high yields. Starting from a mixture of C-brominated trisilacyclohexanes the isomeric 2,2,9-tribromo-1,3,3,5,5,8,8,10,10,13,13-undecamethyl-1,3,5,8,10,13- hexasilabicyclo[7.2.2]tridec-6-yne (6) had been obtained in very low yield in an attempt to establish a preparative route to adamantanes with a C₄Si₆ skeleton, i.e., with C bridgeheads and SiR₂ bridges. By ICl-cleavage of a Si - methyl bond in 2 and subsequent substitution with Br₃CLi, the trisilacyclohexane 4 with functional groups in opposite positions of the ring can be obtained. Linking the step-by-step synthesized Cl-Me₂Si-C≡C-SiMe₂-CH₂-SiMe ₂-Ph to the CBr₃ group of 4 results after HBr-cleavage of the Si - Ph bond in (ω-bromo-octynyl)-trisilacyclohexane (12). A ring closure of 12 would result in an isomeric hexasilabicyclo[7.2.2]tridec-6-yne. The compounds were characterized by ¹H, ¹³C, and ²⁹Si NMR spectra. Additionally, the molecular structures of 4 and 6 were confirmed by X-ray single crystal investigations.

Full text of DOI:10.1139/v00-035

1388
Formation of organosilicon compounds 117:1
C-brominated 1,3,5-trisilacyclohexanes and their
reactions with BuLi
G. Fritz, M. Keuthen, F. Kirschner, E. Matern, H. Goesmann, K. Peters,
E.-M. Peters, and H.G. von Schnering
Abstract: The photobromination of 1,1,3,3,5,5-hexamethyl-1,3,5-trisilacyclohexane (1) almost exclusively attacks
CH2 groups and results in 2,2-dibromo-trisilacyclohexane (2) as well as 2,2,4,4-tetrabromo-trisilacyclohexane
(3) in high yields. Starting from a mixture of C-brominated trisilacyclohexanes the isomeric 2,2,9-tribromo-
1,3,3,5,5,8,8,10,10,13,13-undecamethyl-1,3,5,8,10,13-hexasilabicyclo[7.2.2]tridec-6-yne (6) had been obtained in very
low yield in an attempt to establish a preparative route to adamantanes with a C4Si6 skeleton, i.e., with C bridgeheads
and SiR2 bridges. By ICl-cleavage of a Si—methyl bond in 2 and subsequent substitution with Br3CLi, the
trisilacyclohexane 4 with functional groups in opposite positions of the ring can be obtained. Linking the step-by-step
synthesized Cl-Me2Si-CϵC-SiMe2-CH2-SiMe2-Ph to the CBr3 group of 4 results after HBr-cleavage of the Si—Ph bond
in (ω-bromo-octynyl)-trisilacyclohexane (12). A ring closure of 12 would result in an isomeric hexasila-
1
bicyclo[7.2.2]tridec-6-yne. The compounds were characterized by H, 13C, and 29Si NMR spectra. Additionally, the
molecular structures of 4 and 6 were confirmed by X-ray single crystal investigations.
Key words: 1,1,3,3,5,5-hexamethyl-1,3,5-trisilacyclohexane, bromination, 2,2,9-tribromo-1,3,3,5,5,8,8,10,10,13,13-
undecamethyl-1,3,5,8,10,13-hexasilabicyclo[7.2.2]tridec-6-yne, carbosilane synthesis, NMR data, crystal structure
investigation.
Résumé : La photobromation du 1,1,3,3,5,5-hexaméthyl-1,3,5-trisilacyclohexane (1) se fait presque exclusivement par
une attaque des groupes CH2 et elle conduit à la formation de 2,2-dibromotrisilacyclohexane (2) ainsi que de 2,2,4,4-
tétrabromo-trisilacyclohexane (3) avec des rendements élevés. Dans un essai de démontrer la possibilité de préparer des
adamantanes comportant un squelette C4Si6, c’est-à-dire les atomes de ponts C et les ponts SiR2, on a déterminé que,
en partant d’un mélange de trisilacyclohexanes C-bromé, on obtient le 2,2,9-tribromo-1,3,3,5,5,8,8,10,10,13,13-
undécaméthyl-1,3,5,8,10,13-hexasilabicyclo[7.2.2]tridéc-6-yne (6) avec un faible rendement. Un clivage d’une liaison
C—méthyle du composé 2 à l’aide de ICl, suivie d’une substitution subséquente avec du Br3CLi permet d’obtenir le
trisilacyclohexane 4 portant des groupes fonctionnels dans des positions opposées du cycle. La liaison du groupe CBr3
du produit 4 au Cl-Me2Si-CϵC-SiMe2-CH2-Ph synthétisé étape par étape conduit au (ω-bromo-octynyl)-trisilacyclo-
hexane (12) après clivage de la liaison Si—Ph avec du HBr. Une fermeture du cycle du produit 12 devrait conduire à
1
l’hexasilabicyclo[7.2.2]tridéc-6-yne. Les composés ont été caractérisés par spectroscopie RMN du H, du 13C et du 29Si.
De plus, on a confirmé les structures des composés 4 et 6 par diffraction des rayons X par des cristaux uniques.
Mots clés : 1,1,3,3,5,5-hexaméthyl-1,3,5-trisilacyclohexane, bromation, 2,2,9-tribromo-1,3,3,5,5,8,8,10,10,13,13-
undécaméthyl-1,3,5,8,10,13-hexasilabicyclo[7.2.2]tridéc-6-ène, synthèse de carbosilanes, données de RMN, étude de
structures cristallines.
[Traduit par la Rédaction] Fritz et al. 1395
1. Introduction
only those with a Si4C6 skeleton were observed, i.e., Si oc-
cupies the bridgehead positions and CH2 groups form the
bridges in these adamantanes (1). Nevertheless, C bridge-
heads exist in the barrelane Si4C11H28 (2), in the asterane
The gas phase pyrolysis of methylsilanes, among many
other carbosilanes, yields 1,3,5,7-tetrasilaadamantanes, but
Received July 2, 1999. Published on the NRC Research Press website on October 20, 2000.
Dedicated to Professor Adrian G. Brook on the occasion of his 75th birthday.
G. Fritz,2 M. Keuthen, F. Kirschner, E. Matern, and H. Goesmann. Institut für Anorganische Chemie der Universität (TH),
Karlsruhe, Germany.
K. Peters, E.-M. Peters, and H.G. von Schnering. Max-Planck-Institut für Festkörperforschung, Stuttgart, Germany.
1Part 116: H. Wendt, I. Odenwald, P. Scheer, E. Matern, G. Fritz, Z. Anorg. Allg. Chem. 623, 853 (1997).
2Author to whom correspondence may be addressed. Prof. Dr. Dr.h.c. G. Fritz, Institut für Anorganische Chemie der Universität
(TH), Engesserstr. Geb. Nr. 30.45, D-76128 Karlsruhe, Germany. E-mail: gerhard.fritz@chemie.uni-karlsruhe.de
Can. J. Chem. 78: 1388–1395 (2000)
© 2000 NRC Canada

Fritz et al.
1389
Scheme 1. Photobromination products of (Me2Si-CH2)3 (1).
Scheme 2. Organometallic preparation of 4.
Fig. 1. Structure of 2,2,9-tribromo-1,3,3,5,5,8,8,10,10,13,13-
undecamethyl-1,3,5,8,10,13-hexasilabicyclo[7.2.2]tridec-6-yne
Si6C18H39Br3 (6) with the atomic labelling (hydrogen atoms
omitted). For bond lengths and angles see Tables 4 and 5, re-
spectively.
According to the sum formula it might consist of two triply
C-C linked trisilacyclohexanes in a wurtzite-like structure.
The single crystal structure investigation of this product con-
firmed the heavy-atom formula Si6C18Br3, while revealing a
completely different and unexpected molecular structure
(molecule 6 in Fig. 1). In this paper we first present a conve-
nient preparative access to C-brominated derivatives of
(Me2Si-CH2)3 1, then we deal with the reactions of such de-
rivatives with metal alkyls, and eventually we describe a re-
action path for the synthesis of 6 as far as these steps have
already been realized.
Si6Cl8C6H8 (3), in the di-adamantane Si7C16H36 (4) and in
the octamethyl-octasilaheptacyclooctadecane Si8C18H40 (5),
e.g., obtained either by pyrolysis or by organometallic syn-
thesis. The synthetic approach to adamantanes with the “in-
verse” skeleton C4Si6 has to start from C-functional
carbosilanes which at the same time carry less or non-
reactive groups at Si. Unfortunately, these cannot simply be
obtained by alkylating the readily available perchlorinated
carbosilanes, because MeMgCl, for example, does not sub-
stitute the Si atoms in (Cl2Si-CCl2)3. Instead, the reaction
starts at a ring C atom and proceeds via ring contraction and
ring cleavage to eventually lead to the linear Me3Si-CH2-
SiMe2-CϵC-SiMe3 (6).
In addition, the photochlorination of Si-methylated carbo-
silanes does not provide a suitable pathway to the desired
starting material, because photochlorination attacks both CH3
and CH2 groups at the same time. In contrast, the photo-
bromination of hexamethyl-1,3,5-trisilacyclohexane 1 can be
guided in a way that nearly exclusively attacks CH2 groups
(7) (Scheme 1).
The structures of all mentioned compounds are to be
found in Schemes 1–4 and NMR data are given in Table 1.
2. Results and discussion
2.1 Photobromination of (Me2Si-CH2)3 (1)
A solution of (Me2Si-CH2)3 (1) and N-bromosuccinimide
in CCl4 was irradiated (UV) until 2 and 3 reached a ratio of
1:1 and about 10% of 1 remained. If 2 and 3 are isolated at
this point the compounds can be obtained in pure form. In
the same manner 2 was further brominated to give 3 in an
overall yield of about 90%. Only when trying to gain higher
yields of the ring 3 with two CBr2 groups or when trying to
introduce a third CBr2 group, some of the remaining ring 2
suffers bromination of a CH3 to a CBr3 group in opposite
position to the CBr2 group. This side product 4 is formed in
a yield of 5–10% at most.
In a first attempt to link the CBr2 groups in a mixture of
such C-functional trisilacyclohexanes by adding BuLi, the
compound Si6C18H39Br3 was isolated in a very low yield.
2.2 The reactions of 3 and 4 with metal alkyls
n
Whereas 3 is lithiated and subsequently alkylated by nBuLi,
t
the reactions with BuLi or MeMgBr proceed under ring
© 2000 NRC Canada

1390
Can. J. Chem. Vol. 78, 2000
Scheme 3. Preparation of the functional trisilaoct-3-ynes 10a and
10b.
Scheme 4. Linking of 10 and 4, and reaction with HBr to give
(ω-bromo-octynyl)-trisilacyclohexane 12.
contraction and ring opening to BrMe2Si-CH2-SiMe2-CϵC-
SiMe2Br or Me3Si-CH2-SiMe2-CϵC-SiMe3, respectively,
quite analogously to the reactions of (Cl2Si-CCl2)3 (6) or of
1,3-disila-4-trimethylsilylcyclopentene (8) with MeMgCl.
t
In 4, BuLi first attacks the CBr3 group as was proven by
trapping the lithiated intermediate with MeI.
terminal Si atom of 11 could be functionalized using HBr.
As CBr2 groups are readily lithiated by BuLi, a cyclization
n
With the formation of 4 in the photobromination of 2, and
with BrMe2Si-CH2-SiMe2-CϵC-SiMe2Br resulting from the
rearrangement of the lithiated derivative of 3, two molecules
were independently isolated which may have been combined
to give 6 in our orientating experiment. However, 6 could
not be detected in repeated experiments when treating a mix-
of 12 via elimination of LiBr was attempted. However, only
oxygen-containing products, probably resulting from ether
cleavage, were obtained. Therefore, a ring closure using me-
tallic Li was tried. In mass spectra of two out of fifteen
HPLC fractions (11) of this product mixture, a compound
with the molecular mass of Si6C18H41Br could be detected
which can be tentatively assigned to a partially hydroge-
nated isomer of 6.
t
ture of 3 and 4 with BuLi.
2.3 Organometallic synthesis of 6
On the way to 6, the organometallic synthesis of 4 is quite
a decisive step because 4 was not obtained in sufficient
amounts by photobromination. Starting from 2, a Si—Me
bond has to be cleaved using ICl (9), then the intermediate
-SiCl- unit can be substituted by Br3CLi. In this manner, 4
could be prepared in a yield of 70% as shown in Scheme 2.
As a second building block, initially compound 7 was
synthesized (Scheme 2). However it wasdecomposed by HBr
to BrMe2Si-CH2-SiMe2Br and further products while cleav-
ing the Ph—Si bonds. Therefore, the Si-functional 2,5,7-
trisilaoct-3-yne 10 had to be synthesized according to a
slightly changed reaction path (Scheme 3).
2.4 Conclusion
As only few crystals of 6 were initially obtained, it is not
reasonable to derive a synthetic procedure from this single
experiment. However, the investigations demonstrate a
reproducible synthetic path from (Me2Si-CH2)3 to 2,2,9-tri-
bromo-1,3,3,5,5,8,8,10,10,13,13-undecamethyl-1,3,5,8,10,13-
hexasilabicyclo[7.2.2]tridec-6-yne 6 even if the last step, i.e.,
the final ring closure has not yet been done.
2.5 Molecular structure determinations
The crystal structures of 6 and 4 were determined by X-
ray single crystal methods. The crystallographic data are
The linking of 4 with 10 was achieved via the Br2CLi in-
termediate 4a (10) as shown in Scheme 4. Then the remaining
© 2000 NRC Canada

Fritz et al.
1391
1
Table 1. H, 13C, and 29Si NMR data (δ in ppm) of the compounds 2–12. Bruker AMX 300, standard TMS (ext.); solvent CDCl3,
1
where indicated for H also C6D6; high field shifts with negative sign. Where the same number is used to mark different positions in
this table and in the reaction schemes, the signals could not unambigously be assigned.
Compound
2
3
4
5
δ1H
(1) 0.35; (2) 0.03; (3) 0.09
(1) –0.3; (2) 1.7; (3) 3.0; (4) 56.7
(1) 10.95; (3) –0.3
(3) 0.74; (2) 0.39; (1) 0.24
(1) –1.4; (2) 0.5; (3) –2.8; (4) 47.5
(3) 10.9; (2) 11.4
(3) 0.5 0.33; (2) 0.23 0.83; (1) 0.45; JAB 13.95 Hz
(1) –1.7; (2) –1.4; (3) –0.3; (4) 48.6 53.9
(3) 10.6; (1) 20.5
δ13C
δ29Si
δ1H
δ13C
δ29Si
δ1H
2
δ13C
δ29Si
δ1H, C6D6
δ13C
(3) 0.5 0.42; (2) 0.16 0.23; (1) 0.37; JAB 13.6 Hz
(1) 6.5; (2) 5.2; (3) –0.8; (4) 54.9
(3) 10.5; (1) 27.6
2
δ29Si
6
7
8
δ1H, C6D6
δ1H, C6D6
δ1H, C6D6
δ13C
0.21 (6H); 0.26 (6H); 0.48(6H); 0.59 (3H); 0.86 (AA′); 0.35 (XX′ + 12H + 2H)
(1) 0.39; (2) 0.08; (3) 0.12; (4) 0.37; (Ph) 7.6 7.2
(1) 0.25; (2) 0.32; (4) 0.32; (Ph) 7.25
(1) –0.4; (2) 7.0; (3) 4.5; (Ph) 139.6 139.4 133.2 127.8
(1) –4.5; (3) 30.5
(1) 0.33; (2) 0.0; (3) 0.08; (4) 2.12; (Ph) 7.32
(1) 1.0; (2) 2.6; (3) –0.3; () 90.4 93.9; (Ph) 141.1 140.3 133.2 127.7
(1) –3.7; (3) –17.1
δ29Si
9
δ1H, C6D6
δ13C
δ29Si
10a X = Cl
δ1H, C6D6
δ29Si
(1) 0.36; (2) 0.0; (3) 0.06; (4) 0.44; (Ph) 7.34
(1) –3.9; (3) –18.0; (4) –1.5
(2) 0.09; (1) 0.57; (2) 0.29 1.10; (3,3,5,7,9) 0.18 0.26 0.43 0.56 0.6; (Ph) 7.39; JAB 14.1 Hz
2
11
δ1H, C6D6
δ13C
(1) 0.4; (2) 0.2; (4) 54.9 56.7; (6) 110.6 118.3; (8) 2.5; (3,3,5,7,9) 0.9 –0.2 –0.3 –0.5 –0.6;
(Ph) 140.7 133.3 128.8 127.7
(3) 10.6; (1) 10.3; (5) –14.4; (7) –18.7; (9) –4.3
(8) 0.26; (1) 0.51; (2) 0.25 1.01; (3,3,5,7,9) 0.21 0.25 0.52 0.54 0.58
(1) –0.6; (2) 0.2; (4) 54.8 56.4; (8) 7.7; (9) 5.4; (6) 111.7 116.7; (3,3,5,7) –0.7 –0.5 –0.2 0.7
(3) 10.6; (1) 10.3; (5) –14.6; (7) –20.0; (9) 25.6
δ29Si
12
δ1H, C6D6
δ13C
δ29Si
listed in Table 2, and structural information is summarized
in Tables 3–8.3
184(2) pm. At first glance the Si—CC bond shortening may
indicate an influence of the CϵC triple bond. On the other
hand, the internal tension of the seven-membered section
(C7 → Si13) is reflected by the small bond angle C5-Si13-
C12 (105.8°) and the large bond angles Si6-C5-Si13
(118.2°), Si8-C9-Si10 (122.0°), and Si2-C7-Si8 (125.5°) as
well as by the deviation of the Si10-C11-C12-Si13 fragment
from linearity (7–8°). This tension seems to be responsible
for the shortening and bending of this fragment (Fig. 1). The
axial ligands C21, C41, C61, and Si13 are shifted away from
the central axis of the trisilacyclohexane ring. This demon-
strates the six-membered ring to be arched, which is indi-
cated, too, by the enlarged endocyclic angles at C1 and C3
(117–120°) and the smaller endocyclic angles at Si2 and Si4
(108–109°) as described by Fritz et al. (13).
2,2,9-Tribromo-1,3,3,5,5,8,8,10,10,13,13-undecamethyl-
1,3,5,8,10,13-hexasilabicyclo[7.2.2]tridec-6-yne Si6C18H39Br3
(6) crystallizes in the monoclinic space group P21/a with 4
molecules in the unit cell (12). The molecule of 6 with the
atomic labelling is shown in Fig. 1. Bond lengths and angles
are listed in Table 4 and 5, respectively.
The standard deviations of about 1 pm do not allow for a
sophisticated analysis. Nevertheless, some tendencies are
obvious because the standard deviation of the averaged bond
length d(Si—C) = 188(3) is three times as large as those for
the single values. This results from two significantly differ-
ent groups of bonds to be considered, namely the elongated
ones to the atoms C5 and C7, which are bonded to bromine
(d(Si—C5,7) = 192(1) pm); and the other ones (d(Si—C) =
186(1) pm). Moreover, the latter show a small differentiation
of d(Si—C)endo = 187(1) > d(Si—C)exo = 186(1) > d(Si—CC) =
Neglecting the CϵC group, the framework of 6 almost
perfectly displays a diamond fragment with all six-membered
rings being in the chair conformation (Fig. 2). It is noticeable,
3 Additional material has been deposited as supplementary material and may be purchased from the Depository of Unpublished Data, Docu-
ment Delivery, CISTI, National Research Council Canada, Ottawa, ON, Canada K1A 0S2 (http://www.nrc.ca/cisti/irm/unpub_e.shtml).
Crystallographic data for the structure(s) reported in this paper have been deposited with the Cambridge Crystallographic Data Centre
(CCDC 132465). Copies of the data can be obtained, free of charge, on application to the Director, CCDC, 12 Union Road, Cambridge CB2
1EZ, U.K. (Fax: 44-1223-336033 or e-mail: deposit@ccdc.cam.ac.uk).
© 2000 NRC Canada

1392
Can. J. Chem. Vol. 78, 2000
Table 2. Crystallographic data for 6 and 4.
Compound; formula
4; C9H19Br5Si3
6; C18H39Br3Si6
Molar mass
Crystal system
Space group
a
b
c
β
V
663.7 amu
monoclinic
P21/a (no. 14)
1425.2(4) pm
1856.2(5) pm
1152.7(3) pm
101.81(2)°
2986 × 106 pm3
1.477 g cm–3
4
610.5 amu
orthorhombic
Pbca (no. 61)
1262.3(12) pm
2355.5(18) pm
1285.2(26) pm
3821 × 106 pm3
2.124 g cm–3
8
ρ calc
Z
Diffractometer
Radiation
Syntex P3
Mo Kα
Siemens AED 2
Mo Kα
graphite monochromator
ω
graphite monochromator
ω – 2θ
Scan type
2θ range
3.5–55°
3–56°
Temperature
293 K
293 K
Total observed reflections
Total independent reflections
Reflections
Absorption correction
Absorption coefficient
R; Rw
7077
4636
3937
2728
4279 with I > 3σ(I)
Ψ-scan
45.33 cm–1
0.067; 0.054
1452 with I > 2σ(I)
Ψ-scan
105.9 cm–1
0.078; 0.065
Table 3. Atomic positional parameters and Ueq (104 pm2) of 6.
Estimated standard deviations are given in parentheses.
that only minor modifications are necessary to insert the
CϵC group into this diamond fragment, namely rotations
around the C9—Si10 and Si8—C9 bonds by angles of about
60° and 30°, respectively. These modifications result in the
skew boat conformation of the six-membered ring fragment
C81-Si8-C9-Si10-C11 (one atom is missing between C81
and C11).
A colourless single crystal of 2,2-dibromo-5-tribro-
momethyl-1,1,3,3,5-pentamethyl-1,3,5-trisilacyclohexane
Si3C9H19Br5 4 was selected for the X-ray structure determi-
nation and the results are summarized in Table 2. The mole-
cule of (4) with the atomic labelling is shown in Fig. 3;
fractional coordinates are given in Table 6; bond lengths and
angles are listed in Table 7 and 8, respectively.
The trisilacyclohexane ring adopts the chair conformation
with an equatorial CBr3 group, which is the necessary con-
formation for a successful further synthesis. Also in this
molecule the bond distances from the brominated carbon
atoms to Si as well as to Br are significantly larger than stan-
dard average values (Table 7). The endocyclic angles Si-C-Si
with an average value of 118.2° are larger than the tetrahe-
dral angle whereas the angles C-Si-C don’t show a com-
mon tendency. Due to the sterically requiring CBr3 group,
the angle C4-Si2-C7 of 111.1° is remarkably larger than
those at Si1 and Si3 (Table 8). The mean value of 113.0° for
all endocyclic angles indicates a slightly flattened conforma-
tion of the trisilacyclohexane chair.
a
Atom
x/a
y/b
z/c
Ueq
Br(5)
Br(71)
Br(72)
Si(2)
Si(4)
Si(6)
0.3283(1)
0.0188(1)
0.2245(1)
0.1940(2)
0.3666(2)
0.1637(2)
0.0958(2)
0.2260(2)
0.2445(2)
0.1134(5)
0.3083(5)
0.2702(3)
0.1349(5)
0.2058(6)
0.2166(6)
0.2213(6)
0.2160(6)
0.4295(6)
0.4603(6)
0.2010(7)
0.0681(6)
0.0073(6)
0.0362(7)
0.1355(7)
0.3489(7)
0.1416(7)
0.3457(7)
0.9461(1)
0.6908(1)
0.6818(1)
0.7667(1)
0.8657(1)
0.8591(1)
0.8175(1)
0.9634(1)
1.0112(1)
0.8236(4)
0.8180(4)
0.9191(4)
0.7464(4)
0.8646(4)
0.9844(4)
0.9950(4)
0.6787(5)
0.7977(5)
0.9245(5)
0.7844(5)
0.9061(5)
0.8790(5)
0.7701(5)
1.0171(5)
0.9861(5)
1.0598(5)
1.0770(5)
1.0745(1)
0.5455(1)
0.4771(1)
0.7008(2)
0.8465(2)
0.9198(2)
0.4165(2)
0.4158(2)
0.8341(2)
0.7696(7)
0.7068(7)
0.9066(7)
0.5394(7)
0.3882(8)
0.5711(7)
0.6748(8)
0.7767(8)
0.9543(8)
0.8090(9)
1.0306(8)
0.9795(9)
0.4630(9)
0.2768(8)
0.3137(8)
0.4047(8)
0.8711(9)
0.8682(9)
0.052(1)
0.051(1)
0.049(1)
0.031(1)
0.037(1)
0.035(1)
0.037(1)
0.040(1)
0.037(1)
0.033(3)
0.033(3)
0.031(3)
0.035(3)
0.049(4)
0.037(3)
0.037(3)
0.051(4)
0.061(4)
0.066(4)
0.054(4)
0.061(4)
0.058(4)
0.058(4)
0.058(4)
0.058(4)
0.063(4)
0.065(4)
Si(8)
Si(10)
Si(13)
C(1)
C(3)
C(5)
C(7)
C(9)
C(11)
C(12)
C(21)
C(41)
C(42)
C(61)
C(62)
C(81)
C(82)
C(101)
C(102)
C(131)
C(132)
3. Experimental
3.1 General procedures
All solvents were thoroughly dried, and the reactions run
in a N2 atmosphere. Gaseous Cl2 was passed through a P2O5
aEquivalent isotropic U defined as one third of the trace of the
orthogonalized tensor.
© 2000 NRC Canada

Fritz et al.
1393
Table 4. Bond lengths (in pm) for 6.
Bond
Length
Bond
Length
Bond
Length
Br(5)-C(5)
Si(2)-C(1)
Si(2)-C(21)
Si(4)-C(41)
Si(6)-C(5)
200.6 (7)
185.5 (9)
184.8 (9)
186.6 (9)
191.4 (8)
193.3 (8)
187.9 (9)
185.0 (9)
182.3 (9)
120.0 (9)
Br(71)-C(7)
Si(2)-C(3)
Si(4)-C(3)
Si(4)-C(42)
Si(6)-C(61)
Si(8)-C(9)
Si(10)-C(9)
Si(10)-C(102)
Si(13)-C(131)
196.3 (8)
187.6 (8)
187.7 (8)
184.4 (9)
188.6 (9)
188.1 (9)
187.4 (9)
183.1 (9)
184.4 (9)
Br(72)-C(7)
Si(2)-C(7)
Si(4)-C(5)
199.2 (8)
191.6 (8)
193.4 (8)
185.4 (8)
186.4 (9)
185.9 (9)
186.3 (9)
190.6 (8)
186.8 (9)
Si(6)-C(1)
Si(6)-C(62)
Si(8)-C(81)
Si(10)-C(11)
Si(13)-C(5)
Si(13)-C(132)
Si(8)-C(7)
Si(8)-C(82)
Si(10)-C(101)
Si(13)-C(12)
C(11)-C(12)
Table 5. Bond angles (in deg) for 6.
Fig. 2. Below: the diamond fragment of the molecule 6 after ab-
stracting the CϵC group (arrows). Real bonds are shown by full
lines, dotted lines help to complete the six-membered rings (all
in chair conformation) of the diamond fragment. Above: very
small modifications allow for the insertion of the CϵC group
(see text) by formation of only one skew boat fragment (atoms
81, 8, 9, 10, 11, and one atom missing between atoms 81 and 11).
Bond
Angle
Bond
Angle
C(1)-Si(2)-C(3)
C(1)-Si(2)-C(21)
C(3)-Si(2)-C(21)
C(3)-Si(4)-C(5)
C(3)-Si(4)-C(42)
C(5)-Si(4)-C(42)
C(1)-Si(6)-C(5)
C(1)-Si(6)-C(62)
C(5)-Si(6)-C(62)
C(7)-Si(8)-C(9)
C(7)-Si(8)-C(82)
C(9)-Si(8)-C(82)
C(9)-Si(10)-C(11)
C(9)-Si(10)-C(102)
107.7(4) C(1)-Si(2)-C(7)
111.2(4) C(3)-Si(2)-C(7)
111.7(4) C(7)-Si(2)-C(21)
109.1(3) C(3)-Si(4)-C(41)
107.0(4) C(5)-Si(4)-C(41)
112.4(4) C(41)-Si(4)-C(42)
107.1(4) C(1)-Si(6)-C(61)
109.8(4) C(5)-Si(6)-C(61)
114.1(4) C(61)-Si(6)-C(62)
108.6(4) C(7)-Si(8)-C(81)
108.7(4) C(9)-Si(8)-C(81)
108.4(4) C(81)-Si(8)-C(82)
109.3(4) C(9)-Si(10)-C(101)
109.2(4) C(11)-Si(10)-C(101)
109.8(3)
110.0(4)
106.5(4)
108.7(4)
112.8(4)
106.6(4)
111.9(4)
111.3(4)
102.7(4)
108.6(4)
113.4(4)
108.7(4)
110.8(4)
108.7(4)
C(11)-Si(10)-C(102) 106.1(4) C(101)-Si(10)-C(102) 112.6(4)
C(5)-Si(13)-C(12)
C(5)-Si(13)-C(132)
105.8(4) C(5)-Si(13)-C(131)
115.0(4) C(12)-Si(13)-C(131)
115.3(4)
108.9(4)
C(12)-Si(13)-C(132) 107.0(4) C(131)-Si(13)-C(132) 104.4(4)
Si(2)-C(1)-Si(6)
Br(5)-C(5)-Si(4)
Br(5)-C(5)-Si(13)
Si(4)-C(5)-Si(13)
Br(71)-C(7)-Br(72)
Br(71)-C(7)-Si(8)
Br(72)-C(7)-Si(8)
Si(8)-C(9)-Si(10)
C(11)-C(12)-Si(13)
116.5(4) Si(2)-C(3)-Si(4)
106.9(3) Br(5)-C(5)-Si(6)
101.7(3) Si(4)-C(5)-Si(6)
112.9(4) Si(6)-C(5)-Si(13)
107.7(4) Br(71)-C(7)-Si(2)
105.5(3) Br(72)-C(7)-Si(2)
104.9(4) Si(2)-C(7)-Si(8)
122.0(5) Si(10)-C(11)-C(12)
172.9(8)
120.0(5)
104.3(4)
111.3(4)
118.2(4)
106.0(4)
106.3(3)
125.5(4)
172.3(7)
tube (2 m × 8 cm). The photoreactions were run in quartz
flasks using a water cooled Original Hanau mercury high
pressure immersion lamp. HPLC analyses were performed
according to ref. 11.
diluted in pentane and prepurified using a silicagel column
(Merck 40–63 µm). Pentane and remaining 1 (10.6 g;
49 mmol) were removed, then 11.3 g (30.6 mmol, 60%) of 2
could be obtained as colourless crystals (mp 52°C).
Method 2: 21.6 g (100 mmol) of 1 + 19.6 g (110 mmol)
of N-bromosuccinimide in CCl4 were irradiated for 7 d. The
insoluble succinimide was removed by filtration, Br2 and
CCl4 stripped, the oily brown residue eluted with 1 L of
pentane from an Al2O3 column (10 × 2 cm; Riedel de Haen
Al2O3 S neutral active), and the solvent removed. With
3.2 Photobromination of (Me2Si-CH2)3 1 to give 2,2-
dibromo-1,1,3,3,5,5-hexamethyl-1,3,5-trisilacyclohexane (2)
Method 1: 21.6 g (100 mmol) of 1 + 3 drops of Br2 with-
out any solvent were irradiated. Only after the colour of Br2
disappeared was a new portion of Br2 added. Evolving HBr
was collected in a cooling trap (slightly reduced pressure,
liquid N2). Whenever the mixture turned black from sus-
pended carbon it had to be filtered. These steps were re-
1
peated until 3 became detectable in the H NMR spectrum
after a total irradiation time of about 150 h. The mixture was
© 2000 NRC Canada

1394
Can. J. Chem. Vol. 78, 2000
Table 6. Fractional atomic coordinates and Ueq of 4 with stan-
dard deviations in parentheses.
Fig. 3. Structure of 2,2-dibromo-5-tribromomethyl-1,1,3,3,5-
pentamethyl-1,3,5-trisilacyclohexane Si3C9H19Br5 (4) with the
atomic labelling (hydrogen atoms omitted). For bond lengths and
angles see Table 7 and 8, respectively.
Atom
x/a
y/b
z/c
Ueq
Br(1)
Br(2)
Br(3)
Br(4)
Br(5)
Si(1)
Si(2)
Si(3)
C(1)
C(2)
C(3)
C(4)
C(5)
C(6)
C(7)
C(8)
C(9)
0.3701(1)
0.2553(1)
0.3554(4)
0.3303(2)
0.1377(2)
0.1381(3)
0.2202(3)
0.3365(3)
0.2706(9)
0.038(1)
0.101(1)
0.157(1)
0.128(1)
0.259(1)
0.343(1)
0.255(1)
0.469(1)
0.08540(7)
–0.01401(6)
0.28645(8)
0.32302(7)
0.33136(7)
0.1009(2)
0.2103(2)
0.0939(2)
0.0683(5)
0.0785(7)
0.0765(7)
0.1790(6)
0.2174(7)
0.2864(6)
0.1725(6)
0.0734(7)
0.0605(8)
0.3884(1)
0.5127(1)
0.4277(1)
0.6618(1)
0.5098(2)
0.4620(3)
0.5814(3)
0.6259(3)
0.5025(9)
0.559(1)
0.330(1)
0.466(1)
0.693(1)
0.546(1)
0.615(1)
0.740(1)
0.638(1)
0.0511(5)
0.0513(6)
0.0625(7)
0.0684(7)
0.0742(8)
0.030(1)
0.027(1)
0.031(1)
0.025(4)
0.047(6)
0.053(6)
0.032(5)
0.044(5)
0.042(5)
0.033(5)
0.050(6)
0.057(7)
Table 7. Bond lengths (in pm) for 4.
Bond
Length
Bond
Length
Bond
Length
Br(1)—C(1)
Si(2)—C(6)
Si(1)—C(2)
Br(2)—C(1)
Si(3)—C(1)
Si(2)—C(5)
197(1)
191(1)
185(2)
195(1)
189(1)
185(1)
Br(5)—C(6)
Si(3)—C(9)
Si(2)—C(7)
Si(1)—C(1)
Br(4)—C(6)
Si(3)—C(8)
192(1)
186(2)
184(1)
191(1)
195(1)
185(2)
Si(1)—C(4)
Br(3)—C(6)
Si(3)—C(7)
Si(2)—C(4)
Si(1)—C(3)
186(1)
194(1)
186(1)
184(1)
185(2)
another 19.6 g (110 mmol) of NBS, the full procedure was
repeated. In a short path distillation apparatus (oil pump vac-
uum) first 2.3 g (11 mmol) of 1, then 26.6 g (55 mmol) of 2
were collected. A third fraction of 8.3 g consisted of about
50 mol% of 2 and 50 mol% of 3. 20.6 g of 2 (55%) re-
mained after recrystallization from pentane. As the recovered
starting material was re-used for further photobrominations,
2 could be obtained in an overall yield of 90%.
of dry Cl2 was applied for 2 min, and the mixture stirred for
15 min. This was repeated until the signals of 2 disappeared
from the 1H NMR spectrum of the solution. After all volatile
components had been removed, 9.9 g (25.2 mmol; 94%) of 5
crystallized (mp 87.5°C).
3.5 Preparation of 4 from 5
At –110°C, 9 mL (13.5 mmol) of nBuLi were added
dropwise to a solution of 5.0 g (15.2 mmol) of CBr4 in
200 mL THF:pentane:Et2O (4:1:1). After stirring for 1 h at
this temperature the solution was frozen (liquid N2). A solu-
tion of 5.4 g (13.7 mmol) of 5 in 50 mL THF–pentane–Et2O
was added, the reaction flask allowed to warm to –80°C and
the mixture stirred for another 3 h at this temperature. After
gradually warming to room temperature the solvents were
removed, hexane added and the solution filtered. 5.9 g of 4
(9.7 mmol; 72%) were recrystallized from hexane (mp
137°C, decomp.).
3.3 Photobromination of (Me2Si-CH2)3 1 to give 2,2,4,4-
tetrabromo-1,1,3,3,5,5-hexamethyl-1,3,5-
trisilacyclohexane 3 and 2,2-dibromo-5-tribromomethyl-
1,1,3,3,5-pentamethyl-1,3,5-trisilacyclohexane (4)
According to method 2, 36.4 g of a mixture of about 10
mol% of 1, 80 mol% of 2 and 10 mol% of 3 (1H NMR anal-
ysis) + 19.6 g (110 mmol) of NBS in CCl4 were irradiated
for 10 d, and the procedure repeated for a second time. At
30–50°C, the collected residues were distilled in the men-
tioned apparatus yielding a main fraction which after
recrystallization gave 12.4 g (33 mmol; 33%) of 2, 16.5 g
(31 mmol; 31%) of 3 as well as 1.8 g (3 mmol; 3%) of 4.
From the distillation residue another small amount of 4 and
3 was obtained by crystallization.
3.6 Preparation of 5-phenyl-3,3,5-trimethyl-3,5-
disilahex-1-yne (9)
75 g (0.31 mol) of 8 in 100 mL THF were added at 0°C to
a solution of HCϵCMgBr (14) (8.5 g (0.35 mol) Mg turn-
ings; 21.6 mL (0.35 mol) EtBr; 400 mL THF). After
refluxing for 2 h, the ether was removed, the residue sus-
pended in pentane and filtered, and the solvent stripped. A
vacuum distillation yielded 51.7 g (0.22 mol; 71%) of 9 (bp
60°C; 2 × 10–3 mm Hg).
3.4 Cleavage of a Si—Me bond in 2 to give 2,2-
dibromo-5-chloro-1,1,3,3,5-pentamethyl-1,3,5-
trisilacyclohexane (5)
To 10 g (26.8 mmol) of 2 in 60 mL of CCl4, a solution of
1.4 g of I2 (20 mol%) was added; in the dark a slow stream
© 2000 NRC Canada

Fritz et al.
1395
Table 8. Bond angles (in deg) for 4.
3.10 Reaction of 11 with HBr to give (ω-bromo-
octynyl)-trisilacyclohexane (12)
Bond
Angle
Bond
Angle
4.3 g (5.25 mmol) of 11 were placed in an ampoule and
frozen (liquid N2). 6 mmol of HBr were added, the ampoule
sealed and stored at –78°C. After 1 d, the ampoule was
warmed to room temperature, vigorously shaken and stored
again at –78°C. This was repeated for 7 d. A raw yield of
3.9 g of 12 remained when all volatile components had been
removed.
C(1)-Si(1)-C(2)
C(4)-Si(2)-C(7)
C(1)-Si(3)-C(7)
Br(1)-C(1)-Br(2)
Si(1)-C(4)-Si(2)
Br(4)-C(6)-Br(5)
C(2)-Si(1)-C(4)
C(7)-Si(2)-C(6)
C(7)-Si(3)-C(9)
Br(2)-C(1)-Si(3)
Br(3)-C(6)-Br(4)
C(1)-Si(1)-C(4)
C(4)-Si(2)-C(6)
C(1)-Si(3)-C(9)
Br(1)-C(1)-Si(3)
Si(2)-C(6)-Br(4)
107.6(6)
111.1(6)
105.9(6)
108.4(6)
118.3(7)
108.0(7)
110.7(7)
107.1(6)
112.7(7)
107.7(6)
107.8(7)
106.1(6)
106.9(6)
109.2(6)
106.2(6)
110.4(7)
C(2)-Si(1)-C(3)
C(7)-Si(2)-C(5)
C(7)-Si(3)-C(8)
Br(2)-C(1)-Si(1)
Si(2)-C(6)-Br(5)
C(1)-Si(1)-C(3)
C(4)-Si(2)-C(5)
C(1)-Si(3)-C(8)
Br(1)-C(1)-Si(1)
Si(2)-C(6)-Br(3)
Si(2)-C(7)-Si(3)
C(3)-Si(1)-C(4)
C(5)-Si(2)-C(6)
C(8)-Si(3)-C(9)
Si(1)-C(1)-Si(3)
Br(3)-C(6)-Br(5)
110.9(8)
112.9(6)
110.3(7)
109.2(6)
111.9(7)
110.2(7)
113.1(7)
109.7(6)
105.8(6)
110.4(7)
117.3(7)
111.3(7)
105.2(7)
108.9(7)
119.0(6)
108.3(7)
3.11 Reaction of 12 with Li granules
Over 11 h, 1.95 g (2.4 mmol) of 12 in 500 mL THF were
added to a vigorously stirred suspension of 44.3 mg
(6.4 mmol) of Li (metal granules) in 50 mL THF. Dry work-
up gave 1.28 g of raw product of which 533 mg were sepa-
rated by HPLC.
Acknowledgements
We gratefully acknowledge the financial support of this
work by “Deutsche Forschungsgemeinschaft” and “Fonds
der Chemischen Industrie.”
3.7 Preparation of Me2SiBr2
To 4.2 g (70 mmol) of Me2SiH2 (no solvent) at –35°C
(cooling bath) very cautiously and as slowly as possible
22.1 g (140 mmol) of Br2 were dropwise added. The reflux
condenser of the apparatus was connected to a cryostate
(–35°C). When the mixture turned brown, the remaining Br2
was added at room temperature. Eventually the mixture was
References
1. G. Fritz and E. Matern. Carbosilanes. Springer–Verlag, Berlin,
Heidelberg, New York, Tokyo. 1986.
2. G. Fritz and M. Hähnke. Z. Anorg. Allg. Chem. 390, 191
(1972); A. Lipka, H.G. v. Schnering, Z. Anorg. Allg. Chem.
419, 20 (1976).
1
heated to 55°C until the H NMR peak of Me2SiHBr was no
longer detectable. Distillation at normal pressure yielded
12.3 g (40 mmol) of Me2SiBr2.
3. A. Lipka and H.G. v. Schnering. Z. Anorg. Allg. Chem. 399,
257 (1973); G. Fritz, H.J. Dannappel, and E. Matern. Z.
Anorg. Allg. Chem. 399, 263 (1973).
4. G. Fritz, H. Köhler, and D. Kummer. Z. Anorg. Allg. Chem.
374, 54 (1970); G. Fritz and G. Marquardt. Z. Anorg. Allg.
Chem. 404, 1 (1974); G. Sawitzki and H.G. v. Schnering. Z.
Anorg. Allg. Chem. 425, 1 (1976).
5. G. Fritz, K.-P. Wörns, K. Peters, E.-M. Peters, and H.G. v.
Schnering. Z. Anorg. Allg. Chem. 512, 126 (1984).
6. G. Fritz and N. Szczepansky. Z. Anorg. Allg. Chem. 367, 44
(1969).
7. G. Fritz, U. Finke, and W. Speck. Z. Anorg. Allg. Chem. 481,
51 (1981); G. Fritz and W. Speck. Z. Anorg. Allg. Chem. 481,
60 (1981).
8. G. Fritz and P. Böttinger. Z. Anorg. Allg. Chem. 395, 159
(1973).
9. G. Fritz and J. Honold. Z. Anorg. Allg. Chem. 556, 23 (1988);
M. Bordeau, S.M. Djamei, J. Dunogues, and R. Calas. Bull.
Soc. Chim. Fr. 1982, II-159.
10. R.H. Fischer and G. Köbrich. Chem. Ber. 101, 3230 (1968).
11. G. Fritz and K.-P. Wörns. Z. Anorg. Allg. Chem. 512, 103
(1984); G.Fritz, H. Reuss, K.-P. Wörns, and A. Wörsching. Z.
Anorg. Allg. Chem. 512, 93 (1984).
12. G.M. Sheldrick. SHELXTL-Plus, Siemens Industrial Automa-
tion, Inc. 1994.
3.8 Preparation of 2-chloro-7-phenyl-2,5,5,7-
tetramethyl-2,5,7-trisilaoct-3-yne 10a and of 2-bromo-7-
phenyl-2,5,5,7-tetramethyl-2,5,7-trisilaoct-3-yne (10b)
n
At 0°C, 15.7 mL (23.8 mmol) of BuLi were added over
45 min to 5.8 g (25 mmol) of 9 dissolved in 50 mL hexane.
After another 45 min stirring at room temperature, the
mixture is transferred with 250 mL hexane into a dropping
funnel and added over 7 h to 3.25 g (25 mmol) of Me2SiCl2
in 100ml hexane. Dry work-up with filtration and distillation
yielded 3.3 g (10.2 mmol; 43%) of 10a (bp 78°C; 5 ×
10–3 mm Hg).
Similarly, from 13.5 g (58.2 mmol) of 9, 35.5 mL
n
(57.5mmol) of BuLi and 12.6 g (58.3 mmol) of Me2SiBr2 a
yield of 10.1 g (27.4 mmol; 48%) of 10b (bp 91°C; 5 ×
10–3 mm Hg) resulted.
3.9 Linking of 10 and 4 to give (ω-phenyl-octynyl)-
trisilacyclohexane (11)
At –105°C, 4.0 mL (6.3 mmol) of BuLi were added to
n
3.87 g (6.4 mmol) of 4 in 90 mL THF:pentane:Et2O (4:1:1).
After stirring for 30 min, 2.1g (6.5 mmol) of 10a in 50 mL
pentane were added at this temperature. The mixture was
stirred at –105 to –85°C for another 2 h, then filtered and
distilled in vacuo. 4.93 g of 11 were obtained as a slightly
yellow, highly viscous oil.
13. G. Fritz, H. Volk, M. Straub, H.G. v. Schnering, K. Peters, and
E.-M. Peters. J. Organomet. Chem. 341, 109 (1988).
14. L. Skattebol, E.R.H. Jones, and M.C. Withing. Org. Synthesis.
4, 792 (1963).
Similarly, 8.2 g of 11 resulted from 6.8 g (11.2 mmol) of
4, 6.9 mL nBuLi (11.2 mmol) and 4.2 g (11.4 mmol) of 10b.
© 2000 NRC Canada