Tridentate Lewis Acids Based on 1,3,5-Trisilacyclohexane Backbones and an Example of Their Host-Guest Chemistry

Article abstract of DOI:10.1002/chem.201501683

Directed tridentate Lewis acids based on the 1,3,5-trisilacyclohexane skeleton with three ethynyl groups [CH₂Si(Me)(C₂H)]₃ were synthesised and functionalised by hydroboration with HB(C₆F₅)₂, yielding the ethenylborane {CH₂Si(Me)[C₂H₂B(C₆F₅)₂]}₃, and by metalation with gallium and indium organyls affording {CH₂Si(Me)[C₂M(R)₂]}₃ (M=Ga, In, R=Me, Et). In the synthesis of the backbone the influence of substituents (MeO, EtO and iPrO groups at Si) on the orientation of the methyl group was studied with the aim to increase the abundance of the all-cis isomer. New compounds were identified by elemental analyses, multi-nuclear NMR spectroscopy and in some cases by IR spectroscopy. Crystal structures were obtained for cis-trans-[CH₂Si(Me)(Cl)]₃, all-cis-[CH₂Si(Me)(H)]₃, all-cis-[CH₂Si(Me)(C₂H)]₃, cis-trans-[CH₂Si(Me)(C₂H)]₃ and all-cis-[CH₂Si(Me)(C₂SiMe₃)]₃. A gas-phase electron diffraction experiment for all-cis-[CH₂Si(Me)(C₂H)]₃ provides information on the relative stabilities of the all-equatorial and all-axial form; the first is preferred in both solid and gas phase. The gallium-based Lewis acid {CH₂Si(Me)[C₂Ga(Et)₂]}₃ was reacted with a tridentate Lewis base (1,3,5-trimethyl-1,3,5-triazacyclohexane) in an NMR titration experiment. The generated host-guest complexes involved in the equilibria during this reaction were identified by DOSY NMR spectroscopy by comparing measured diffusion coefficients with those of the suitable reference compounds of same size and shape.

Full text of DOI:10.1002/chem.201501683

DOI: 10.1002/chem.201501683
Full Paper
&
Tridentate Lewis Acids
Tridentate Lewis Acids Based on 1,3,5-Trisilacyclohexane
Backbones and an Example of Their Host–Guest Chemistry
Eugen Weisheim, Christian G. Reuter, Peter Heinrichs, Yury V. Vishnevskiy, Andreas Mix,
Beate Neumann, Hans-Georg Stammler, and Norbert W. Mitzel*[a]
Dedicated to Professor Manfred Scheer on the occasion of his 60th birthday
Abstract: Directed tridentate Lewis acids based on the 1,3,5-
trisilacyclohexane skeleton with three ethynyl groups
[CH2Si(Me)(C2H)]3 were synthesised and functionalised by hy-
droboration with HB(C6F5)2, yielding the ethenylborane
{CH2Si(Me)[C2H2B(C6F5)2]}3, and by metalation with gallium
and indium organyls affording {CH2Si(Me)[C2M(R)2]}3 (M=Ga,
In, R=Me, Et). In the synthesis of the backbone the influ-
ence of substituents (MeO, EtO and iPrO groups at Si) on the
orientation of the methyl group was studied with the aim to
increase the abundance of the all-cis isomer. New com-
pounds were identified by elemental analyses, multi-nuclear
NMR spectroscopy and in some cases by IR spectroscopy.
Crystal structures were obtained for cis-trans-[CH2Si(Me)(Cl)]3,
all-cis-[CH2Si(Me)(H)]3,
all-cis-[CH2Si(Me)(C2H)]3,
cis-trans-
gas-
[CH2Si(Me)(C2H)]3 and all-cis-[CH2Si(Me)(C2SiMe3)]3.
A
phase electron diffraction experiment for all-cis-
[CH2Si(Me)(C2H)]3 provides information on the relative stabili-
ties of the all-equatorial and all-axial form; the first is pre-
ferred in both solid and gas phase. The gallium-based Lewis
acid {CH2Si(Me)[C2Ga(Et)2]}3 was reacted with a tridentate
Lewis base (1,3,5-trimethyl-1,3,5-triazacyclohexane) in an
NMR titration experiment. The generated host–guest com-
plexes involved in the equilibria during this reaction were
identified by DOSY NMR spectroscopy by comparing mea-
sured diffusion coefficients with those of the suitable refer-
ence compounds of same size and shape.
Introduction
dentate Lewis acid shows the complexation of halogens as
a polyanionic chain in the crystal.[9] Investigations with other
substrates like molecules with carbonyl groups[7,10] and acety-
lenes[11] have revealed the potential of this mercury compound
in coordination chemistry.
The first attempts to establish poly-Lewis acids were docu-
mented in 1968[1] and this subject has continued to grow since
then. Such functional compounds are able to form complexes
with Lewis bases and anions. Issues of molecular recognition
have been put into the foreground of such investigations. Par-
allel to these investigations poly-cations were studied as recep-
tors for certain anions. The first prototypes were synthesised
by Park and Simmons.[1] These catapinands[2] were able to com-
plex halides selectively due to the fitting sizes of cavity and
anion. Later on in 1977 Schmidtchen established spherical me-
thylated poly-ammonium systems with high stability constants
of their complexes with anions in water.[3] Further on some
cyclic poly-Lewis acids based on silicon,[4] tin[5] and mercury[6,7]
were reported. A prominent example of a mercury-based poly-
Lewis acid was synthesised by Sartori and Golloch.[8] This tri-
Besides cyclic poly-Lewis acids some representatives with di-
rected Lewis acid functions have also been synthesised. Ben-
zene, naphthalene and anthracene have been used as underly-
ing rigid skeletons. In the case of benzene, functionalities with
aluminium, gallium[12] and mercury[13] have been introduced. In
this way bidentate acids have been created that are capable of
selectively binding certain Lewis bases. Naphthalene offers also
a rigid backbone for selective binding of Lewis bases; exam-
ples include compounds with mercury,[14] boron,[15] indium[16]
and gallium.[17] Based on boron, Katz has synthesised 1,8-an-
thracenediethynylbis(catecholboronate),[18] a bidentate Lewis
acid with a fixed distance between the acid functions. Using
1,8-diethynylanthracene we have recently reported a series of
bidentate earth metal Lewis acids[19] and also a selective com-
plexation of such bidentate Lewis acids with pyridine and pyri-
midine as substrates.[20] Encouraged by these results with the
ethynyl-functionalised scaffolds we have now addressed triden-
tate backbones based on 1,3,5-trisilacyclohexane to establish
directed tridentate Lewis acids.
[a] E. Weisheim, C. G. Reuter, P. Heinrichs, Dr. Yu. V. Vishnevskiy, Dr. A. Mix,
B. Neumann, Dr. H.-G. Stammler, Prof. Dr. N. W. Mitzel
Lehrstuhl für Anorganische Chemie und Strukturchemie
Centrum für Molekulare Materialien CM2
Fakultät für Chemie, Universität Bielefeld
Universitätsstraße 25, 33615 Bielefeld (Germany)
Fax: (+49)521-106-6026
Herein we present syntheses of several 1,3,5-trisilacyclohex-
anes and demonstrate their use as versatile backbones for tri-
dentate Lewis acids. We demonstrate that all-cis-1,3,5-triethyn-
E-mail: mitzel@uni-bielefeld.de
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201501683.
Chem. Eur. J. 2015, 21, 12436 – 12448
12436
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
yl-1,3,5-trimethyl-1,3,5-trisilacyclohexane offers a possibility to
access tridentate Lewis acids by hydroboration and metalation
using gallium- and indiumtrialkyls under alkane elimination.
methyldimethoxymethylsilane (2) with magnesium led to an
overall yield of 41% of a mixture of all-cis-4 and cis-trans-4 in
a ratio of 1:1. The diastereomer mixture of 5 was obtained in
lower yield (32%), but contains 64% of the desired all-cis ste-
reoisomer.
Results and Discussion
All attempts to separate the diastereomers failed and conse-
quently 5 was employed in subsequent reactions.
In order to place three Lewis acidic functionalities on one side
of a molecule that point in the same direction, we chose 1,3,5-
trisilacyclohexanes as backbones (Scheme 1). The strategy was
all-cis-1,3,5-Triethynyl-1,3,5-trimethyl-1,3,5-trisilacyclohex-
ane
A direct route to the desired trisilacyclohexane by reacting eth-
ynyl magnesium bromide with 5 turned out to be not possible.
Consequently, we transformed 5 into the trichloro derivative 6
by treatment with boron trichloride. All attempts to separate
the diastereomers of 6 failed. Therefore we treacted 5 with
LiAlH4 to afford the corresponding trihydride 7 (Scheme 2),
a compound synthesised earlier in a different way.[28,29] In con-
trast to 5 and 6, the diastereomers of 7 could be separated by
fractionated condensation, as was earlier reported.[28] The de-
sired all-cis stereoisomer of 7 was obtained in a yield of 58%
starting from the mixture of 5.
In order to obtain the all-cis isomer of 6, all-cis-7 was reacted
with chlorine gas dissolved in carbon tetrachloride. This proce-
dure has been reported to chlorinate the cyclic 1-chloro-1-
phenyl-1-sila-1,2,3,4-tetrahydronaphthalene under retention of
configuration.[30] However, this protocol failed when applied to
7 and led instead to a diastereomeric mixture of all-cis- and
cis-trans-6, that is, a loss of stereo information.
For that reason the detour via all-cis-7 was not further pur-
sued and instead isomer mixture 6 was reacted with ethynyl
magnesium bromide to yield an isomer mixture of the desired
trisilacyclohexane 8 (Scheme 2). The diastereomeric ratio in
several implementations of this reaction was 60:40 in favour of
all-cis-8. At this step the diastereomers could be separated by
fractional sublimation at room temperature under high
vacuum conditions.
Scheme 1. Two possible diastereomers resulting from the orientation of sub-
stituents at the ring silicon atoms.
to bind three ethynyl groups to the silicon atoms, which will
be subsequently carry typical Lewis acidic functions with ele-
ments of the boron group.
With one function at each silicon atom there are two possi-
bilities for diastereomers—the desired all-cis isomer with three
equally oriented substituents and the cis-trans isomer, the
second product of the cyclisation reaction. The dynamics of
ring inversion will change the orientation of these substituents,
in the case of all-cis isomer from all-equatorial (denoted eee) to
all-axial (aaa). Because the barriers of inversion for 1,3,5-trisila-
cyclohexanes are known to be very small,[21,22] a trifunctional
system as a host will be easily adaptable in binding to a guest
molecule regarding this change in conformation. The challeng-
es lie in finding conditions for a preferred formation of the all-
cis isomer, the separation of diastereomers and their conver-
sions under retention of configuration.
Formation of 1,3,5-trisilacyclohexanes
Kriner has established the first
elegant preparative access to
1,3,5-trisilacyclohexanes.[23,24]
Based on this and other
work[25,26] dichloro(chloromethyl)-
methylsilane (1) was protected
with methoxy- and ethoxy-sub-
stituents (Scheme 2). In this way
the formation of 1,3,5-trisilacy-
clohexanes is favoured over that
of 1,3-disilacyclobutanes.[27] Me-
thoxy and ethoxy substituents
were tested for an enhanced se-
lectivity in formation of the all-
cis stereoisomer (compounds 2
and 3); however, with isopro-
poxy groups, no formation of
1,3,5-trisilacyclohexane was ob-
Scheme 2. Synthesis of 1,3,5-trisilacyclohexanes and the backbone 8. Yields: 2: 83%; 3: 86%; 4: 41%; 5: 32%; 6
served. The reaction of chloro- with BCl3: 89%; 6 with Cl2: 100%; all-cis-7: 58%; 8: 95%.
Chem. Eur. J. 2015, 21, 12436 – 12448
www.chemeurj.org 12437
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
The compounds described in this section were characterised
by multinuclear NMR spectroscopy and elemental analyses,
some in addition by FT-IR spectroscopy, mass spectrometry
and single-crystal X-ray diffraction. Some discussion of the
spectroscopic and structural data follows here.
NMR spectroscopy
NMR spectra are provided in the Supporting Information, Fig-
ures S6–S69, along with a compilation of NMR shifts in
Table S6. The chemical shifts of the geminal hydrogen atoms
at the rings cover the range between 0.47 and À0.64 ppm. Var-
iation of the ring substituents has no appreciable effect on the
chemical shifts of methyl or methylene hydrogen atoms. In
contrast, the 29Si chemical shifts depend strongly on a change
in substituents from chlorine to ethynyl (26.4 to À18.5 ppm),
in the case of all-cis diastereomers of 6 and 8.
The dynamics of ring inversion was a subject of earlier stud-
ies, but so far there are only theoretically obtained values for
the barrier to ring inversion of 1,3,5-trisilacyclohexanes[21,22]
that estimated this value to be about 5 kcalmolÀ1. We ran low-
temperature NMR spectra for all-cis-8 in toluene at tempera-
tures as low as À908C, but, besides some broadening of the
lines, no splitting of signals was observable that would allow
us to extract a value for an inversion barrier.
Figure 1. Molecular structure of cis-trans-6 in the crystal. Displacements ellip-
soids are drawn at the 50% probability level. Methyl hydrogen atoms are
omitted for clarity. Selected distances [Š]: Si(1)ÀC(1) 1.862(1), Si(1)ÀC(3)
1.858(1), Si(2)ÀC(1) 1.863(1), Si(2)ÀC(2) 1.865(1), Si(3)ÀC(2) 1.862(1), Si(3)ÀC(3)
1.861(1), Si(1)ÀC(4) 1.866(1), Si(2)ÀC(5) 1.863(1), Si(3)ÀC(6) 1.870(1), Si(1)À
Cl(1) 2.090(1), Si(2)ÀCl(2) 2.086(1), Si(3)ÀCl(3) 2.089(1). Selected angles [8]:
C(1)-Si(1)-C(3) 110.3(1), C(2)-Si(2)-C(1) 110.0(1), C(3)-Si(3)-C(2) 109.2(1), Si(1)-
C(1)-Si(2) 118.4(1), Si(2)-C(2)-Si(3) 116.0(1), Si(3)-C(3)-Si(1) 116.9(1).
Crystal structures
During these studies we obtained single-crystalline material
suitable for structure determination by X-ray diffraction of cis-
trans-6, all-cis-7 and both diastereomers of 8. The structures
will be presented in comparison below. Single crystals suitable
for X-ray diffraction experiments were obtained of cis-trans-6
by sublimation, and of 7 and all-cis-8 (X-ray structure of cis-
trans-8 is presented in Supporting Information, Figure S3) by
gradual evaporation of a solution in n-pentane. The molecular
structures are displayed in Figures 1–3.
Figure 2. Molecular structure of all-cis-7 in the crystalline state. Displacement
ellipsoids are shown at the 50% probability level. Hydrogen atoms of the
methyl groups are omitted for clarity. Selected distances [Š]: Si(1)ÀC(1)
1.875(3), Si(1)ÀC(3) 1.874(3), Si(2)ÀC(1) 1.874(3), Si(2)ÀC(2) 1.870(3), Si(3)ÀC(2)
1.877(3), Si(3)ÀC(3) 1.870(3), Si(1)ÀC(4) 1.863(3), Si(2)ÀC(5) 1.870(3), Si(3)ÀC(6)
1.868(3), Si(1)ÀCl(1) 1.223(1), Si(2)ÀCl(2) 1.214(1), Si(3)ÀCl(3) 1.190(1). Selected
angles [8]: C(1)-Si(1)-C(3) 109.0(1), C(2)-Si(2)-C(1) 108.7(1), C(3)-Si(3)-C(2)
108.7(1), Si(1)-C(1)-Si(2) 113.7(2), Si(2)-C(2)-Si(3) 114.0(2), Si(3)-C(3)-Si(1)
113.7(2).
The trisilacyclohexane rings of cis-trans-6, all-cis-7 and all-cis-
8 all exhibit chair conformations. The SiÀC bond lengths of the
rings have values that are similar to those reported for
[SiCl2CH2]3.[31] On average, the shortest SiÀC bonds are found in
the chlorinated species, [SiCl2CH2]3 and cis-trans-6 (average
1.862(1) Š), while those in the ethynyl substitute all-cis-8 (aver-
age 1.869(1) Š) are longer and those in the hydride species all-
cis-7 are the longest (average 1.873(3) Š).
The SiÀCl bond lengths in the chlorine derivative cis-trans-6
vary from 2.086(1) to 2.090(1) Š and are longer than those in
[SiCl2CH2]3 (2.043(1)–2.059(1) Š). Increasing substitution by
chlorine at silicon leads to a shortening of all bonds to this
atom.
tential binding sites for Lewis acidic functions in compounds
with all-cis-8 being used as a backbone.
This structure of all-cis-8 in the crystalline state does not
necessarily reflect the behaviour of a free molecule in the gas
phase or in solution, in which an easy change between the all-
axial (aaa) or all-equatorial (eee) orientation of the ethynyl
groups can be expected. More light on this behaviour is shed
by the data from a gas-phase structure investigation reported
below.
Compound all-cis-8 crystallises in the monoclinic space
group P21/n with eight molecules per unit cell, that is, two in-
dependent ones in the asymmetric unit. In both molecules the
ethynyl groups are arranged equatorially. This leads to large
distances between the terminal ethynyl carbon atoms:
C(5)···C(8) 7.648(1), C(8)···C(11) 7.849(1) and C(5)···C(11)
7.695(1) Š. These values define the distance between the po-
Axial verus equatorial orientation of ethynyl groups—quan-
tum-chemical calculations for all-cis-8
A series of quantum-chemical calculations of the energies of
the conformers of compound all-cis-8 was carried out to esti-
Chem. Eur. J. 2015, 21, 12436 – 12448
www.chemeurj.org
12438
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 3. Molecular structure of one of the two independent molecules of
all-cis-8 in the crystal. Displacement ellipsoids are shown at the 50% proba-
bility level. Hydrogen atoms of methyl groups are omitted for clarity. Select-
ed distances[Š]: Si(1)ÀC(1) 1.866(1), Si(1)ÀC(3) 1.870(1), Si(2)ÀC(1) 1.867(1),
Si(2)ÀC(2) 1.868(1), Si(3)ÀC(2) 1.870(1), Si(3)ÀC(3) 1.870(1), Si(1)ÀC(6) 1.862(1),
Si(2)ÀC(9) 1.865(1), Si(3)ÀC(12) 1.863(1), Si(1)ÀC(4) 1.857(1), Si(2)ÀC(7)
1.844(1), Si(3)ÀC(10) 1.836(1), C(4)ꢀC(5) 1.195(2), C(7)ꢀC(8) 1.194(2), C(10)ꢀ
C(11) 1.194(2). Selected angles [8]: C(1)-Si(1)-C(3) 109.6(1), C(2)-Si(2)-C(1)
108.7(1), C(3)-Si(3)-C(2) 107.7(1), Si(1)-C(2)-Si(2) 116.5(1), Si(2)-C(2)-Si(3)
115.4(1), Si(3)-C(3)-Si(1) 116.5(1).
Figure 4. Experimental (open circles) and model (line) radial distribution
curves of all-cis-1,3,5-triethynyl-1,3,5-trimethyl-1,3,5-trisilacyclohexane (eee
all-cis-8) as determined by GED. Vertical bars indicate the contributions of in-
dividual interatomic distances, some are labelled. The difference curve for
the eee conformer model is shown below.
phase. The best fit with the model was achieved for a content
of 99% eee conformer. The fit of this model is best seen in
Figure 4, which displays the radial distribution curve obtained
from the GED experiment. However, the accuracy of this com-
position value is rather limited. The error 3s is 30%, meaning
that the amount of eee all-cis-8 in the gas phase at the temper-
ature of the experiment (381–388 K) is between 84 and 100%.
This is consistent with the DFT calculations, but contradicts the
results of the MP2 calculation. The result is also consistent with
the data from the solid state, which often represent the con-
formational ground state in absence of pronounce intermolec-
ular forces.
Table 1. Comparison of quantum-chemical results of conformational
composition concerning the ratio of axial to equatorial orientation of eth-
ynyl groups for all-cis-8. DE=EaaaÀEeee; DG=GaaaÀGeee; ceee calculated for
T=379 K from DG.
Method
DE [kcalmolÀ1
]
DG [kcalmolÀ1
]
ceee [%]
B3LYP/6-31G(d,p)
PBE0/6-31G(d,p)
PBE0-D3/6-31G(d,p)
MP2/cc-pVTZ
1.02
1.47
1.51
À0.27
0.77
1.52
1.76
À0.28
74
88
91
41
mate their relative abundance. The used methods and their re-
sults are given in Table 1. The DFT calculations at the B3LYP
and PBE0 approximations using the double-z basis set indicate
that the dominant component of all-cis-8 is that with equatori-
ally (eee) orientated ethynyl groups. Using corrections for dis-
persion, the aaa conformer was not found to be significantly
stabilised by dispersive interactions. In contrast to the DFT re-
sults, calculations at the MP2 level resulted in a different con-
formational composition with an aaa/eee conformer ratio of
59:41 (Table 1). An attempt to determine a barrier to ring in-
version turned out to be complicated due to the flatness of
the potential energy surface and in the light of the expected
limited reliability was not further pursued.
Experimental structure parameter values of eee all-cis-8 are
listed in Table 2 together with those from quantum-chemical
calculations for comparison. Compound all-cis-8 has similar
angles and distances in the crystal and in the gas phase. The
angles in the cycle in the gas phase deviate less than one
Table 2. Experimental and calculated structural parameters (re, a) of
equatorially oriented [CH2Si(Me)(C2H)]3 (eee all-cis-8). Levels of theory DFT
(PBE0-D3/6-31G(d,p)), MP2/cc-pVTZ and XRD average values [distances in
Š, angles in 8].
Parameters
GED
PBE0-D3 MP2
XRD
re or a
rg
r(SiÀCH2)
1.860(5) 1.871(5)
1.885(5) 1.899(5)
1.843(5) 1.854(5)
1.204(3) 1.211(3)
1.879
1.877
1.843
1.225
1.881
1.869
r(SiÀCH3)
1.880
1.843
1.215
1.863
1.846
1.195
Gas-phase structure of all-cis-8
r(SiÀCꢀ)
r(CꢀC)
Experimental studies of the composition and structure of free
molecules of all-cis-8 were undertaken by gas-phase electron
diffraction (GED). A model for the composition of all-cis-8 con-
sisted of the eee (all ethynyl groups equatorial) and aaa con-
formers (all ethynyl groups axially). The experiment showed
that conformer eee all-cis-8 is the most abundant in the gas
a(SiÀCH2ÀSi)
a(CH2ÀSiÀCH2)
a(CH3ÀSiÀCꢀ)
a(CH2ÀSiÀCꢀ)
a(CH3ÀSiÀCH2)
f(SiÀCH2ÀSiÀCH2)
117.2(9)
116.3
116.0
116.1
106.4(18)
108.8(28)
109.0(10)
111.8(16)
54.0(32)
108.9
106.7
108.4
112.1
51.5
109.6
107.0
108.2
111.9
50.8
108.9
106.9
107.6
112.1
51.8
Chem. Eur. J. 2015, 21, 12436 – 12448
www.chemeurj.org
12439
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
degree from those in the crystal (Table 2). The distances be-
tween the carbon atoms in the ethynyl groups are 0.009 Š
shorter in the solid state than in the gas phase.
Table 3. Selected NMR shifts in [ppm] of solutions of the tridentate Lewis
acids 9 (C6D6), 10 (C6D6/Et2O), 11 (C6D6/Et2O), 12 ([D8]THF).
Compound
9
10
11
12
In summary, the ethynyl groups are arranged equatorially,
but it is known from the work of Arnason et al.[21,22]—as men-
tioned above—that the 1,3,5-trisilacyclohexane cycle has an
energy barrier for inversion (estimated to 5.5 kcalmolÀ1) of half
the height calculated for cyclohexane. Consequently, the three
equally oriented functions can easily swing in con-rotatorial
fashion for simultaneous binding to a substrate. For that
reason compound all-cis-8 was provided with Lewis acid func-
tionalities and converted with a tridentate Lewis base. For de-
tails, follow the next chapters.
-SiCH2Si-
-SiCH2Si-
-SiCH3
0.10
0.62
0.17
0.32
115.6
no resonance
À22.3
0.45
0.17
0.23
0.11
À0.06
À0.06
0.23
0.13
-SiCꢀC-M/-SiC=CÀM 173.2
no resonance 115.9
no resonance 135.6
À21.2
-SiCꢀC-M/-SiC=CÀM 150.7
-CH2SiCH2-
À4.6
À24.2
a single broad resonance at 59 ppm corresponding to a tri-co-
ordinate boron atom. The 29Si NMR spectrum contains a single
resonance at À4.6 ppm (Table 3).
Tridentate gallium and indium Lewis acids were synthesised
using GaMe3, GaEt3 or InMe3 in alkane elimination reactions in
analogy to earlier work of our group, but varying condi-
tions.[19,34] Compound 10 was afforded by a solvent-free reac-
tion of 8 with trimethylgallium and heating to 428C for 24 h
(Scheme 3), while compound 11 formed from 8 and triethyl-
gallium also within 24h without heating; both reactions pro-
ceeded quantitatively. Due to the solid nature of trimethylindi-
um at room temperature, the reaction with this reagent was
performed in toluene, affording compound 12 in 80% yield.
Products 10–12 are colourless, less soluble (in donor free sol-
vents) solids and were characterised by NMR (Table 3) and FT-
IR spectroscopy, as well as elemental analyses. They are likely
aggregated in the solid state, but so far all attempts to grow
single crystals for structure elucidation failed.
Synthesis of tridentate Lewis acids
Having established a synthesis for all-cis-8, we now have
a backbone available with C3 symmetry, free of donor units
and with ethynyl groups oriented to the same side of the mol-
ecule that can be converted into Lewis acid functions, so that
the synthesis of directed tridentate Lewis acids becomes feasi-
ble.
With the aim of preparing an example for a boron-based tri-
dentate Lewis acid all-cis-8 was reacted with Piers’ borane,[32,33]
HB(C6F5)2, in n-pentane (Scheme 3). The reaction proceeded
The solubilities of 10–12 are generally very low in hydrocar-
bon solvents and on the order of about 1 mgmLÀ1 for 11. This
is likely due to aggregation. In order to record good quality so-
lution NMR spectra, compounds 10 and 11 were dissolved in
C6D6/Et2O mixtures and 12 in [D8]THF. Compounds 10 and 11
are stable in the named solvents, but compound 12 undergoes
redistribution as is indicated by the observation of further res-
onances in all NMR spectra in addition to the expected signals.
Only the resonances assigned to the desired compound 12 are
listed in Table 3 and in the Experimental Section. The supposed
redistribution reaction is shown in Scheme 4.
The metalation of ethynyl groups with gallium organyl
groups causes a small downfield shift of the resonances of the
methyl and methylene groups. The functionalisation with
indium organyl groups (compound 12) has barely an influence
on those resonances (compare Table 3 and Table S6 in the Sup-
Scheme 3. Syntheses of the directed tridentate Lewis acids 9, 10, 11 and 12.
quantitatively and yielded product 9 as beige solid in high
purity. Compound 9 was characterised by elemental analysis
and multinuclear NMR spectroscopy. All attempts to grow suit-
able crystals for X-ray diffraction failed. All NMR spectra were
1
porting Information). As an example the H NMR spectrum of
compound 11 is shown in Figure 5. The corresponding reso-
nances are assigned in the figure. The identity of 11 is con-
firmed by comparing the NMR shifts of triethylgallium
(1.43 ppm [triplet] and 0.93 ppm [quartet], neat, referenced to
a C6D6 capillary) with those of the tridentate Lewis acid 11
(1.35 ppm [triplet] and 0.61 ppm [quartet]). The same down-
field shift was observed by comparing the proton resonances
of 10 (À0.04 ppm) with those of trimethylgallium (0.21 ppm,
1
recorded in C6D6 at ambient temperature. The H NMR spec-
trum shows a typical signal pattern for the olefin protons of
the trans-product: two doublets at 7.64 and 7.48 ppm with
coupling constants of 21 Hz. Two doublet resonances of the
methylene protons in the cycle are observed at 0.10 and
À0.06 ppm with a geminal coupling constant of 13.9 Hz
(Table 3). The 19F NMR spectrum shows the three expected res-
onances: two multiplets at À129.05 and À161.04 ppm (ortho-
neat)
and
12
(À0.44 ppm)
with
trimethylindium
and meta-fluorine) and one as
a
triplet of triplets at
(À0.24 ppm[35]). Resonances of the carbon atoms in the ethynyl
À146.71 ppm (para-fluorine). The 11B NMR spectrum shows
groups of the gallium compounds could not be observed,
Chem. Eur. J. 2015, 21, 12436 – 12448
www.chemeurj.org
12440
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Host–guest complex formation observed by DOSY NMR in-
vestigations
We have recently reported the complexation of a bidentate
gallium Lewis acid based on an anthracene framework with
pyridine and pyrimidine under dynamic conditions in solution
by a combination of NMR titration and diffusion NMR experi-
ments.[20] Here we describe the conversion of 11 (called host,
H) with the tridentate Lewis base 1,3,5-trimethyl-1,3,5-triazacy-
clohexane (TMTAC, called guest, G) as an example for molecu-
lar recognition of a tridentate Lewis acid.
Diffusion NMR spectroscopy is a versatile tool for the analy-
sis of molecular size,[36] analysis of compound mixtures and de-
termination of complex formation in solution.[37] The method
provides self-transitional diffusion coefficients D that depend
on the size and shape of the investigated system. For a reliable
interpretation of hydrodynamic radii it is indispensable to com-
pare the diffusion coefficients D of the analyte with values of
suitable model systems that are similar in structure (if this is
not close to spherical). For these reasons some model com-
pounds (see Scheme 5: 13–16) were synthesised that mimic
the structures of likely host–guest adducts in size and shape—
as far as achievable with a reasonable synthetic effort. Their
syntheses are described in the Supporting Information. Their
diffusion coefficients D were determined in [D8]toluene. They
are shown below the reference compounds and listed in
Table 4.
Scheme 4. Supposed redistribution of compound 12 in THF.
Table 4. Results of the diffusion NMR measurements of host (11), TMTAC
(guest) and reference compounds in [D8]toluene at 294 K.
Compound
D [10À10 m2 sÀ1
]
Compound
D [10À10 m2 sÀ1
]
11
TMTAC
13 (monomer)
4.2
12.4
8.6
14
15
16
6.5
5.9
5.6
Figure 5. 1H NMR spectrum of 11 in C6D6 with 10 equivalents of Et2O. # de-
notes the resonances of Et2O. The resonances of 11 are assigned in the
figure.
Figure 6 displays the NMR titration of 11 with TMTAC. Ratios
of compound 11 (H) and TMTAC (G) are listed on the left side
of the figure. Resonances marked with * represent the signals
of TMTAC (in spectra b, c and d the resonances of TMTAC are
overlaid by the methyl resonance of the residual protons of
likely due to the strong quadrupole broadening caused by the
gallium nucleus. The corresponding chemical shifts of 12
(Table 3) receive a strong downfield shift in comparison to 8
(94.5 and 90.1 ppm). The chemical shifts of the silicon atoms of
8, 10, 11 and 12 have almost the same value (compare Table 3
and Table S6 in the Supporting Information), confirming that
the silicon atoms all bear ethynyl groups.
1
toluene). H NMR spectrum Figure 6a shows host 11 without
the guest. The resonances of 11 are marked by ꢁ1 in all spec-
tra. The diffusion coefficients were determined at resonances
labelled with a N.
Further proof for the identity of these CꢀC units stems from
FT-IR spectra. Bands were observed at 2034 (10, 11 and 12)
and 2029 cmÀ1 (12). Elemental analysis data confirm the above
structural assignments.
The host 11 dissolved in toluene has a diffusion coefficient
of D=4.2”10À10 m2 sÀ1 (Table 4). Comparison with the diffusion
coefficient of compound 13, as reference for a monomeric
host, indicates that compound 11 does not adopt a monomeric
structure in solution. In this case, the method of diffusion coef-
ficient formula weight correlation was applied (for more infor-
mation on this methodology see reference [38]). Therefore, the
logarithmic diffusion coefficients D were plotted against the
logarithmic molecular weights of TMTAC and model com-
Chem. Eur. J. 2015, 21, 12436 – 12448
www.chemeurj.org
12441
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 5. Model of the monomer 13 and references 14–16 for host molecule 11 and likely structures of host–guest complexes of 11 with TMTAC (1,3,5-tri-
methyl-1,3,5-triazacyclohexane) 17–19. Experimental diffusion coefficients D for 14--16 measured in [D8]toluene at 294 K are listed below the compound num-
bers.
of 1378 gmolÀ1 for this aggre-
gate of 11 was estimated (taking
into account the different
atomic weight of Si and Ga). This
corresponds to trimeric aggre-
gate of 11 in solution.
With further addition of
TMTAC the resonances of com-
ponent 11 (Figure 6b, ꢁ1 )
become smaller and finally dis-
appear at a ratio of H:G=1:1
(Figure 6c), while new signals
appear (labelled ꢁ2 ), which rep-
resent more than one new com-
plex species. This relation can be
explained by the fact, that the
Lewis base molecules can bind
more than one host molecule
11.
The corresponding diffusion
coefficients are listed in Table 5
and are a plot of the diffusion
coefficients against the equiva-
Figure 6. 1H NMR spectra of the host compound 11 (H), the guest TMTAC (G) and the conversion of 11 with
TMTAC in different mixtures in [D8]toluene at 294 K (600 MHz); # denotes the residual proton resonance of tolu-
lents of TMTAC added is dis-
played in Figure 7. Mixture ratios
*
ene; resonances labelled with are those of TMTAC.
from 1:0.1 to 1:0.5 (H:G) result in
diffusion coefficients in the
pounds 14–16. This yields the equation of a regression line,
log D=À0.4597xÀ7.9336 (x=log MW) (Figure S5, Supporting
information). Using this equation and the diffusion coefficient
derived from the signals of compound 11, a molecular weight
range between 4.4”10À10 and 4.7”10À10 m2 sÀ1. By applying
the equation from above and considering the small values of
the diffusion coefficient for the ratio 1:0.5 (H:G), we infer the
presence of a dimeric host structure in interaction with one
Chem. Eur. J. 2015, 21, 12436 – 12448
www.chemeurj.org
12442
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 5. Results of the diffusion NMR measurements of the conversion of
host 11 and TMTAC (guest).
Molar ratio
H:G
D (host)
[10À10 m2 sÀ1
Molar ratio
H:G
D (host)
[10À10 m2 sÀ1
]
]
1:0.1
1:0.2
1:0.4
1:0.5
1:0.7
4.4
4.4
4.6
4.7
6.6
1:0.8
1:1
1:1.2
1:1.5
6.8
6.8
7.0
7.0
Figure 7. Diffusion coefficients of the conversion of 11 with TMTAC. Data are
listed in Table 5. Horizontal lines shows the reference compound of mono-
mer 13 (dashed line) and a host–guest complex 1:1 14 (dash-and-dot line).
1
Figure 8. Temperature-dependent H NMR spectra of a 1:1.2 mixture of host
11 and guest TMTAC in [D8]toluene; # denotes the residual proton reso-
TMTAC molecule (calculated molecular weight: 1079 gmolÀ1
compare
2MW,13 +MW,TMTAC =(2”463.07+129.20) gmolÀ1
1055.34 gmolÀ1).
;
*
nance of toluene; with labelled resonances are those of TMTAC.
=
Spectra of host–guest mixtures of around 1:1 show broad
resonances ꢁ2 (Figure 6c), which points to a dynamic arrange-
ment of host and guest molecules. Therefore temperature-de-
Table 6. Results of the diffusion NMR measurements of the conversion of
11 (host) and TMTAC (guest). Listed are the molar ratio H:G and the diffu-
sion coefficients D [10À10 m2 sÀ1].
1
pendent H NMR spectra of a 1:1.2 (H:G) mixture were record-
ed to prove the dynamics.
H:G
D (Host)
D (Host)
D (Host)
D (Guest)
Figure 8 shows this dynamic behaviour at both the reso-
nance of TMTAC (3.5 ppm, 353 K) and that of component 11
(triplet, 1.5 ppm). At a temperature of 273 K additional reso-
nances appear and upon cooling to 193 K further broad reso-
nances occur. These additional signals represent further coordi-
nation species. Diffusion coefficients of 1:0.7 to 1:1.5 mixtures
were determined to 6.6”10À10 and 7.0”10À10 m2 sÀ1 at room
temperature (Table 5), which are comparable with that of refer-
ence molecule 14 (6.5”10À10 m2 sÀ1, Table 4).
Despite the dynamic processes at room temperature, com-
pound 11 evidently forms a one-to-one complex with TMTAC
at a ratio close to 1:1. The fact that the diffusion coefficient of
the host–guest complex is slightly larger than the value of ref-
erence 14 indicates a slightly smaller hydrodynamic radius.
This suggests that TMTAC coordinates to the three gallium
1:1.5
1:2
1:5
7.0
6.2
6.2
5.7
5.6
11.2
10.5
8.1
8.7
13.1
6.6
5.9
1:10
9.5
atoms of host 11 with all three nitrogen atoms to give the 1:1
aggregate 17 (Scheme 5).
If the ratio is increased to ꢂ1:1.5, other species form
(Figure 6, triplets ꢁ3 , doublets ꢁ4 , quartets ꢁ5 , singlets ꢁ6 ).
Their diffusion coefficients (Table 6) are comparable to those of
model compounds 15 and 16 (Table 4). This underlines the
fact that TMTAC is dynamically bonded to 11, and —if avail-
able—further TMTAC molecules can cleave aggregate 17 and
coordinate additional guest molecules at ratios above 1:1.5.
Chem. Eur. J. 2015, 21, 12436 – 12448
www.chemeurj.org
12443
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
lower ratios of the experiments (1:0.5 and higher) aggregates
of 11 with additional TMTAC ligands are present in solution,
while the pure host 11 seems to be trimeric in solution.
In essence, the all-cis form of 1,3,5-triethynyl-1,3,5-trisilacy-
clohexanes offers a well accessible framework for the genera-
tion of systems with three concordantly oriented functions. De-
spite the predefined orientation, the high conformational flexi-
bility of the 1,3,5-trisilacyclohexane rings (inversion) allows for
a simultaneous movement of reactive sites to a given sub-
strate. We are currently exploring this ability in our laboratories
with a multitude of different reactive sites.
Scheme 6. Host–guest chemistry of the system 11/TMTAC during the titra-
tion with TMTAC. H designates host 11 and guest TMTAC is denoted by G.
In essence, the host–guest chemistry of the system 11/
TMTAC shows a dynamic behaviour, so that at H:G ratios be-
tween 1:0 and 1:0.5 host 11 seems to be a dimer with addi-
tional TMTAC coordination in solution (Scheme 6, H2G) and at
ratios of 1:1, a one-to-one complex is present (Scheme 6, HG).
At higher concentrations of TMTAC (H:G ꢂ1:1.5) a two- and
threefold coordination of TMTAC to 11 is observed.
Experimental Section
General methods
All manipulations were performed under dried argon or nitrogen
using Schlenk and glove-box techniques. Tetrahydrofuran was
dried with potassium, pentane, hexane and diethyl ether with
LiAlH4. They were distilled prior to use. Dichloro(chloromethyl)me-
thylsilane (1) (Sigma–Aldrich) was distilled before use. Piers’ borane
Conclusion
Tridentate Lewis acids with unidirectional functions were syn-
thesised with 1,3,5-trisilacyclohexanes as basic skeletons that
bear three symmetrically bonded ethynyl units carrying the
acid functions. Preparative pathways were explored to gener-
ate these skeletons in a way with optimised selectivity con-
cerning the formation of the all-cis isomers of the 1,3,5-trie-
thynyl-1,3,5-trisilacyclohexanes. Such a way was found in first
was synthesised according to
a
reported protocol.[32,33] C6D6,
[D8]THF and [D8]toluene were dried with Na/K alloy and con-
densed. NMR measurements were performed on Bruker Avance III
500, Bruker Avance III 300 and Bruker DRX 500 instruments. NMR
spectra were referenced to the residual signal of used protonated
solvents (1H, 13C) or external standards (11B: BF3·OEt3, 19F: CCl3F; 29Si:
TMS). GC/EI-MS analyses were done using a Shimadzu GC-2010/
GCMS-QP 2010S instrument (capillary column: Rtx-200 Crossbond,
trifluoropropylmethylpolysiloxane, 30 m, 25 mm, 0.25 mm). EI and
CI mass spectra were recorded using an Autospec X magnetic
sector mass spectrometer with EBE geometry (Vacuum Generators,
Manchester, UK) equipped with a standard EI source. FT-IR spectra
were recorded on a Bruker ALPHA FT-IR-spectrometer. Elemental
analyses were performed with CHNS elemental analyser HEKAtech
EURO EA (too low values for carbon are due to the known forma-
tion of silicon carbide).
cyclising
diethoxy(chloromethyl)methylsilane,
(EtO)2Si(Me)CH2Cl, with magnesium to afford 1,3,5-triethoxy-
1,3,5-trimethyl-1,3,5-trisilacyclohexane (5), which was then
chlorinated with boron trichloride to give 1,3,5-trichloro-1,3,5-
trimethyl-1,3,5-trisilacyclohexane (6). This was reacted with eth-
ynyl magnesium bromide to give 1,3,5-triethynyl-1,3,5-trimeth-
yl-1,3,5-trisilacyclohexane (8). Only at this step was it possible
to separate the diastereomers by fractional sublimation to
afford pure all-cis-8. Spectroscopy and structural analysis by X-
ray diffraction proofed the identity unidirectional functionality.
Gas-phase electron diffraction (GED) and quantum chemical in-
vestigations showed that the free molecules adopt an all-equa-
torial arrangement of the ethynyl functions; however, there is
a low barrier to ring inversion to the all-axial conformer (too
low to be determined in solution by NMR spectroscopy).
Quantum chemical calculations
Quantum-chemical calculations using DFT[39] and MP2[40] approxi-
mations with built-in 6-31G(d,p) and cc-pVTZ basis sets have been
carried out by using Gaussian 03 program package.[41] In all cases
for optimised structures calculations of frequencies have been
done to prove the existence of minima on the potential energy
surface. Calculations of numeric cubic force fields were performed
on the B3LYP/6–31G(d,p) level of theory.
Some examples for tridentate Lewis acids were obtained.
The reaction of all-cis-8 with three equivalents of Piers’ borane
yielded 1,3,5-tris[bis(pentafluorophenyl)boranylethenyl]-1,3,5-
trimethyl-1,3,5-trisilacyclohexane (9) under reduction of the
ethynyl units. Metalation with gallium- and indium trialkyls
gave the 1,3,5-tris(dialkylmetallynylethynyl)-1,3,5-trimethyl-
1,3,5-trisilacyclohexanes, [(Me)(R2M-CꢀC)SiCH2]3 (M/R=Ga/Me
(10), Ga/Et (11), In/Me (12)) under alkane elimination.
The host–guest chemistry of [(Me)(Et2GaÀCꢀC)SiCH2]3 (11)
with the tridentate guest 1,3,5-trimethyl-1,3,5-triazacyclohex-
ane (TMTAC) was investigated. NMR titration and DOSY NMR
experiments were performed to analyse this dynamic system
in solution. The experiments revealed, that 11 forms a one-to-
one host guest complex with TMTAC only at ratios around 1:1
(host/guest). At ratios ꢂ1:1.5 (host:guest) two and three
TMTAC molecules bind to the tridentate Lewis acid 11. At
Gas-phase electron diffraction experiment
The electron diffraction patterns were recorded using the im-
proved Balzers Eldigraph KD-G2 gas-phase electron diffractome-
ter[42] at Bielefeld University. The experimental conditions are pre-
sented in Table 7. The electron diffraction patterns were measured
on Fuji BAS-IP MP 2025 imaging plates, which were scanned using
a calibrated Fuji BAS-1800II scanner. The intensity curves (Figure S1
in Supporting Information) were obtained by applying a method
described earlier.[43] Sector function and electron wavelengths were
refined[44] using diffraction patterns of CS2, recorded along with the
substances under investigation and comparing with standard pa-
rameters for CS2.[45]
Chem. Eur. J. 2015, 21, 12436 – 12448
www.chemeurj.org
12444
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
49.7 (-SiOCH3), 3.8 (s, -SiCH2Si-), 3.4 (s, -SiCH2Si-), 0.7 (s, -SiCH3),
0.6 ppm (s, -SiCH3); 29Si{1H} NMR (99 MHz, C6D6, 298 K): d=16.8 (s,
-CH2SiCH2-), 16.4 ppm (s, -CH2SiCH2-); GC/EI-MS: retention time
8.13 min (all-cis): m/z (%): 249.0 (100) [M+ÀCH3], retention time
7.96 min (cis-trans): m/z (%): 264.0 [M+], 249.0 (100) [M+ÀCH3]; FT-
Table 7. Details of the GED experiments for 8 for the measurements at
short and long camera distances.
Parameter
Short
Long
dnozzleÀdetector [mm]
Uacceleration [kV]
250.0
60
500.0
60
~
IR (KBr): n=2956, 2913, 2830, 1464, 1409, 1353, 1254, 1190, 1047,
1046, 1044, 1041, 1039, 1038, 1035, 1033, 1029, 940, 847, 842, 840,
837, 834, 832, 830, 828, 826, 824, 820, 815, 814, 809, 807, 803, 791,
789, 785, 782, 762, 741, 623 cmÀ1; elemental analysis calcd (%) for
C9H24O3Si3 (264.54): C 40.86, H 9.14; found: C 40.41, H 9.11.
Ifast electrons [mA]
lelectron [Š]
1.1
0.048718
381
1.7”10À6
8.3”10À8
5
0.9
0.048736
388
1.4”10À6
2.3”10À7
5
[a]
Tnozzle [K]
[b]
pchamber [mbar]
presidual gas[c] [mbar]
texposure [s]
s range [Š
inflection points[d]
À1
]
7.0–31.2
4
2.2–16.6
2
1,3,5-Triethoxy-1,3,5-trimetyl-1,3,5-trisilacyclohexane (5)
Compound 5 was synthesized by using a protocol analogous to
that for 4, but with compound 3 as the starting material. B.p. 838C
at 1 mmHg (1.3 mbar) (Literature values:[23,24] 68–698C at 0.2–
0.25 mmHg/978C at 1.7 mmHg). Yield: 32% (Literature values:
30.4–33.2%). The product was a mixture of two diastereomers, all-
[a] Determined from CS2 diffraction patterns measured in the same ex-
periment. [b] During the measurement. [c] Between measurements.
[d] Number of inflection points on the background lines (see Figure S1 of
Supporting Information).
1
cis and cis-trans. Assignments for all-cis-5: H NMR (500 MHz, C6D6,
3
298 K): d=3.58 (q, 3JH,H =7.0 Hz, 6H; -SiOCH2CH3), 1.16 (t, JH,H
=
Gas-phase electron diffraction structural analysis
2
7.0 Hz, 9H; -SiOCH2CH3), 0.25 (d, JH,H =13.6 Hz, 3H; -SiCH2Si-), 0.15
(s, 9H; -SiCH3), À0.06 ppm (d, JH,H =13.6 Hz, 3H; -SiCH2Si-); 13C{1H}
2
The structural analysis was performed with the UNEX program.[46]
All refinements were done using two intensity curves simultane-
ously (see Figure S1 of Supporting Information), one from the
short and another from the long camera distance. These were ob-
tained by averaging independently measured intensity curves. The
total intensities were transformed into molecular intensity curves
(Figure S2 of Supporting Information) by background elimination
using cubic splines. The structure of 8 was assumed to be of C3v
symmetry. For the definition of the geometry in the form of a Z-
matrix and grouping of structure parameters in least-squares re-
finements see Table S1 (Supporting Information). The differences
between values of parameters in one group were kept fixed at the
values taken from B3LYP/6-31G(d,p) calculations. Scale factors for
mean square amplitudes have been refined in groups (see Sup-
porting Information Table S2). Thus, the ratios between different
amplitudes in one group were fixed on the theoretical values cal-
culated using the SHRINK program.[47] Correlation coefficients are
provided in Table S3 (Supporting Information).
NMR (126 MHz, C6D6, 298 K): d=57.9 (s, -SiOCH2CH3), 18.9 (s,
-SiOCH2CH3), 4.6 (s, -SiCH2Si-), 1.6 ppm (s, -SiCH3); 29Si{1H} NMR
(99 MHz, C6D6, 298 K): d=12.7 ppm (s, -CH2SiCH2-); assignments for
3
cis-trans-5: 1H NMR (500 MHz, C6D6, 298 K): d=3.57 (d, JH,H
=
7.0 Hz, 4H; -SiOCH2CH3), 3.47 (q, 3JH,H =7.0 Hz, 2H; -SiOCH2CH3),
1.16 (t, 3JH,H =7.0 Hz, 6H; -SiOCH2CH3), 1.10 (t, 3JH,H =7.0 Hz, 3H;
2
-SiOCH2CH3), 0.26 (s, 6H; -SiCH3), 0.23 (s, 3H; -SiCH3), 0.19 (d, JH,H
=
2
13.9 Hz, 1H; -SiCH2Si-), 0.04 (d, JH,H =13.6 Hz, 2H; -SiCH2Si-), À0.06
2
2
(d, JH,H =13.9 Hz, 1H; -SiCH2Si-), À0.08 ppm (d, JH,H =13.6 Hz, 2H;
-SiCH2Si-); 13C{1H} NMR (126 MHz, C6D6, 298 K): d=58.0 (s,
-SiOCH2CH3), 57.9 (s, -SiOCH2CH3), 18.9 (s, -SiOCH2CH3), 18.7 (s,
-SiOCH2CH3), 4.8 (s, -SiCH2Si-), 4.5 (s, -SiCH2Si-), 1.4 (s, -SiCH3),
1.4 ppm (s, -SiCH3); 29Si{1H} NMR (99 MHz, C6D6, 298 K): d=14.1 (s,
-CH2SiCH2-), 13.8 ppm (s, -CH2SiCH2-); GC/EI-MS: retention time
8.5 min (all-cis and cis-trans): m/z (%): 291 (100) [M+ÀCH3], 261
[M+ÀOEt], 247 [M+ÀCH3ÀOEt], 217, 203, 189, 175, 145, 117, 103,
85, 73, 59, 45; MS (ESI, positive): m/z=329.2 (100) [M++Na], 307.2
[M++H]; accurate mass (MS ESI): m/z calcd for C12H30O3Si3Na+:
~
329.13950; found: 329.13980; FT-IR (KBr): n=2973, 2925, 2896,
2873, 1481, 1443, 1391, 1354, 1293, 1255, 1166, 1109, 1084, 1037,
947, 824, 783, 761, 739, 625 cmÀ1; elemental analysis calcd (%) for
C12H30O3Si3 (Mr =306.62): C 47.01, H 9.86; found: C 46.64, H 9.89.
1,3,5-Trimethoxy-1,3,5-trimetyl-1,3,5-trisilacyclohexane (4)
A solution of compound 2 (27.85 g, 180 mmol) in THF (250 mL)
was added dropwise to a mixture of THF (250 mL) and magnesium
turnings (5.0 g, 206 mmol, surface activated with 1,2-dibromo-
ethane) while heating to 708C. After the addition of 2 was com-
pleted, the grey suspension was refluxed for 4 h. Subsequently the
solvent was removed and the remaining grey solid was extracted
with pentane. The pentane was removed under reduced pressure
(50 mbar) and the remaining oil distilled to afford product 4, b.p.
56–598C at 2.7 mbar. Yield: 6.49 g (24.5 mmol, 41%). The product
was a mixture of two diastereomers: all-cis- and cis-trans-4. Assign-
1,3,5-Trichloro-1,3,5-trimethyl-1,3,5-trisilacyclohexane (6)
a) Compound 5 (7.54 g, 24.6 mmol) was dissolved in n-pentane
(80 mL). After three freeze–pump–thaw cycles boron trichloride
(81.2 mmol measured as gas) was condensed onto the frozen solu-
tion of 5 at À1988C. The reaction mixture was warmed to À208C.
At this temperature the reaction started under formation of a waxy
solid. After warming to room temperature overnight all volatiles
were removed under reduced pressure and the residual colourless
solid was sublimed at 508C and 0.03 mbar. Product 6 was obtained
as a colourless solid in a yield of 89% (6.08 g, 21.9 mmol).
1
ments for all-cis-4: H NMR (500 MHz, C6D6, 298 K): d=3.31 (s, 9H;
2
-SiOCH3), 0.22 (d, JH,H =13.6 Hz, 3H; -SiCH2Si-), 0.12 (s, 9H; -SiCH3),
À0.12 ppm (d, 2JH,H =13. 6 Hz, 3H; -SiCH2Si-); 13C{1H} NMR
(126 MHz, C6D6, 298 K): d=49.7 (s, -SiOCH3), 3.5 (s, -SiCH2Si-),
1.0 ppm (s, -SiCH3); 29Si{1H} NMR (99 MHz, C6D6, 298 K): d=
15.3 ppm (s, -CH2SiCH2-); assignments for cis-trans-4: 1H NMR
b) Compound 7 (39 mg, 0.22 mmol) was dissolved in tetrachloro-
methane (2 mL). A saturated solution of chlorine in tetrachlorome-
thane (2 mL) was added dropwise over a period of 1 min. The
formed hydrogen chloride and excess of chlorine were removed
by purging nitrogen through the solution. Tetrachloromethane was
removed under reduced pressure to afford the product as a waxy
solid; yield 100% (60.5 mg, 22 mmol).
(500 MHz, C6D6, 298 K): d=3.30 (s, 6H; -SiOCH3), 3.20 (s, 3H;
2
-SiOCH3), 0.23 (s, 6H; -SiCH3), 0.19 (s, 3H; -SiCH3), 0.16 (d, JH,H
=
2
13.7 Hz, 1H; -SiCH2Si-), 0.01 (d, JH,H =13.8 Hz, 2H; -SiCH2Si-), À0.10
2
2
(d, JH,H =13.7 Hz, 1H; -SiCH2Si-), À0.13 ppm (d, JH,H =13.8 Hz, 2H;
-SiCH2Si-); 13C{1H} NMR (126 MHz, C6D6, 298 K): d=49.7 (s, -SiOCH3),
Chem. Eur. J. 2015, 21, 12436 – 12448
www.chemeurj.org
12445
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
1
Assignments for all-cis-6: 1H NMR (500 MHz, C6D6, 298 K): d=0.34
(s, 9H; -SiCH3), 0.18 (d, JH,H =14.0 Hz, 3H; -SiCH2Si-), 0.13 ppm (d,
); assignments for cis-trans-8: H NMR (500 MHz, C6D6, 298 K): d=
2.13 (s, 2H; -SiCꢀCH), 2.01 (s, 1H; -SiCꢀCH), 0.35 (s, 6H; -SiCH3),
2
2JH,H =14.0 Hz, -SiCH2Si-); 13C{1H} NMR (126 MHz, C6D6, 298 K): d=
10.1 (s, -SiCH2Si-), 5.2 ppm (s, -SiCH3); 29Si{1H} NMR (99 MHz, C6D6,
298 K): d=26.4 ppm (s, -CH2SiCH2-); assignments for cis-trans-6:
1H NMR (500 MHz, C6D6, 298 K): d=0.47 (d, 2JH,H =14.2 Hz, 1H;
0.34 (s, 3H; -SiCH3), 0.25 (d, JH,H =14.1 Hz, 2H; -SiCH2Si-), 0.01 (d,
2
2JH,H =14.0 Hz, 1H; SiCH2Si), À0.03 (d, 2JH,H =14.0 Hz, 1H; SiCH2Si),
À0.16 ppm (d, JH,H =14.1 Hz, 2H; SiCH2Si); 13C{1H} NMR (126 MHz,
2
C6D6, 298 K): d=94.8 (s, -SiCꢀCH), 94.4 (s, -SiCꢀCH), 90.7 (s, -SiCꢀ
CH), 90.5 (s, SiCꢀCH), 2.3 (s, -SiCH2Si-), 2.1 (s, -SiCH2Si-), 1.9 (s,
SiCH3), 1.4 ppm (s, SiCH3); 29Si{1H} NMR (99 MHz, C6D6, 298 K): d=
À18.7 (s, -SiCH3), À18.5 ppm (s, -SiCH3); GC/EI-MS (all-cis): m/z (%):
231 (100) [M+ÀMe], retention time: 7.3 min; GC/EI-MS (cis-trans):
m/z (%): 231 (100) [M+ÀMe], retention time: 7.0 min; EI-MS: m/z
(%): 245.0 [M+ÀH], 231.0 (100) [M+ÀMe], 93.0, 67.0; FT-IR (KBr)
2
-SiCH2Si-), 0.39 (d, JH,H =14.2 Hz, 2H; -SiCH2Si-), 0.32 (s, 6H; -SiCH3),
0.16 (d, 2JH,H =14.2 Hz, 1H; -SiCH2Si-), 0.12 (d, 2JH,H =14.2 Hz, 2H;
-SiCH2Si-), 0.08 ppm (s, 3H; -SiCH3); 13C{1H} NMR (126 MHz, C6D6,
298 K): d=10.9 (s, -SiCH2Si-), 10.5 (s, -SiCH2Si-), 5.4 (s, -SiCH3),
4.8 ppm (s, -SiCH3); 29Si{1H} NMR (99 MHz, C6D6, 298 K): d=26.1 (s,
-CH2SiCH2-), 25.4 ppm (s, -CH2SiCH2-); GC/EI-MS: m/z (%): 261 (100),
[M+ÀMe], 241 [M+ÀCl], 225 [M+ÀHClÀMe], 169, 147, 133, 113,
93, 79, 63, 43; MS (ESI, negative): m/z (%): 257.0 (100) [MÀCl+O]À;
~
(all-cis): n=3282 (CꢀCÀH), 3265 (CꢀCÀH), 2962, 2925, 2905, 2877,
2035 (CꢀC), 2030 (CꢀC), 1411, 1361, 1349, 1261, 1047, 820, 783,
716, 693, 680, 592, 578 cmÀ1; elemental analysis calcd (%) for
C12H18Si3 (Mr =246.53): C 58.46, H 7.36; found: C 57.92, H 7.48.
~
FT-IR (KBr): n=2963, 2926, 2901, 2872, 2858, 1406, 1349, 1259,
1160, 1050, 951, 831, 772, 748, 719, 633, 621, 610, 530, 508, 486,
466 cmÀ1; elemental analysis calcd (%) for C6H15Si3Cl3 (Mr =277.80):
C 25.94, H 5.44; found: C 25.77, H 5.42.
1,3,5-Tris[bis(pentafluorophenyl)boranylethenyl]-1,3,5-tri-
methyl-1,3,5-trisilacyclohexane (9)
1,3,5-Trimethyl-1,3,5-trisilacyclohexane (7)
Compound 3 (3.29 g, 12.5 mmol) was added dropwise to a suspen-
sion of LiAlH4 (95%, 576 mg, 14.4 mmol) in diethyl ether (80 mL) at
08C. During this addition the suspension became white. The reac-
tion mixture was refluxed for 24 h. All volatiles were passed
through a series of cold traps (À45 and À1988C) under vacuum
and the product was collected in the À458C trap, while diethyl
ether passed to the À1988C trap. Yielded was 1.55 g of a product
mixture of all-cis and cis–trans (8.9 mmol, 71%). The diastereomers
were separated according to a protocol reported by I. Arnason[28]
in a fractional condensation with the help of an U-tube (yield of
Piers’ borane (HB(C6F5)2, 212 mg, 0.6 mmol) and
8 (50 mg,
0.2 mmol) were placed in a flask equipped with a Young greaseless
tap. The mixture was frozen in liquid nitrogen, n-pentane (10 mL)
was condensed onto it and warmed after that to RT. After a few
minutes both solids were dissolved and the reaction was over. All
volatiles were removed under reduced pressure and a solid residue
of
9
remained. Yield: 259.1 mg (0.2 mmol, 100%). 1H NMR
3
(300 MHz, C6D6, 298 K): d=7.64 (d, JH,H =21.0 Hz, 3H; -SiHC=CHB-),
3
7.48 (d, JH,H =21 Hz, 3H; -SiHC=CHB-) 0.23 (s, 9H; -SiCH3), 0.10 (d,
all-cis: 58%); Assignments for all-cis-7: 1H NMR (500 MHz, C6D6,
2JH,H =13.9 Hz, 3H; -SiCH2Si-), À0.06 ppm (d, 2JH,H =13.9 Hz, 3H;
-SiCH2Si-); 11B{1H} NMR (96 MHz, C6D6, 298 K): d=58.5 ppm (brs);
13C{1H} NMR (75 MHz, C6D6, 298 K): d=173.2 (s, (FPh)2BCH=CH-),
298 K): d=4.36 (ttq, 3JH,H =1.3 Hz (He), 3JH,H =3.7 Hz (HMe), JH,H
=
3
3
7.6 Hz (Ha), 3H; -SiH), 0.14 (d, JH,H =3.7 Hz, 9H; -SiCH3), À0.13 (d,
2JH,H =13.6 Hz, 3H; -SiCHeSi-), À0.64 ppm (dd, 2JH,H =13.6 Hz, 3H;
-SiCHaSi-); 13C{1H} NMR (126 MHz, C6D6, 298 K): d=0.4 (s, -SiCH3),
À1.4 ppm (s, -SiCH2Si-); 29Si{1H} NMR (99 MHz, C6D6, 298 K): d=
À14.4 ppm; GC/EI-MS (all-cis): m/z (%): 173 [M+ÀH], 159 (100) [M+
ÀMe], 129, 113, 99, 85, 73, 59, 43, retention time: 4.1 min; FT-IR
150.7 (brs, (FPh)2BCH=CH-), 148.1 (dm, JF,C =250.6 Hz, m-C), 143.8
1
1
1
(dm, JF,C =266.2 Hz, p-C), 137.8 (dm, JF,C =250.2 Hz), 113.6 (brs, i-
C), À0.2 (s, -SiCH3), À1.6 ppm (s, -SiCH2Si-); 19F NMR (282 MHz, C6D6,
298 K): d=À129.05 (m, 2F; o-F), À146.71 (tt, 3JF,F =20.7 Hz, JF,F
=
4
4.8 Hz, 1F; p-F), À161.04 ppm (m, 2F; m-F); 29Si{1H} NMR (60 MHz,
C6D6, 298 K): d=À4.6 ppm; elemental analysis calcd (%) for
C48H21B3F30Si3 (Mr =1284.32): C 44.89, H 1.65; found: C 45.05, H
1.81.
~
(KBr): n=2958, 2916, 2871, 2102 (SiÀH), 1358, 1261, 1253, 1243,
1056, 1044, 1002, 868, 788, 738, 694, 681, 668, 614, 574 cmÀ1; ele-
mental analysis calculated (%) for C6H18Si3 (Mr =174.43): C 41.31, H
10.41; found: C 38.55, H 10.33.
1,3,5-Triethynyl-1,3,5-trimethyl-1,3,5-trisilacyclohexane (8)
1,3,5-Tris(dimethylgallanylethynyl)-1,3,5-trimethyl-1,3,5-trisi-
lacyclohexane (10)
A solution of ethynyl magnesium bromide (70 mL, 0.5m, 35 mmol)
was added dropwise to a solution of compound 8 (2.14 g,
7.7 mmol) in THF (100 mL). The mixture was refluxed for 37 h. After
hydrolysis with a saturated aqueous solution of ammonia chloride,
the organic phase was separated and the aqueous phase was ex-
tracted three times with n-pentane (3”100 mL). The combined or-
ganic phases were washed twice with water (2”100 mL) and dried
over MgSO4. The solvent was removed and the remaining solid
was sublimed at room temperature and 0.003 mbar. Compound 8
was obtained as colourless solid. Yield: 1.79 g (7.3 mmol, 95% dia-
stereomeric mixture, ratio all-cis:cis-trans 60:40). The separation of
the diastereomers was performed by fractional sublimation at
room temperature and 0.003 mbar. Assignments for all-cis-8:
1H NMR (500 MHz, C6D6, 298 K): d=2.15 (s, 3H; -SiCꢀCH), 0.38 (d,
2JH,H =14.1 Hz, 3H; -SiCH2Si-), 0.11 (s, 9H; -SiCH3), À0.20 ppm (d,
2JH,H =14.1 Hz, 3H; -SiCH2Si-); 13C{1H} NMR (126 MHz, C6D6, 298 K):
d=94.5 (s, -SiCꢀCH), 90.1 (s, -SiCꢀCH), 2.2 (s, -SiCH2Si-), 1.8 ppm (s,
-SiCH3); 29Si{1H} NMR (99 MHz, C6D6, 298 K): d=À18.5 (s, -CH2SiCH2-
Compound 8 (70 mg, 0.28 mmol) was dissolved in trimethylgallium
(1.00 mL, 1.13 g, 9.8 mmol) and the reaction solution was stirred at
room temperature for 72 h (or 24 h at 428C). During the reaction
a colourless solid was formed. After the reaction was finished, re-
sidual trimethylgallium was removed by condensation and the re-
maining colourless solid was dried in vacuum. Yield: 152.0 mg
1
(0.26 mmol, 100%); m.p. 978C (decomp); H NMR (500 MHz, C6D6,
2
Et2O, 298 K): d=0.62 (d, JH,H =13.9 Hz, 3H; -SiCH2Si-), 0.32 (s, 9H;
-SiCH3), 0.17 (d, 2JH,H =13.9 Hz, 3H; -SiCH2Si-), À0.04 ppm
(s, -Ga(CH3)2); 13C{1H} NMR (126 MHz, C6D6, Et2O, 298 K): d=115.6
(brs, -SiCꢀC-Ga), 3.8 (s, -SiCH2Si-), 3.1 (s, SiCH3), À4.2 ppm (s,
-Ga(CH3)2); 29Si{1H} NMR (99 MHz, C6D6, Et2O, 298 K): d=À22.3 ppm
~
(-CH2SiCH2-); FT-IR (KBr): n=3436, 3288, 2961, 2920, 2034 (CꢀC),
1359, 1260, 1043, 826, 782, 756, 712, 669, 666, 614, 585, 538 cmÀ1
;
elemental analysis calcd (%) for C18H33Ga3Si3 (Mr =542.88): C 39.82,
H 6.13; found: C 39.19, H 6.32.
Chem. Eur. J. 2015, 21, 12436 – 12448
www.chemeurj.org
12446
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
TION, Fuji, Japan). Their diffusion coefficients were determined ac-
cording to the procedure described in general remarks. Results are
listed in Table 5.
1,3,5-Tris(diethylgallanylethynyl)-1,3,5-trimethyl-1,3,5-trisila-
cyclohexane (11)
Compound 8 (63 mg, 0.26 mmol) was dissolved in triethylgallium
(1.00 mL, 1.07 g, 6.8 mmol) and the reaction solution was stirred at
room temperature for 24 h. During the reaction colourless solid
was formed. After the reaction was finished, triethylgallium was
condensed off and the residual colourless solid was dried in
vacuum. Yield: 163.0 mg (0.26 mmol, 100%); m.p. 788C (decomp);
Crystallographic structure determination
Single crystals suitable for X-ray diffraction measurement were
picked, suspended in a Paratone-N/paraffin oil mixture, mounted
on a glass fibre and transferred onto the goniometer of the diffrac-
tometer. The measurements were carried out with Mo-Ka radiation
(l=0.71073 Š). The structures were solved by direct methods and
refined by full-matrix least-squares cycles (programs: SHELX-97).[48]
Crystallographic details of compounds cis-trans-6, all-cis-7, all-cis-8
3
1H NMR (500 MHz, C6D6, Et2O, 298 K): d=1.35 (t, JH,H =8.1 Hz, 18H;
3
2
CH3CH2Ga-), 0.61 (q, JH,H =8.1 Hz, 12H; CH3CH2Ga-), 0.45 (d, JH,H
=
=
2
14.1 Hz, 3H; -SiCH2Si-), 0.23 (s, 9H; SiCH3), 0.17 ppm (d, JH,H
14.1 Hz, 3H; -SiCH2Si-); 13C{1H} NMR (126 MHz, C6D6, Et2O, 298 K):
d=10.7 (s, CH3CH2Ga-), 6.3 (s, CH3CH2Ga-), 4.1 (s, -SiCH2Si-),
2.5 ppm (s, -SiCH3); 29Si{1H} NMR (99 MHz, C6D6, Et2O, 298 K): d=
~
À21.2 ppm (-CH2SiCH2-); FT-IR (KBr): n=3290, 3223, 2950, 2901,
2865, 2814, 2034 (CꢀC), 1460, 1418, 1375, 1359, 1261, 1192, 1043,
1003, 961, 940, 827, 780, 758, 713, 655, 609, 565, 515 cmÀ1; elemen-
tal analysis calcd (%) for C24H45Ga3Si3 (Mr =627.05): C 45.97, H 7.23;
found: C 44.53, H 7.34.
Table 8. Crystallographic data for compounds cis-trans-6, all-cis-7 and all-
cis-8.
cis-trans-6
all-cis-7
all-cis-8
formula
Mr
C6H15Cl3Si3
277.80
C6H18Si3
174.47
C12H18Si3
246.53
1,3,5-Tris(dimethylindanylethynyl)-1,3,5-trimethyl-1,3,5-trisi-
lacyclohexane (12)
T [K]
crystal size [mm]
100(2)
100.0(2)
100.0(2)
0.30”0.29”0.20 0.38”0.09”0.09 0.27”0.25”0.13
crystal system
space group
a [Š]
b [Š]
c [Š]
monoclinic
P21/n
orthorhombic
Pna21
16.8261(2)
12.7130(2)
5.0042(2)
90
1070.45(3)
4
1.083
monoclinic
P21/n
Compound 8 (50 mg, 0.20 mmol) and trimethylindium (97.4 mg,
0.61 mmol) were placed in toluene (4 mL). The solution received
was heated for four days to 608C. During the reaction a colourless
solid was formed. All volatiles were removed in vacuum and the
colourless solid was washed with toluene (1 mL). The residual col-
ourless solid was dried in vacuum. Yield: 109.0 mg (0.16 mmol,
6.3145(2)
13.3052(2)
15.9629(2)
98.3904(8)
1326.78(3)
4
1.391
0.917
576
3.00 to 30.00
À8ꢃhꢃ8
À18ꢃkꢃ18
À22ꢃlꢃ22
51627
13.7663(2)
12.4516(2)
17.3185(2)
96.615(2)
2948.83(6)
8
1.111
0.293
1056.0
5.76 to 60.07
À19ꢃhꢃ19
À17ꢃkꢃ17
À24ꢃlꢃ24
11 0551
8627
7872
0.030
8627/0/277
b [8]
V [Š3]
1
Z
80%); m.p. 1818C (decomp). H NMR (500 MHz, [D8]THF, 298 K): d=
1calcd [gcmÀ1
]
2
0.13 (s, 9H; SiCH3), 0.11 (d, JH,H =13.9 Hz, 3H; -SiCH2Si-), À0.06 (d,
m [mmÀ1
F(000)
]
0.377
384.0
2JH,H =13.9 Hz, 3H; -SiCH2Si-), À0.44 ppm (s, 18H; -In(CH3)2); 13C{1H}
NMR (126 MHz, [D8]THF, 298 K): d=135.6 (s, -SiCꢀCIn-), 115.9
(s, -SiCꢀCIn-), 5.7 (s, -SiCH2Si-), 3.3 (s, -SiCH3), À7.7 ppm (s,
-In(CH3)2); 29Si{1H} NMR (99 MHz, [D8]THF, 298 K): d=À24.2 ppm (s,
2q range [8]
index range
4.02 to 52.73
À21ꢃhꢃ21
À16ꢃkꢃ16
À6ꢃlꢃ6
37800
3736
3359
0.059
3736/1/155
~
-CH2SiCH2-); FT-IR (KBr): n=3287, 3282, 3272, 3265, 2966, 2922,
2034 (CꢀC), 2029 (CꢀC), 1407, 1359, 1261, 1160, 1041, 1000, 824,
780, 756, 709, 681, 650, 604, 582, 522 cmÀ1; elemental analysis
calcd (%) for C18H33In3Si3 (Mr =678.17): C 31.88, H 4.90; found: C
32.39, H 5.06.
reflns collected
unique reflns
observed reflns (2s) 3483
3860
Rint
0.041
3860/0/112
data/restraints/pa-
rameters
GoF (F2)
Diffusion NMR and titration experiments
1.048
1.006
1.096
R1, wR2 [I> 2s(I)]
0.0238, 0.0597 0.0319, 0.0809 0.0258, 0.0749
0.0277, 0.0613 0.0363, 0.0831 0.0287, 0.0767
1
General remarks: Diffusion H NMR measurements were performed
R1, wR2 (all data)
on a Bruker Avance 600 NMR spectrometer in [D8]toluene (294 K,
d=7.09, 7.00, 6.98, 2.09 ppm) using the LED sequence with bipolar
gradients (ledbpgp2s). The NMR tube with the sample was allowed
to equilibrate for at 3 h within the probe/magnet prior to data re-
cording. The duration of the gradients was incremented linearly in
16 steps. The maximum length of the gradient pulse d and the dif-
fusion time were set to d=3.80 ms and D=79.95 ms, respectively,
for all experiments. The diffusion coefficients have been calculated
by using the relaxation module of the Bruker software TOPSPIN.
À3
D1(max/min) [eŠ
]
0.32/À0.29
0.34/À0.23
0.39/À0.26
reported in Table 8. CCDC 1059729 (cis-trans-6), 1059730 (all-cis-7)
and 1059731 (all-cis-8) contain the supplementary crystallographic
data. These data can be obtained free of charge by The Cambridge
Crystallographic Data Centre
Diffusion coefficients of host 11, TMTAC and reference compounds:
Small amounts of the reference compounds 13, 14, 15, 16, host 11
and TMTAC (1,3,5-trimethyl-1,3,5-triazacyclohexane, guest) were
dissolved in [D8]toluene. Their diffusion coefficients were deter-
mined according to the procedure described in the general re-
marks. Results are listed in Table 4.
Acknowledgements
We thank Klaus-Peter Mester and Gerd Lipinski for recording
NMR spectra, Brigitte Michel for performing the elemental
analyses, Mian Qi and Julia Wegner for help with a HPLC sepa-
ration. We gratefully acknowledge the financial support of the
Fonds der Chemischen Industrie (stipend for E.W.) and the
Deutsche Forschungsgemeinschaft.
Conversion of 11 with TMTAC: Compound 11 (2 mg, 3.2 mmol) was
dissolved in [D8]toluene (0.6 mL); small amounts solution of the
TMTAC (TMTAC: 35 mg, 270.9 mmol in 3.669 g [D8]toluene;
0.064 mmolmLÀ1) were added using a 50 mL syringe (ITO CORPORA-
Chem. Eur. J. 2015, 21, 12436 – 12448
www.chemeurj.org
12447
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[23] W. A. Kriner, J. Org. Chem. 1964, 29, 1601–1606.
[24] W. A. Kriner, US Patent US3178392A, 1962.
Keywords: NMR spectroscopy · hydroboration · Lewis acids ·
metalation · trisilacyclohexane
[25] a) R. Tacke, J. Sperlich, C. Strohmann, G. Mattern, Chem. Ber. 1991, 124,
1491–1496; b) R. Tacke, J. Pikies, H. Linoh, R. Rohr-Aehle, S. Gçnne, Lie-
bigs Ann. Chem. 1987, 51–57.
[26] S. Bienz, A. Chapeaurouge, Helv. Chim. Acta 1991, 74, 1477–1488.
[27] N. Auner, J. Grobe, J. Organomet. Chem. 1980, 188, 151–177.
[28] I. Arnason, Teilbromierte Carbosilane, Dissertation, Karlsruhe, 1973.
[29] G. Fritz, K. Gompper, H. Wilhelm, Z. Anorg. Allg. Chem. 1981, 478, 20–
38.
[1] C. H. Park, H. E. Simmons, J. Am. Chem. Soc. 1968, 90, 2431–2432.
[2] F. Vçgtle, E. Blasius, Host–Guest Complex Chemistry I, Springer, Heidel-
berg, 1981.
[3] F. P. Schmidtchen, Angew. Chem. Int. Ed. Engl. 1977, 16, 720–721;
Angew. Chem. 1977, 89, 751–752.
[4] D. Brondani, F. H. CarrØ, R. J. P. Corriu, J. J. E. Moreau, M. Wong Chi Man,
Angew. Chem. Int. Ed. Engl. 1996, 35, 324–326; Angew. Chem. 1996, 108,
349–352.
[30] M. Oestreich, U. K. Schmid, G. Auer, M. Keller, Synthesis 2003, 2725–
2739.
[31] S. H. Hong, S. I. Hyun, I. N. Jung, W.-S. Han, M.-H. Kim, H. Yun, S.-W.
Nam, S. O. Kang, Organometallics 2010, 29, 687–691.
[32] D. J. Parks, W. E. Piers, G. P. A. Yap, Organometallics 1998, 17, 5492–
5503.
[33] D. J. Parks, R. E. v. H. Spence, W. E. Piers, Angew. Chem. Int. Ed. Engl.
1995, 34, 809–811; Angew. Chem. 1995, 107, 895–897.
[5] a) M. Newcomb, J. H. Horner, M. T. Blanda, J. Am. Chem. Soc. 1987, 109,
7878–7879; b) K. Jurkschat, H. G. Kuivila, S. Liu, J. A. Zubieta, Organome-
tallics 1989, 8, 2755–2759; c) M. Schulte, M. Schürmann, K. Jurkschat,
Chem. Eur. J. 2001, 7, 347–355; d) M. Schulte, G. Gabriele, M. Schür-
mann, K. Jurkschat, A. Duthie, D. Dakternieks, Organometallics 2003, 22,
328–336.
[34] D. Winkelhaus, B. Neumann, H.-G. Stammler, N. W. Mitzel, Dalton Trans.
2012, 41, 9143–9150.
[35] S. J. Schauer, C. L. Watkins, L. K. Krannich, R. B. Gala, E. M. Gundy, C. B.
Lagrone, Polyhedron 1995, 14, 3505–3514.
[6] a) A. L. Chistyakov, I. V. Stankevich, N. P. Gambaryan, I. A. Tikhonova, V. B.
Shur, Russ. Chem. Bull. 1995, 44, 997–1004; b) A. L. Chistyakov, I. V. Stan-
kevich, N. P. Gambaryan, Y. Struchkov, A. I. Yanovsky, I. A. Tikhonova,
´
V. B. Shur, J. Organomet. Chem. 1997, 536–537, 413–424; c) D. Grdenic,
[36] a) Y. Cohen, L. Avram, L. Frish, Angew. Chem. Int. Ed. 2005, 44, 520–554;
Angew. Chem. 2005, 117, 524–560; b) C. Zuccaccia, G. Bellachioma, G.
Cardaci, A. Macchioni, Organometallics 2000, 19, 4663–4665; c) C. Zuc-
caccia, N. G. Stahl, A. Macchioni, M.-C. Chen, J. A. Roberts, T. J. Marks, J.
Am. Chem. Soc. 2004, 126, 1448–1464.
Chem. Ber. 1959, 92, 231–234; d) J. D. Wuest, B. Zacharie, J. Am. Chem.
Soc. 1987, 109, 4714–4715; e) X. Yang, C. B. Knobler, M. F. Hawthorne,
Angew. Chem. Int. Ed. Engl. 1991, 30, 1507–1508; Angew. Chem. 1991,
103, 1519–1521.
[7] J. B. King, M. Tsunoda, F. P. Gabba¨ı, Organometallics 2002, 21, 4201–
4205.
[37] L. Frish, S. E. Matthews, V. Bçhmer, Y. Cohen, J. Chem. Soc. Perkin Trans.
2 1999, 669–672.
[8] P. Sartori, A. Golloch, Chem. Ber. 1968, 101, 2004–2009.
[9] V. B. Shur, I. A. Tikhonova, A. I. Yanovsky, Y. Struchkov, P. V. Petrovskii, S.
Panov, G. G. Furin, M. E. Vol’pin, J. Organomet. Chem. 1991, 418, C29.
[10] S. P. Fisher, E. W. Reinheimer, J. Chem. Crystallogr. 2013, 43, 478–483.
[11] a) T. J. Taylor, F. P. Gabba¨ı, Organometallics 2006, 25, 2143–2147; b) T. J.
Taylor, V. I. Bakhmutov, F. P. Gabba¨ı, Angew. Chem. Int. Ed. 2006, 45,
7030–7033; Angew. Chem. 2006, 118, 7188–7191.
[38] a) W. Li, H. Chung, C. Daeffler, J. A. Johnson, R. H. Grubbs, Macromole-
cules 2012, 45, 9595–9603; b) W. Li, G. Kagan, H. Yang, C. Cai, R.
Hopson, W. Dai, D. A. Sweigart, P. G. Williard, Organometallics 2010, 29,
1309–1311; c) D. Stalke, R. Neufeld, Chem. Sci. 2015, 6, 3354–3364;
d) H. Subramanian, C. P. Jasperse, M. P. Sibi, Org. Lett. 2015, 17, 1429–
1432.
[39] R. G. Parr, W. Yang, Density Functional Theory of Atoms and Molecules,
Oxford University Press, Oxford, 1989.
[12] a) W. Uhl, H. R. Bock, F. Breher, M. Claesener, S. Haddadpour, B. Jasper,
A. Hepp, Organometallics 2007, 26, 2363–2369; b) W. Uhl, H. R. Bock, M.
Claesener, M. Layh, I. Tiesmeyer, E.-U. Würthwein, Chem. Eur. J. 2008, 14,
11557–11564; c) W. Uhl, M. Claesener, S. Haddadpour, B. Jasper, A.
Hepp, Dalton Trans. 2007, 417–423; d) W. Uhl, M. Claesener, Inorg.
Chem. 2008, 47, 4463–4470; e) W. Uhl, A. Hepp, H. Westenberg, S.
Zemke, E.-U. Würthwein, J. Hellmann, Organometallics 2010, 29, 1406–
1412.
[13] J. D. Beckwith, M. Tschinkl, A. Picot, M. Tsunoda, R. Bachman, F. P.
Gabba¨ı, Organometallics 2001, 20, 3169–3174.
[14] H. Schmidbaur, H.-J. Öller, D. L. Wilkinson, B. Huber, G. Müller, Chem. Ber.
1989, 122, 31–36.
[15] a) C. Jiang, O. Blacque, H. Berke, Chem. Commun. 2009, 5518–5520;
b) H. E. Katz, J. Am. Chem. Soc. 1985, 107, 1420–1421; c) H. E. Katz, J.
Org. Chem. 1985, 50, 5027–5032; d) M. Mela¨ımi, S. SolØ, C.-W. Chiu, H.
Wang, F. P. Gabba¨ı, Inorg. Chem. 2006, 45, 8136–8143; e) S. SolØ, F. P.
Gabba¨ı, Chem. Commun. 2004, 1284–1285; f) H. Zhao, F. P. Gabbai, Or-
ganometallics 2012, 31, 2327–2335.
[16] F. P. Gabba¨ı, A. Schier, J. Riede, A. Sladek, H. W. Gçrlitzer, Inorg. Chem.
1997, 36, 5694–5698.
[17] J. D. Hoefelmeyer, D. L. Brode, F. P. Gabba¨ı, Organometallics 2001, 20,
5653–5657.
[40] C. Møller, M. S. Plesset, Phys. Rev. 1934, 46, 618–622.
[41] Gaussian03, 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. Tomasi, 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. Hratchian, J. B.
Cross, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev,
A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Moroku-
ma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dap-
prich, A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, 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, 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, Inc., Wallingford CT, 2004..
[42] R. J. F. Berger, M. Hoffmann, S. A. Hayes, N. W. Mitzel, Z. Naturforsch.
b 2009, 64, 1259–1268.
[43] Yu. V. Vishnevskiy, J. Mol. Struct. 2007, 833, 30–41.
[44] Yu. V. Vishnevskiy, J. Mol. Struct. 2007, 871, 24–32.
[45] Y. Morino, T. Iijima, Bull. Chem. Soc. Jpn. 1962, 35, 1661–1667.
[46] Yu. V. Vishnevskiy, UNEX Version 1.6.0 beta.
[47] a) V. A. Sipachev, J. Mol. Struct. 1985, 22, 143–151; b) V. A. Sipachev, J.
Mol. Struct. 2001, 567–568, 67–72; c) V. A. Sipachev, J. Mol. Struct. 2004,
693, 235–240.
[18] H. E. Katz, J. Org. Chem. 1989, 54, 2179–2183.
[19] a) J. Chmiel, B. Neumann, H.-G. Stammler, N. W. Mitzel, Chem. Eur. J.
2010, 16, 11906–11914; b) J.-H. Lamm, J. Horstmann, J. H. Nissen, J.-H.
Weddeling, B. Neumann, H.-G. Stammler, N. W. Mitzel, Eur. J. Inorg.
Chem. 2014, 4294–4301.
[20] J.-H. Lamm, P. Niermeier, A. Mix, J. Chmiel, B. Neumann, H.-G. Stammler,
N. W. Mitzel, Angew. Chem. Int. Ed. 2014, 53, 7938–7942; Angew. Chem.
2014, 126, 8072–8076.
[48] G. M. Sheldrick, Acta Crystallogr. Sect. A 2008, 64, 112–122.
[21] I. Arnason, G. K. Thorarinsson, E. Matern, J. Mol. Struct. 1998, 454, 91–
102.
Received: April 29, 2015
[22] I. Arnason, E. Matern, J. Mol. Struct. 2001, 544, 61–68.
Published online on July 21, 2015
Chem. Eur. J. 2015, 21, 12436 – 12448
www.chemeurj.org
12448
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim