The synthesis, crystal structures and NMR spectroscopic investigation of 3,7,10-trimethylsilatranes and carbasilatranes

Article abstract of DOI:10.1016/j.jorganchem.2008.09.055

The first representatives of aryloxy-trimethylsilatranes, C₆H₅OSi[OCH(CH₃)CH₂] ₃N, of aryloxy-carbasilatranes, C₆H₅OSi(CH₂CH₂CH₂ )(OCH₂

Full text of DOI:10.1016/j.jorganchem.2008.09.055

Journal of Organometallic Chemistry 694 (2009) 14–20
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
The synthesis, crystal structures and NMR spectroscopic investigation
of 3,7,10-trimethylsilatranes and carbasilatranes
b
b
b
c
Ilona Kovács a, , Eberhard Matern , Ewald Sattler , Christopher E. Anson , László Párkányi
*
a Budapest University of Technology and Economics, Department of Inorganic and Analytical Chemistry, H-1111, Szt. Gellér tér 4, Budapest, Hungary
b Universität Karlsruhe (TH), Institut für Anorganische Chemie, D-76128, Engesserstr. 15, Karlsruhe, Germany
c Institute of Chemistry, Chemical Research Center, Hungarian Academy of Sciences, H-1025, Pusztaszeri út 59-67, Budapest, Hungary
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 4 July 2008
Accepted 29 September 2008
Available online 4 October 2008
The first representatives of aryloxy-trimethylsilatranes, C6H5OSi[OCH(CH3)CH2]3N, of aryloxy-carbasi-
latranes, C6H5OSi(CH2CH2CH2)(OCH2CH2)2N, and number of novel 3,7,10-trimethylsilatranes
a
(RSi[OCH(CH3)CH2]3N) and carbasilatranes (RSi(CH2CH2CH2)(OCH2CH2)2N) have been prepared and char-
acterised, both structurally and by NMR spectroscopy. The influence of various Si substituents (R = alkyl,
aryl, alkoxy and aryloxy) as well as the influence of the substitution in the skeleton of the trimethylsilatr-
anes and carbasilatranes on the chemical shift in the 29Si, 1H and 13C NMR spectra has been investigated.
The crystal structures of both diastereomers of phenyl-3,7,10-trimethylsilatrane and of the symmetrical
isomer of p-tolyl-3,7,10-trimethylsilatrane were determined by X-ray diffraction after separation of the
diastereomers by means of HPLC. All three crystal structures are centrosymmetric.
Keywords:
Trimethylsilatranes
Carbasilatranes
NMR spectroscopy
X-ray scattering
Ó 2008 Elsevier B.V. All rights reserved.
1. Introduction
13C NMR spectra are discussed. Additionally, the single crystal
structure investigations of the diastereomers of the phenyl- and
The silatranes RSi(OCH2CH2)3N adopt a special position among
the hypervalent neutral penta- and hexa-coordinated silicon com-
pounds as a result of the characteristic structure of their bonding
[1]. The central silicon atom is linked by means of three triatomic
bridges to a nitrogen atom, the free electron pair of which is en-
gaged in a relatively weak, easily distortable, coordinative bond
to the silicon atom, dependent on the conditions of aggregation
[2]. The length and strength of this bond also depends on the axial
substituent R on the silicon atom and upon the properties of the
bridges, i.e. on the electronic influence of the neighbouring O, N,
p-tolyl-3,7,10-trimethylsilatranes are reported. It is of special value
that we can present data for several compounds as well in the solid
as in the liquid state, and such can consider their properties from a
more general point of view.
2. Results and discussion
2.1. Synthesis of investigated compounds
C atoms in the
a-positions defining the equatorial plane, on the
The compounds investigated in this study are presented on
Fig. 1. The alkyl-trimethylsilatranes 1a–5a and the aryl-trimethyl-
silatranes 6a, 7a were prepared from silicon-substituted trimeth-
oxysilanes and triisopropanolamine by a modification of the
method of Frey [7], the alkoxy-trimethylsilatranes 8a, 9a were syn-
thesised from tetraethoxysilane, triisopropanolamine and the
appropriate alcohol using a modification of the method of Voron-
kov and Zelchan [8]. The carbasilatranes 1b and 6b–8b were ob-
number of atoms present in the bridges, and possibly also on steric
influences [3]. This Si N interaction seems to be responsible for
many important chemical [4] and biological [5] properties of the
silatranes. Some aspects of this dative bond between Si and N are
still not fully understood, therefore the synthesis of new silatranes
as well as the comprehensive characterisation of as many as possi-
ble derivatives is highly desirable. Recently, silatranes have found
some applications in material science [6].
tained from silicon-substituted (c-chloropropyl)dimethoxysilanes
In the present paper, the preparation and characterisation of
several 3,7,10-trimethylsilatranes and carbasilatranes are reported,
and the influences of various substituents (R = alkyl, aryl, alkoxy,
aryloxy) and the substitution in the skeleton of the trimethylsilatr-
anes and carbasilatranes on the chemical shift in the 29Si, 1H and
and diethanolamine (in the presence of triethylamine as HCl accep-
tor) using improved literature methods [9]. To our knowledge no
aryloxy-trimethylsilatranes and aryloxy-carbasilatranes have been
previously described. We report here the syntheses of the first rep-
resentatives of these compounds. The synthesis of phenoxy-tri-
methylsilatrane 10a was not possible using the same route as for
the alkoxy-trimethylsilatranes, but the compound could be pre-
pared Eq. (1):
* Corresponding author. Tel.: +36 1 463 1294; fax: +36 1 463 3642.
E-mail address: ikovacs@mail.bme.hu (I. Kovács).
0022-328X/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2008.09.055

I. Kovács et al. / Journal of Organometallic Chemistry 694 (2009) 14–20
15
Table 1
Crystal data and structure refinement of 6asym, 6aasym and 7asym
.
Data
6asym
6aasym
7asym
Empirical formula
Formula weight
Temperature, K
Radiation,
wavelength, Å
Crystal system
Space group
a(Å)
C15H23NO3Si
293.43
100(2)
C15H23NO3Si
293.43
100(2)
C16H25NO3Si
307.46
293(2)
Mo Ka
,
Mo Ka
,
Cu Ka, k = 1.5418
k = 0.71073
Monoclinic
P21/c
10.9566(5)
16.9950(8)
16.6884(8)
97.491(1)
3081.0(3)
8
k = 0.71073
Monoclinic
P21/n
18.3613(19)
9.7768(10)
18.6127(18)
113.059(2)
3074.3(5)
8
Monoclinic
P21/c
11.824(1)
12.161(1)
13.116(1)
116.55(1)
1687.1(3)
4
R
R’= CH3
X = O
R’= H
X = CH2
b(Å)
c(Å)
b(°)
Volume, Å3
Z
CH3
Cl(CH2)3
HS(CH2)3
1a [4a]
2a
3a
1b [4a]
Dcalc, Mg mꢀ3
1.265
0.159
1.268
0.160
1.210
1.306
l
, mmꢀ1
CH2=C(CH3)COO(CH2)3
H2N(CH2)2NH(CH2)3
C6H5
p-CH3C6H4
CH3O
4a
5a
6a [4a]
7a [11]
8a [10]
9a
F(000)
1264
Colourless
Chunk
1264
Colourless
Needle
664
Colourless
Prism
Crystal colour
Crystal description
Crystal size, mm
2hmax (°)
Reflections
collected
6b [4a]
7b [12]
8b [4a]
0.47 ꢁ 0.44 ꢁ 0.39 0.22 ꢁ 0.08 ꢁ 0.06 0.50 ꢁ 0.40 ꢁ 0.15
56.02
15327
54.04
11347
151.50
3693
n-C8H17O
C6H5O
Rint
0.0254
6935
0.0861
6144
0.0287
3417
10a
10b
Unique reflections
Reflections I > 2
Final wR2 (all data)
S (all data)
r
(I) 5760
2166
0.0857
0.875
3100
0.1524
1.041
Fig. 1. The investigated 3,7,10-trimethylsilatranes and carbasilatranes.(See above-
mentioned references for further information[10]).
0.1418
1.024
Final R1 [I > 2
r
(I)]
0.0521
0.49/ꢀ0.56
0.0591
0.73/ꢀ0.42
0.0502
0.39/ꢀ0.37
Largest difference
in peak/hole,
e Åꢀ3
C2H5OSi[OCH(CH3)CH2]3N + C6H5OH
C6H5OSi[OCH(CH3)CH2]3N + C2H5OH
ð1Þ
The phenoxy-carbasilatrane 10b was synthesised in the follow-
ing manner:
2.2. Crystal structure determinations
Cl(CH2)3Si(OCH3)3 + C6H5OH + HN(CH2CH2OH)2 + (C2H5)3N
;
Crystallographic data for 6asym, 6aasym and 7asym are summa-
rised in Table 1, the molecular diagrams are depicted in Fig. 3
and selected bond distances and angles are listed in Table 2. In
the crystal structures of both 6asym and 6aasym there are two inde-
pendent molecules in the asymmetric unit.
C6H5OSi(CH2CH2CH2)(OCH2CH2)2N + 3 CH3OH + (C2H5)3N.HCl
ð2Þ
The new compounds were characterised by 1H, 13C, 29Si NMR,
IR, MS and elemental analyses. The syntheses of the 3,7,10-tri-
methylsilatranes lead to mixtures of diastereomers (Fig. 2).
The symmetrical diastereomer has idealised C3 molecular
symmetry, the asymmetric only C1. The theoretical isomeric ratio
is sym:asym = 1:3. The diastereomers of compounds 6a and 7a
could be separated by means of HPLC chromatography; the minor
diastereoisomers had the shorter retention time. Single crystals
suitable for X-ray analysis could be obtained from both stereoiso-
mers of 6a and from the minor stereoisomer of 7a.
Silatranes with an unsubstituted atrane skeleton usually form
ordered structures. When they are disordered, the carbon atoms
a
to the nitrogen appear in split positions. These partially filled
atom sites are positioned on both sides of the planes formed by
Si, O, C (b to N) and N. Methyl substitution at the C3, C7, C10 atoms
dramatically changes this situation as the disorder mostly affects
these methyl substituted carbons (and the methyl moiety) and
these now appear on both sides of the Si, O, C (a to N) and N chains.
Fig. 2. Symmetrical (right) and asymmetrical (left) diastereomers of silatranes.

16
I. Kovács et al. / Journal of Organometallic Chemistry 694 (2009) 14–20
Fig. 3. Molecular structures of 6asym (a), 6aasym (b) and 7asym (c). In (a) and (b), Molecule 1 from the asymmetric unit is shown; in (a) and (c) atoms of the minor disorder
components are omitted for clarity.
Table 2
suggesting that disorder was present but unresolved. The Si N
Selected bond lengths (Å) and angles (°).
bonds at 2.167(3) and 2.182(3) Å are longer than the values in
the present low temperature study, 2.155(1) and 2.164(1) Å,
respectively.
Compound
6asym/1
6asym/2
6aasym/1
6aasym/2
7asym
Si1 N5
2.155(1)
1.662(1)
1.657(1)
1.665(1)
1.902(2)
118.06(7)
120.29(7)
117.64(7)
96.67(6)
98.36(6)
95.07(6)
83.43(6)
83.66(5)
82.85(6)
177.62(6)
2.164(1)
1.663(1)
1.656(1)
1.663(1)
1.893(2)
117.82(6)
118.07(8)
119.76(8)
97.52(7)
97.46(7)
95.94(6)
82.78(6)
83.57(6)
82.74(6)
178.62(6)
2.144(3)
1.657(3)
1.664(3)
1.673(3)
1.895(4)
118.8(1)
117.6(2)
119.7(2)
98.1(2)
96.0(2)
95.8(2)
83.8(1)
82.8(2)
2.182(3)
1.656(3)
1.659(3)
1.659(3)
1.889(4)
118.2(2)
118.7(1)
118.7(2)
97.4(2)
96.1(2)
97.5(2)
83.1(1)
82.9(1)
2.223(2)
1.648(2)
1.655(1)
1.655(1)
1.887(2)
119.36(9)
117.90(9)
116.96(9)
98.25(8)
97.73(7)
98.19(7)
82.04(7)
81.75(7)
82.05(6)
179.48(7)
The crystal structure of the major stereoisomeric component of
7aasym was published earlier [11]. The Si N distance, 2.236(3) Å,
is longer by 0.013 Å than in 7asym, possibly due to a more strained
atrane skeleton, but crystal packing effects may be the primary
cause (see below).
In comparison with the unsubstituted silatranes the methyl
substitution on the C atoms 3, 7 and 10 of the skeleton in the p-to-
lyl derivatives causes a considerable lengthening of the Si–N dis-
tance (7c: 2.169 Å [11]). This effect cannot be observed for the
Si1–O2
Si1–O8
Si1–O9
Si1–C12
O8–Si1–O2
O2–Si1–O9
O8–Si1–O9
O2–Si1–C12
O8–Si1–C12
O9–Si1–C12
O2–Si1 N5
O8–Si1 N5
O9–Si1 N5
C12–Si1 N5
phenyl derivatives (6c a: 2.193 Å [13a], 6c b: 2.156 Å [13b], 6c c:
2.132 Å [13c]). More comprehensive comparison is not possible be-
cause not sufficient structural data are known.
83.6(1)
178.1(2)
83.0(1)
179.0(2)
It has been established earlier [14] that the displacement of the
silicon atom out of the plane of its three equatorial oxygens (
and the deviation of the nitrogen from the plane of its carbon sub-
stituents (
DSi)
Both molecules in the asymmetric unit of 6asym are disordered, the
site occupation factors (s.o.f.) are 0.74:0.26 for Molecule 1 and
0.63:0.37 for Molecule 2. One molecule in the asymmetric unit of
6aasym is ordered, while the second is disordered with a s.o.f. ratio
of 0.70:0.30. There is only one molecule in the asymmetric unit of
7asym. The atrane skeleton is disordered and the disordered atoms
have a s.o.f. ratio of 0.665:0.335.
D
N) are dependent on the length of the Si N dative
bond. Short dative bonds were found to result in more coplanar
SiO3 moieties while, conversely, the nitrogen becomes less tetrahe-
dral for longer Si N bonds. Here, this rule holds approximately
for
DSi, but no such trend can be observed for DN (Fig. 4).
All three crystal structures are centrosymmetric, and half the
sites in the unit cell should contain one enantiomer, with the other
sites containing the opposite enantiomer. The observed disorder
corresponds to a small proportion of the ‘‘wrong” enantiomer in
a given site. In all three cases, the two enantiomers have a very
similar external shape, with the methyl substituents in similar
positions on the periphery of the molecule, so that such packing er-
rors do not greatly affect the neighbouring molecules, and such
disorder is therefore not unlikely.
The Si N transannular bonds vary over the range 2.14–2.22 Å.
In particular, these bond lengths differ significantly between the
two independent molecules in the structures of both compounds
6asym and 6aasym. The silatrane skeleton undergoes rapid confor-
mational changes (flapping of the O–C–C–N chains) in solution,
and it appears that on crystallization the demands of the crystal
packing results in particular ‘‘snapshots” of this motion being fro-
zen out. This may account for the different molecular geometries. A
room temperature crystal structure of 6asym [12] showed no
apparent disorder, though the accuracy of the data was limited
and some of the atomic displacement parameters were rather high,
Fig. 4. The distance of Si from the plane of the oxygen atoms (
best fit) and the distance of N from the plane of C4, C6 and C11 (
function of the Si N approach (Å) in 6asym, 6aasym and 7asym (see Table 2).
D
Si, Å; circles, line of
D
N, Å; +) as a

I. Kovács et al. / Journal of Organometallic Chemistry 694 (2009) 14–20
17
2.3. NMR spectroscopic studies
2.3.2. 1H NMR spectra
The 1H NMR spectroscopical investigations of a series of silatr-
anes of the type RSi(OCH2CH2)3N have shown that the signals of
the NCH2 and OCH2 protons of the silatrane cage each fall in a small
range of 0.25 ppm, and that with increasing electronegativity of
the substituents a low field shift can be observed [16]. As demon-
strated by thorough structural investigations of silatranes [1a], the
increasing electronegativity of the axial substituent R leads to a
shortening of the Si–N bond and thus to an increasing transannular
donor–acceptor interaction between the N and the Si atom.
The introduction of methyl groups in the positions 3, 7 and 10 of
the silatrane cage causes the loss of the C3 symmetry in case of the
asymmetrical diastereomer, resulting in more complex 1H NMR
spectra. The 1H NMR spectra of the trimethylsilatranes 1a–10a
all show the same pattern, and R substituents on Si in the axial po-
sition of the silatrane framework have just a minor influence. In all
spectra, the signals from the protons of the silatrane cage are well
separated, which allows for a reliable integration with respect to
the group intensities. In all cases, mixtures of the symmetrical
and the asymmetrical diastereomers are present (Fig. 2). For the
symmetrical structure, only one spectrum would be expected for
all three bridges, whereas each bridge of the asymmetrical isomer
will give rise to a distinct spectrum. Indeed, four partial spectra are
observed which overlap only to some extent. These all have nearly
the same intensity, which gives a ratio of 1:3 for the symmetrical
to the asymmetrical isomer. By means of 1H COSY spectra, and
using the 1H spectral data from the separated isomers of 6a and
7a, signals could be assigned both to the respective diastereomers
and also to the different protons within the silatrane bridges. By
means of an iterative optimization [17] the chemical shifts and
the coupling constants for 6a in Table 4 were obtained.
2.3.1. 29Si NMR spectra
Structural data [1a] show that electronegative substituents on
Si cause a shortening of the Si–N distance in silatranes. However,
variations in the bridges of the silatrane skeleton cause stronger
changes in the geometry of the heterocycle and in the Si–N
distance than those of the electronic or steric properties of the sub-
stituent in axial position of the silatrane skeleton. As chemical
shifts react sensitively to changes in the geometry in the neigh-
bourhood of the nucleus being measured, changes of the Si–N dis-
tances should be reflected in changes of the value of d29Si.
Additionally, all silatranes should show a strong high field shift
of d29Si due to their nearly trigonal bipyramidal pentacoordinated
Si atoms as compared to the correspondingly substituted trieth-
oxysilanes [15].
The values of the 29Si chemical shifts of the 3,7,10-trimethylsil-
atranes, carbasilatranes and some sil-atranes with unsubstituted
skeletons are compared in Table 3. The signals of the compounds
2a–10a are shifted to high field compared to the methyl derivative
1a. This corresponds to the expected high-field shift of the 29Si res-
onances with increasing electronegativity of the substituents. If the
substituent is linked via an oxygen atom to Si, as in 8a–10a, a
strong high field shift results, and the strongest effect is observed
for R = C6H5O. Comparing the 3,7,10-trimethylsilatranes with
unsubstituted silatranes shows that there is no uniform trend.
The dependence of the chemical shift values from the polarity of
the solvent has not been systematically investigated. In any case,
it is small and did not show an uniform trend when changing from
CDCl3 to C6D6.
The 29Si chemical shifts for the carbasilatranes 1b, 6b–8b and
10b show a dependency comparable to that in the series of the tri-
methylsilatranes. The exchange of an equatorial oxygen against a
methylene group results in the expected strong low field shift.
The very low variation in the 29Si chemical shifts strongly sug-
gests that in solution the silatranes considered here all have rather
similar Si–N distances. It must therefore be the crystal packing
which is of primary importance in determining the conformation
of the three organic bridges between the Si and N atoms, and thus
the Si–N distance; therefore, this distance is not intrinsically im-
posed by the substitution pattern. The Si–N interaction can thus
be considered highly plastic, with the bond length varying readily
in response to different crystal packing environments. This is par-
ticularly apparent when comparing the significantly different Si–N
distances for the crystallographically-independent molecules in
the structures of either 6asym or 6aasym; in these cases the mole-
cules are chemically identical, and the differences in Si–N can only
result from packing effects.
The 1H NMR spectrum of a carbasilatrane consists of two inde-
pendent parts, one resulting from the –(CH2)3– bridge and the other
from the two –O(CH2)2– bridges of the cage, which are of
AA0MM0XX0 and ABXY types, respectively. The apparent CS mirror
symmetry (with the atoms Si–C–C–C–N in the mirror plane) is
not caused by a rigid molecular skeleton but by the high flexibility
of the three bridges in solution at room temperature; only the effec-
tive symmetry for the ”average” of the fast interchanging conforma-
tions is observed. This is mostly due to the fact that non-substituted
five-membered envelope rings have an even lower energy thresh-
old for inversion than the analogous six-membered rings, that no
sterically-demanding substituents are enforcing certain conforma-
tions, that no rigid building blocks such as carbonyl groups are pres-
ent, and that the substituent on the Si atom can rapidly rotate about
the Si–C bond and therefore does not cause a lowering of the
symmetry. Due to their different chemical environments, with the
–(CH2)3– bridge on one side and the second –O(CH2)2– bridge on
the other, each of the geminal H atoms of the –O(CH2)2– bridge
has a different average value for its chemical shift in spite of the
Table 3
Table 4
29Si NMR chemical shifts [ppm] of the silatranes (1–10) a, b, c (solvent CDCl3).
1H NMR data of C6H5Si[OCH3(CH34)CH21,2]3N 6a (solvent CDCl3).
R
a
b
c
sym.
asym.
sym.
asym.
d1
d2
d3
d4
J1,2 (Hz)
J1,3 (Hz)
J2,3 (Hz)
J3,4 (Hz)
C6H5d
2.842
2.313
4.012
1.235
3.001
2.727
4.254
1.317
ꢀ13.26
6.56
3.069
2.405
4.231a
1.221b
ꢀ12.28
4.06
2.901
2.385
4.072
1.259b
ꢀ12.37
4.05
1
2
3
4
5
6
7
8
9
10
CH3
Cl(CH2)3
HS(CH2)3
CH2@C(CH3)COO(CH2)3
H2NCH2CH2NH(CH2)3
C6H5
p-CH3C6H4
CH3O
ꢀ66.3
ꢀ71.6
ꢀ70.0
ꢀ70.2
ꢀ68.6
ꢀ83.8
ꢀ83.1
ꢀ96.7
ꢀ97.0
ꢀ98.4
ꢀ64.3
ꢀ29.1 [4a]
ꢀ65.7 [4a]
ꢀ68.2 [4a]
ꢀ66.7
ꢀ69.1
ꢀ67.5
ꢀ67.7
ꢀ12.21
ꢀ66.2
3.91
ꢀ81.4 [4a]
ꢀ80.6 [11]
ꢀ94.9
ꢀ45.6 [4a]
ꢀ45.0 [12]
ꢀ63.1
ꢀ80.5 [4a]
ꢀ80.6 [11]
ꢀ95.4 [4a]
11.05
5.41
6.56
11.03
6.04
11.15
6.10
6.05
7.246c and 7.781c
n-C8H17
C6H5O
O
ꢀ95.1
a
Not iterated.
ꢀ96.7
ꢀ67.6
ꢀ99.3 [4a]
b
c
Assignment uncertain, values may be exchanged.
Maximum peak in the complex multiplet, no assignment sym/asym possible.
aRSi[OCH(CH3)CH2]3N; b RSi(CH2CH2CH2)(OCH2CH2)2N; c RSi(OCH2CH2)3N.

18
I. Kovács et al. / Journal of Organometallic Chemistry 694 (2009) 14–20
Table 5
symmetrical isomer of p-tolyl-3,7,10-trimethylsilatrane 7asym
were determined by X-ray diffraction. (The crystal structure of
the asymmetrical isomer of 7aasym was published earlier [11].)
All three crystal structures are centrosymmetric. The Si N trans-
annular bonds vary over the range from 2.14 to 2.22 Å.
1H NMR data of C6H5Si(CH21,2CH23,4CH25,6) (OCH27,8CH29,10)2N 6b (solvent C6D6).
(ppm)
(Hz)
d1
d2
d3
d4
d5
d6
d7
d8
d9
d10
0.804
0.804
1.699
1.699
2.613
2.613
3.781
3.869
2.778
2.843
J1,2
J1,3
J1,4
J3,4
J3,5
J3,6
J5,6
J7,8
J7,9
J7,10
J9,10
J8,9
J8,10
ꢀ16.08
7.698
6.338
ꢀ12.522
7.259
4. Experimental
5.021
ꢀ10.766
ꢀ11.19
6.532
4.1. General
1H, 13C and 29Si NMR spectra were recorded on Bruker AC 250
and AMX 300 spectrometers (standard: external TMS), IR spectra
were measured as KBr pellets with a PYE UNICAM SP 2000
spectrometer and mass spectra were obtained with a JEOL JMS
01SG-2 instrument. HPLC separations were performed using a
Hitachi chromatograph equipped with a normal phase Silica gel
column and an UV detector (254 nm); eluant dichloromethane:
hexane:THF = 100:20:4.
Crystallographic data for 6asym, 6aasym were collected on a Bru-
ker SMART Apex CCD diffractometer, and for 7asym on an Enraf–
Nonius CAD4 diffractometer. Crystal data, data collection and
refinement parameters are summarised in Table 1. The structures
were solved by direct methods [18a] and were refined by full-ma-
trix least-squares on F2 using SHELXTL [18b].
5.031
ꢀ12.572
5.070
6.335
molecular dynamic. A higher symmetry than that for the ABXY type
is not possible for this part of the spectrum. The fitting for 6b yields
identical results for the –(CH2)3– bridge (Table 5), whether or not CS
symmetry was imposed [17]. Even if such parameters were refined
in the calculations, the calculated values for the 4JHH coupling con-
stants never differed significantly from zero.
The exchange of an electronegative oxygen atom for a CH2
group leads as expected to a high-field shift for both the neigh-
bouring methylene protons and those of the NCH2 group. The
chemical shift of the NCH2 protons is close to that of simple silatr-
anes and thus confirms that a transannular Si N interaction is
also present in the novel carbasilatranes presented here.
Ethoxy-trimethylsilatrane was prepared according to literature
procedures [8].
4.2. Synthesis of compounds 2a–5a, 8a–10a, and 10b
2.3.3. 13C NMR spectra
4.2.1. X(CH2)3OSi[OCH(CH3)CH2]3N (2a–5a)
As was the case for the 1H NMR spectra, in the 13C NMR spectra
of the 3,7,10-trimethylsilatranes only one spectrum is again ex-
pected for all three bridges of the symmetrical isomer, whereas
each bridge of the asymmetrical isomer must have its own spec-
trum. Due to the more electronegative oxygen, the signals of the
OCH2 group appear at lower field than those of the NCH2 group.
For compounds 6a and 7a, the 13C spectra of the separated isomers
again support the assignment. For the compounds 2a–5a and 8a–
10a the assignment was derived from 1H/13C correlation and 13C
DEPT spectra. A more advanced assignment of the signals to the
particular bridges is not possible.
Compared to those silatranes with no substituents on the ring
atoms, the values of the 13C NMR signals of the OCH2 and NCH2
groups in the carbasilatranes show only a weak low field shift;
the signals for the SiCH2 or CCH2C groups, respectively, appear at
considerably higher field.
0.02 mol of the corresponding X(CH2)3Si(OMe)3 was reacted
with 0.02 mol triisopropanolamine in xylene (20 mL) at 110 °C
with a catalytic amount of KOH. The methanol formed was re-
moved using a Marcusson device, and the reaction mixture was
then stirred for 1–2 h at 110 °C. The solvent was removed in vacuo
and the remaining oil worked up by distillation.
2a (X: Cl) Yield: 4.57 g (78%), b.p. 115–116 °C/0.02 mbar. Anal.
Calc. for C12H24ClNO3Si: C, 49.05; H, 8.23; N, 4.77; Si, 9.56. Found:
C, 49.12; H, 8.50; N, 4.97; Si, 9.59%. 13C{1H} NMR (CDCl3): d 63.9
(OCH sym), 67.3, 65.5, 65.4 (OCH asym), 59.5 (NCH2 sym), 65.7,
62.3, 62.3 (NCH2 asym), 20.9 (CH3 sym), 23.8, 21.3, 21.0 (CH3 asym),
48.8 (ClCH2), 29.6 (CCH2C), 14.6 (SiCH2). IR (cmꢀ1): 1464 m, 1380 s,
1160 s, 1115 s, 1098 m, 885 s, 780s, 662 m, 428 w. EI-MS: m/z (rel-
ative abundance, %) 293.2 (5), 295.2 (2) [M+], 216.1 (100) [M+-
Cl(CH2)3].
3a (X: SH) Yield: 4.65 g (80%), b.p. 121–122 °C/0.02 mbar. Anal.
Calc. for C12H25NO3SSi: C, 49.45; H, 8.64; N, 4.81; Si, 9.64. Found: C,
49.98; H, 8.93; N, 5.09; Si, 9.40%. 13C{1H} NMR (CDCl3): d 63.2 (OCH
sym), 66.6, 64.9, 64.7 (OCH asym), 58.7 (NCH2 sym), 65.0, 61.5, 61.5
(NCH2 asym), 20.3 (CH3 sym), 23.1, 20.7, 20.4 (CH3 asym), 30.6
(SCH2), 28.3 (CCH2C), 16.1 (SiCH2). IR (cmꢀ1): 1460 m, 1381 m,
1160 s, 1115 s, 1098 m, 885 s, 780s, 660 m, 428 w. EI-MS: m/z (rel-
ative abundance, %) 291.2 (2) [M+], 216.2 (100) [M+-HS(CH2)3].
4a (X: CH2@C(CH3)COO) Yield: 4.87 g (71%), b.p. 128–129 °C/
0.02 mbar. Anal. Calc. for C16H29NO5Si: C, 55.95; H, 8.51; N, 4.08;
Si, 8.18. Found: C, 55.33; H, 8.85; N, 4.79; Si, 8.02%. 13C{1H} NMR
(CDCl3): d 63.2 (OCH2 sym), 66.6, 64.9, 64.8 (OCH2 asym), 58.6
(NCH2 sym), 65.2, 61.5, 61.4 (NCH2 asym), 20.2 (CH3 sym), 23.1,
20.5, 20.3 (CH3 asym), 167.5 (COO), 136.7 (CH2@C), 124.6 (CH2@C),
67.9 (COCH2), 24.3 (CCH2C), 18.3 (@CCH3), 12.5 (SiCH2). IR (cmꢀ1):
1460 m, 1380 m, 1158 s, 1116 s, 1100 m, 883 s, 775s, 660 m, 423 w.
EI-MS: m/z (relative abundance, %) 343.3 (3) [M+], 216.2 (100) [M+-
CH2@C(CH3)COO(CH2)3].
3. Conclusion
The present study involves the preparation of several novel
3,7,10-trimethylsilatranes (2a–5a, 9a) and of the carbasilatrane
10b. The synthesis of phenoxy-trimethylsilatrane 10a was not pos-
sible using the same route as for alkoxy-trimethylsilatranes. How-
ever, the compound could be isolated from the reaction between
ethoxy-trimethylsilatrane and phenol. Thorough 29Si, 1H and 13C
NMR investigations show the influence of various substituents R
and of the substitution in the skeleton of trimethylsilatranes and
carbasilatranes on the chemical shift of the measured nuclei. The
introduction of methyl groups in positions 3, 7 and 10 of the sila-
trane cage leads to mixtures of diastereomers causing complex 1H
NMR spectra. By means of 1H COSY spectra and of 1H spectra of the
separated isomers for 6a and 7a the signal groups could be as-
signed to both diastereomers and also to the different protons at-
tached to the silatrane bridges. The crystal structures of both
diastereomers of phenyl-3,7,10-trimethylsilatrane 6a and of the
5a (X: H2N(CH2)2NH) Yield: 5.33 g (84%), b.p. 145–146 °C/
0.01 mbar. Anal. Calc. for C14H31N3O3Si: C, 52.96; H, 9.84; N,

I. Kovács et al. / Journal of Organometallic Chemistry 694 (2009) 14–20
19
13.23; Si, 8.85. Found: C, 51.91; H, 9.55; N, 13.01; Si, 8.21%. 13C{1H}
NMR (CDCl3): d 63.3 (OCH sym), 66.8, 65.0, 64.9 (OCH asym), 58.9
(NCH2 sym), 65.2, 61.8, 61.6 (NCH2 asym), 20.3 (CH3 sym), 23.1,
20.7, 20.4 (CH3 asym), 53.0, 52.6, 42.1(RNCH2), 25.3 (CCH2C), 13.6
(SiCH2). IR (cmꢀ1): 1461 m, 1379 m, 1158 s, 1112 s, 1095 m, 886
s, 780 s, 662 w, 425 w. EI-MS: m/z (relative abundance, %) 216.2
(100) [M+- H2N(CH2)2NH(CH2)3], 287.3 (30) [M+-NH2CH2].
1169 s, 1119 s, 1105 s, 920 k, 880 s, 790 s, 772 s, 760 s, 690 s, 590
w. EI-MS: m/z (relative abundance, %) 265.1 (10) [M+], 172.2 (100)
[M+-C6H5O].
Acknowledgements
We thank Prof. A.K. Powell for the use of X-ray diffractometers,
and Prof. J. Fekete for his valuable contribution in the chromato-
graphic work.
4.2.2. CH3OSi[OCH(CH3)CH2]3N (8a)
A solution of 1.52 g (0.01 mol) tetramethoxysilane and 1.91 g
(0.01 mol) triisopropanolamine in 10 mL xylene was heated to
100–110 °C. After addition of KOH, the methanol formed was dis-
tilled off. The residue was recrystallised from n-hexane. Yield:
1.78 g (72%), m.p. 92–94 °C. Anal. Calc. for C10H21NO4Si: C, 48.56;
H, 8.56; N, 5.66; Si, 11.35. Found: C, 49.22; H, 9.02; N, 5.78; Si,
10.83%. 13C{1H} NMR (CDCl3): d 63.9 (OCH2 sym), 67.0, 65.4, 65.3
(OCH2 asym), 59.6 (NCH2 sym), 65.7, 62.4, 62.1 (NCH2 asym), 20.9
(CH3 sym), 23.9, 21.2, 21.0 (CH3 asym), 51.5 (OCH3). IR (cmꢀ1):
1464 m, 1378 m, 1162 s, 1115 s, 1098 s, 888 s, 791s, 660 w, 425
w. EI-MS: m/z (relative abundance, %) 247.1 (15) [M+], 216.1
(100) [M+-CH3O].
Appendix A. Supplementary material
CCDC 678878, 678879, 678880 contain the supplementary crys-
tallographic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif. Supplementary material
associated with this article can be found, in the online version, at
doi:10.1016/j.jorganchem.2008.09.055.
References
[1] (a) P. Hencsei, L. Párkányi, Rev. Silicon, Germanium, Tin and Lead Comp. 8
(1985) 191;
4.2.3. C8H17OSi[OCH(CH3)CH2]3N (9a)
A
solution of 2.08 g (0.01 mol) tetraethoxysilane, 1.91 g
(b) R.J.P. Corriu, J. Organomet. Chem. 400 (1990) 81;
(c) M.W. Schmidt, T.L. Windus, M.S. Gordon, J. Am. Chem. Soc. 117 (1995)
7480;
(d) J.M. Anglada, C. Bo, J.M. Bofill, R. Crehuet, J.M. Poblet, Organometallics 18
(1999) 5584;
(e) A.B. Trofimov, V.G. Zakrzewski, O. Dolgounitcheva, J.V. Ortiz, V.F. Sidorkin,
E.F. Belogolova, M. Belogolov, V.A. Pestunovich, J. Am. Chem. Soc. 127 (2005)
986;
(f) M.V. Zabalov, S.S. Karlov, G.S. Zaitseva, D.A. Lemenovskii, Russ. Chem. Bull.
55 (2006) 464.
(0.01 mol) triisopropanolamine and 1.43 g (0.011 mol) octylalcohol
in xylene (10 mL) was heated to 100–110 °C. After addition of KOH,
the ethanol formed was distilled off. The reaction mixture was
worked up by distillation. Yield: 2.59 g (75%), b.p. 123–125 °C/
0.02 mbar. Anal. Calc. for C17H35NO4Si: C, 59.09; H, 10.21; N,
4.05; Si, 8.13. Found: C, 59.62; H, 10.67; N, 4.15; Si, 8.36%.
13C{1H} NMR (CDCl3): d 63.4 (OCH sym), 66.5, 65.0, 64.9 (OCH
asym), 59.0 (NCH2 sym), 65.1, 61.8, 61.6 (NCH2 asym), 20.2 (CH3
sym), 23.2, 20.6, 20.4 (CH3 asym), 63.2 (CH2OSi) 32.7, 31.8, 29.6,
29.3, 25.9, 22.6, (RCH2), 14.1 (RCH3). IR (cmꢀ1): 1460 m, 1376 m,
1150 s, 1115 s, 1100 m, 886 s, 784 s, 672 w. EI-MS: m/z (relative
abundance, %) 345.2 (7) [M+], 216.1 (100) [M+-C8H17O].
[2] (a) Q. Shen, R.L. Hilderbrandt, J. Mol. Struct. 64 (1980) 257;
(b) G. Forgács, M. Kolonits, I. Hargittai, Struct. Chem. 1 (1990) 245.
[3] (a) A. Chandrasekaran, R.O. Day, R.R. Holmes, J. Am. Chem. Soc. 122 (2000)
1066;
(b) M.G. Voronkov, A.N. Egorochkin, O.V. Kuznetsova, J. Organomet. Chem. 691
(2006) 159;
(c) A.N. Egorochkin, M.G. Voronkov, O.V. Kuznetsova, O.V. Novikova, J.
Organomet. Chem. 693 (2008) 181.
[4] (a) M.G. Voronkov, V.M. Dyakov, S.V. Kirchipenko, J. Organomet. Chem. 233
(1982) 1. and references cited therein;
4.2.4. C6H5OSi[OCH(CH3)CH2]3N (10a)
6.53 g (0.025 mol) ethoxy-trimethylsilatrane dissolved in
50 mL xylene was added to a solution of 2.35 g (0.025 mol) phenol
in 50 mL xylene. The mixture was heated to 100 °C. After adding a
catalytic amount of KOH, the ethanol formed was distilled off using
a Marcusson device. The residue was recrystallised from n-hexane.
Yield: 2.86 g (37%), m.p. 117–119 °C. Anal. Calc. for C15H23NO4Si: C,
58.21; H, 7.49; N, 4.53; Si, 9.07. Found: C, 58.47; H, 7.21; N, 4.43; Si,
8.73%. 13C{1H} NMR (CDCl3): d 63.7 (OCH sym), 68.0, 65.8, 64.0
(OCH asym), 59.2 (NCH2 sym), 66.6, 65.9, 65.9 (NCH2 asym), 20.4
(CH3 sym), 22.8, 20.7, 20.5 (CH3 asym), 156.4, 128.6, 120.5, 119.6
(PhC). IR (cmꢀ1): 1465 m, 1380 m, 1155 s, 1112 s, 1095 m, 885 s,
794s, 440 w. EI-MS: m/z (relative abundance, %) 309.4 (9) [M+],
216.1 (100) [M+-C6H5O].
(b) S.N. Tandura, M.G. Voronkov, N.V. Alekseev, Top. Curr. Chem. 131 (1986)
99.
[5] (a) M.G. Voronkov, Top. Curr. Chem. 84 (1979) 77;
(b) L. Bihátsi, P. Hencsei, L. Kótai, G. Ripka, Növényvédelem 28 (1992) 180;
(c) L. Zhonghua, S. Xiuyan, S. Huaping, C. Jing, Heterocyclic Commun. 11
(2005) 475;
(d) A. Zablotskaya, I. Segal, S. Belyakov, E. Lukevics, Appl. Organomet. Chem.
20 (2006) 149.
[6] (a) N. Kuanchertchoo, S. Kulprathipanja, P. Aungkavattana, D. Atong, K.
Hemra, T. Rirksomboon, S. Wongkasemjit, Appl. Organomet. Chem. 20
(2006) 775;
(b) N. Thanabodeekij, S. Sadthayanon, E. Gulari, S. Wongkasemjit, Mater.
Chem. Phys. 98 (2006) 131;
(c) N. Kritchayanon, N. Thanabodeekij, S. Jitkarnka, A.M. Jamieson, S.
Wongkasemjit, Appl. Organomet. Chem. 20 (2006) 155.
[7] C.L. Frye, G.E. Vogel, J.A. Hall, J. Am. Chem. Soc. 83 (1961) 996.
[8] M.G. Voronkov, G.I. Zelchan, Khim. Geteroc. Soed. (1966) 511.
[9] E. Lukevics, L.I. Libert, M.G. Voronkov, Zh. Obshch. Khim. 39 (1969) 1784.
[10] L. Bihátsi, P. Hencsei, I. Kovács, L. Kótai, HU 210806, December 01, 1989 [C.A.
1991, 115:280295].
[11] L. Párkányi, P. Hencsei, L. Bihátsi, I. Kovács, A. Szöllo}sy, Polyhedron 4 (1985)
243.
[12] L. Párkányi, V. Fülüp, P. Hencsei, I. Kovács, J. Organomet. Chem. 418 (1991)
173.
[13] (a) J.W. Turley, F.P. Boer, J. Am. Chem. Soc. 90 (1968) 4026;
(b) L. Párkányi, K. Simon, J. Nagy, Acta Crystallogr. B 30 (1974) 2328;
(c) L. Párkányi, J. Nagy, K. Simon, J. Organomet. Chem. 101 (1975) 11.
[14] (a) A. Greenberg, G. Wu, Struct. Chem. 1 (1991) 79;
(b) P. Hencsei, Struct. Chem. 2 (1991) 21.
4.2.5. C6H5OSi[(OCH2CH2)2 (CH2CH2CH2)]N (10b)
A mixture of 4.96 g (0.025 mol)
c-chloropropyltrimethoxysi-
lane, 2.63 g (0.025 mol) diethanolamine, and 2.35 g (0.025 mol)
phenol in xylene (25 mL) was heated to 100 °C. After adding
KOH, the formed methanol was distilled off by means of a Marcus-
son device. After removing the device, 4.9 mL (0.035 mol) triethyl-
amine was added and the reaction mixture refluxed for 20 h. The
precipitated triethylamine hydrochloride was filtered off, the fil-
trate concentrated to dryness, and the residue recrystallised from
n-hexane. Yield: 1.66 g (25%), m.p. 68–71 °C. Anal. Calc. for
C13H19NO3Si: C, 58.84; H, 7.22; N, 5.28; Si, 10.58. Found: C,
59.12; H, 7.48; N, 5,01; Si, 11.17%. 13C{1H} NMR (CDCl3): d 58.6
(3C OCH2), 52.3 (2C NCH2), 51.9 (NCH2), 20.3 (CCH2C), 8.4 (SiCH2),
156.9, 128.7, 120.1, 119.4 (PhC). IR (cmꢀ1): 1600 s, 1498 s, 1269 s,
[15] (a) R.K. Harris, B.E. Mann (Eds.), NMR and the Periodic Table, Academic Press,
London, New York, San Francisco, 1978;
(b) V.A. Pestunovich, S.N. Tandura, M.G. Voronkov, G. Engelhardt, E. Lippmaa,
T. Pehk, V.F. Sidorkin, G.I. Zelchan, V.P. Baryshok, Dokl. Akad. Nauk SSSR 240
(1978) 914.
[16] (a) M.G. Voronkov, Pure Appl. Chem. 13 (1966) 35;
(b) J.M. Bellama, J.D. Nies, N. Ben-Zvi, Magn. Reson. Chem. 240 (1986) 748.

20
I. Kovács et al. / Journal of Organometallic Chemistry 694 (2009) 14–20
[17] (a) Programs WinDaisy and WinNmr, Bruker Daltonik, Bremen, Germany,
1999;
[18] (a) G.M. Sheldrick, SHELXS-97, Program for Crystal Structure Solution, University
of Göttingen, Germany, 1997;
(b) G. Hägele, M. Engelhardt, W. Boenigk, Simulation und automatisierte
Analyse von Kernresonanzspektren, VCH, Weinheim, Germany, 1987.
(b) G.M. Sheldrick, SHELXL-97
University of Göttingen, Germany, 1997.
, Program for Crystal Structure Refinement,