Switching between penta- and hexacoordination with salen-silicon-complexes

Article abstract of DOI:10.1016/j.ica.2005.03.036

The reaction between phenyltrichlorosilane and the tetradentate ligands o-HO-C₆H₄-C(CH₃)=N-(CH₂) _n-N=C(CH₃)-o-C₆H₄-OH (n = 2, 3, 4), supported by an amine base, yields pentacoordinate silicon complexes (C ₆H₅)Si-[o-O-C₆H₄-C(CH ₃=)N-(CH₂)_n-N-C(=CH₂)-o-C ₆H₄-O] with enamine functionalized ligands. This reaction pattern can be transferred onto various ligands of 2-iminomethylphenolate-type. The resulting pentacoordinate silicon complexes react with a variety of Br?nsted acids HY to yield hexacoordinate salen silicon complexes (C ₆H₅)(Y)Si-[o-O-C₆H₄-C(CH ₃)=N-(CH₂)_n-N=C(CH₃)-o-C ₆H₄-O] (Y = benzoate, picrate, 8-oxyquinolinate, 2-oxy-1,4-naphthoquinonate, p-tert-butylphenolate, (5-phenyltetrazol)-2-ide, fluoride, tetrafluoroborate). Hexacoordination of their Si-atoms was confirmed by ²⁹Si NMR spectroscopy and, in some cases, by X-ray crystal structure analysis. Examples for similarities and differences in the coordination behavior of the silicon atom and its heavier congeners (Ge, Sn) in the salen-type coordination sphere as well as data regarding the nucleophilicity of some of these novel enamine complexes are presented.

Full text of DOI:10.1016/j.ica.2005.03.036

Inorganica Chimica Acta 358 (2005) 4270–4286
www.elsevier.com/locate/ica
Switching between penta- and hexacoordination with
salen-silicon-complexes
a
Jo¨rg Wagler , Uwe Bo¨hme , Erica Brendler , Berthold Thomas , Sigrid Goutal ,
a
b
b
c
d
Herbert Mayr , Bernhard Kempf , Grigoriy Ya. Remennikov , Gerhard Roewer
d
d
a,
*
a
Institut fu¨r Anorganische Chemie, Technische Universita¨t Bergakademie Freiberg, Leipziger Str. 29, D-09596 Freiberg, Germany
Institut fu¨r Analytische Chemie, Technische Universita¨t Bergakademie Freiberg, Leipziger Str. 29, D-09596 Freiberg, Germany
b
c
Institut fu¨r Organische Chemie, Technische Universita¨t Dresden, Bergstr. 66, D-01066 Dresden, Germany
Department Chemie und Biochemie, Ludwig-Maximilians-Universita¨t Mu¨nchen, Butenandtstr. 5-13 (F), D-81377 Mu¨nchen, Germany
d
Received 1 February 2005; accepted 14 March 2005
Available online 26 April 2005
Dedicated to Prof. Dr. Hubert Schmidbaur on the occasion of his 70th birthday.
Abstract
The reaction between phenyltrichlorosilane and the tetradentate ligands o-HO-C6H4–C(CH3)@N–(CH2)n–N@C(CH3)-o-C6H4-
OH (n = 2, 3, 4), supported by an amine base, yields pentacoordinate silicon complexes (C6H5)Si-[o-O-C6H4–C(CH3)@N–
(CH2)n–N–C(@CH2)-o-C6H4-O] with enamine functionalized ligands. This reaction pattern can be transferred onto various ligands
of 2-iminomethylphenolate-type. The resulting pentacoordinate silicon complexes react with a variety of Brønsted acids HY to yield
hexacoordinate salen silicon complexes (C6H5)(Y)Si-[o-O-C6H4–C(CH3)@N–(CH2)n–N@C(CH3)-o-C6H4-O] (Y = benzoate, picrate,
8-oxyquinolinate, 2-oxy-1,4-naphthoquinonate, p-tert-butylphenolate, (5-phenyltetrazol)-2-ide, fluoride, tetrafluoroborate). Hexac-
oordination of their Si-atoms was confirmed by 29Si NMR spectroscopy and, in some cases, by X-ray crystal structure analysis.
Examples for similarities and differences in the coordination behavior of the silicon atom and its heavier congeners (Ge, Sn) in
the salen-type coordination sphere as well as data regarding the nucleophilicity of some of these novel enamine complexes are
presented.
ꢀ 2005 Elsevier B.V. All rights reserved.
Keywords: Enamine; Germanium; Hypercoordination; Schiff base; Silicon; Tin
1. Introduction
Thus, hypercoordinate silicon complexes represent min-
ima on a reaction coordinate during a substitution reac-
tion. Both the theoretical aspects of investigating such
model compounds [5] and the practical use of bond acti-
vation by hypercoordination in organic and inorganic
chemistry [6] are important driving forces in the devel-
opment of the coordination chemistry of silicon. Refer-
ring to these basic correlations, controlled switching of
the Si atomꢀs coordination number by variation of sol-
vent [7] or temperature [8] as well as irradiation [9],
exploring the role of hypercoordinate silicon species in
the biological SiO2 transport [10], investigating unique
material properties based on hypercoordinate silicon
The chemistry of hypercoordinate silicon compounds
stimulates the synthesis of novel compounds [1–3] as
well as approaches to vary properties of materials [4].
The increase of the Si atomꢀs coordination number to
five or six by coordinating additional donors provokes
an activation of all bonds around the silicon atom.
*
Corresponding author. Tel.: +49 3731 39 3174; fax: +49 3731 39
4058.
E-mail address: gerhard.roewer@chemie.tu-freiberg.de (G. Roewer).
0020-1693/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2005.03.036

J. Wagler et al. / Inorganica Chimica Acta 358 (2005) 4270–4286
4271
compounds as well as focussed preparation of hyperco-
ordinate silicon compounds for certain applications,
e.g., rhodenticides [11] or anti-cancer drugs [12], are
prospering branches of silicon coordination chemistry.
Recently, we found a novel approach to switch be-
tween penta- and hexacoordination of the silicon atom
by adding Brønsted acids to Si complexes with enamine
functionalized salen-type ligands (Scheme 1) [13–15].
For the first time, Si complexes with enamine function-
alized salen-type ligands were confirmed by X-ray crys-
tal structure analysis. Those compounds allow an easy
access to many hexacoordinate silicon complexes by a
1,4-addition reaction.
In this paper, we report the syntheses of novel
pentacoordinate Si complexes with enamine function-
alized bi- and tetradentate chelating ligands and their
reactions with Brønsted acids to yield hexacoordinate
Si complexes. Selected germanium and tin com-
pounds were also synthesized and characterized for
contrasting their coordination behavior and reactivity
pattern from that of the silicon complexes. Using a
UV/Vis-spectroscopic method, we try to scale the
nucleophilicity of some selected enamine complexes
and compared them with structurally analogous
metal-free enamines.
NMR spectra were recorded on a BRUKER DPX
400 (CDCl3 solution with TMS as internal standard)
and a BRUKER Avance 400WB spectrometer (solid
state) as well as NMR on a Bruker ARX 300 (CDCl3
solution of 12 with TMS as internal standard). Elemen-
tal analyses were carried out on a Foss Heraeus CHN-
O-Rapid (except 12: elementar Vario EL). X-ray single
crystal structure analyses were carried out on an EN-
RAF NONIUS CAD4 diffractometer (4), a BRUKER
NONIUS KappaCCD diffractometer (2b, 2c, 3a Æ 5/8p-
xylene, 3b Æ 2dioxane, 3c Æ thf, 3i Æ 1/2 thf) and on a
BRUKER NONIUS APEX2 CCD diffractometer (3f,
4a, 5, 8a Æ 1/2p-xylene, 8b Æ 1/2p- xylene). Structures were
solved with direct methods, refined with least squares
method, all non-hydrogen atoms were anisotropically
refined. Hydrogen atoms were placed in idealized posi-
tions and isotropically refined.
Kinetics of the reactions of the enamine complexes 2a
and 4 with the benzhydrylium tetrafluoroborates 10a,b-
BF4 were performed by monitoring the Vis-absorbances
of the benzhydrylium cations using the methodology de-
scribed previously [30].
Kinetic measurements of the reactions of benzhydryl
cations (Ar2CH+) 10a,b with the enamine complexes 2a
and 4 in CH2Cl2 at 20 ꢁC:
Ar2CH+
[Ar2CH+]0 (M)
enamine
complex
[2a]0 or [4]0 (M)
[2a]0 or [4]0/[Ar2CH+]0
k2 (20 ꢁC)
(Mꢀ1 sꢀ1
)
10a E = ꢀ 4.72
5.500 · 10ꢀ5
6.500 · 10ꢀ5
1.300 · 10ꢀ4
2a
2a
2a
5.959 · 10ꢀ5
1.192 · 10ꢀ4
1.788 · 10ꢀ4
1.1
1.8
1.4
2.394 · 106
2.409 · 106
2.344 · 106
Average
2.382 · 106
10b E = ꢀ 7.69
2.294 · 10ꢀ5
2.294 · 10ꢀ5
2.294 · 10ꢀ5
4
4
4
1.365 · 10ꢀ4
2.729 · 10ꢀ4
4.094 · 10ꢀ4
5.9
11.9
17.8
7.040 · 104
7.062 · 104
7.980 · 104
Average
7.361 · 104
2. Experimental
Kinetic measurements were performed with a
Hi-Tech SF61DX2 stopped-flow UV/Vis-spectropho-
The used chemicals were commercially available. All
following manipulations were carried out under an inert
atmosphere of dry argon. p-Xylene, dioxane and THF
were distilled from sodium/benzophenone prior to use.
Triethylamine was distilled from calcium hydride and
tometer for 10a at kmax = 672 nm under second-order
conditions (see concentrations) and for 10b at kmax
=
620 nm under pseudo-first-order conditions (see
concentrations).
Ligands 1a, 1b, 1c and 6 were prepared by condensa-
tion of 2-hydroxyacetophenone with the required
amines in isopropanol similar to a previously published
method [33].
˚
stored over molecular sieve 3 A. Chloroform (stabilized
˚
with amylene) was dried over molecular sieve 3 A.
HX
2a was prepared according to a previously described
method [13,14].
2b: In 150 ml thf, 1b (6.00 g; 19.3 mmol) and triethyl-
amine (7.00 g; 69.3 mmol) were stirred at room temper-
ature and phenyltrichlorosilane (4.15 g; 19.6 mmol) was
X
N
O
N
O
N
O
N
O
Si
Si
Ph
Ph
Scheme 1.

4272
J. Wagler et al. / Inorganica Chimica Acta 358 (2005) 4270–4286
added dropwise. The mixture was allowed to stand over-
night, then the precipitated triethylamine hydrochloride
was filtered off and washed with thf (30 ml). The solvents
were removed from the filtrate under vacuum, the solid
product was dissolved in thf (35 ml) and hexane (25 ml)
was added. This mixture was stored at 6 ꢁC for three
weeks. The crystalline product was filtered off, washed
with thf/hexane mixture (1:4) and dried under vacuum.
Yield: 4.60 g (11.1 mmol; 58%) yellow crystals, M.p.
174 ꢁC.
169.8 (C@N); 29Si NMR: ꢀ106.4 (79 MHz, CDCl3);
ꢀ109.6 (CP/MAS).
3a and 3b were prepared according to a previously
described procedure [13,14] and recrystallized from p-
xylene (3a) or dioxane (3b).
3a Æ p-xylene: Anal. Calc. for C36.5H34.875N2O4Si
(found, %): C 73.85 (74.45); H 5.92 (6.04); N 4.72
(5.57). 29Si CP/MAS NMR: ꢀ186.0.
3b Æ (2dioxane): Anal. Calc. for C38H41N5O13Si
(found, %): C 56.78 (55.82); H 5.14 (5.04); N 8.71
(9.01). 29Si CP/MAS NMR: ꢀ172.0.
Anal. Calc. for C25H24N2O2Si (found, %): C 72.78
(72.31); H 5.86 (6.06); N 6.79 (6.80).
3c: In 35 ml thf, 2a (1.30 g; 3.26 mmol) was dissolved
and a solution of 8-hydroxyquinoline (0.48 g; 3.3 mmol)
in 5 ml thf was added. The clear solution was stored at 6
ꢁC for 1 week. Then, the solution was separated from the
crystalline product and its volume was reduced to 20 ml
under vacuum. After cooling down to 6 ꢁC for further 3
weeks, more crystaline product 3ag Æ thf was obtained.
Yield: 1.25 g (2.03 mmol; 62%) light yellowish crys-
tals, M.p. 154 ꢁC.
1H NMR (400 MHz, CDCl3): 1.65–1.80; 2.30–2.45 (2
m, 2, N–CH2–CH2–CH2–N); 2.39 (s, 3, CH3); 3.20–3.70
(m, 4; 4,57 N–CH2–CH2–CH2–N); 4.18; 4.60 (2 d, 2,
2
3
C@CH2, JH–H = 1.2 Hz); 6.65 (dd, 1, ar, JH–H = 8.0
Hz, JH–H = 1.2 Hz); 6.77 (dt, 1, ar); 7.00–7.25 (mm, 9,
4
3
ar); 7.42 (dt, 1, ar); 7.59 (dd, 1, ar, JH–H = 8.0 Hz,
4JH–H = 1.2 Hz); 13C NMR (101 MHz, CDCl3): 16.5
(CH3); 25.8; 44.4; 46.8 (N–CH2–CH2–CH2–N); 81.7
(C@CH2); 119.0; 120.3; 120.6; 120.9; 122.7; 124.4; 125.7;
127.2; 127.9; 128.5; 129.2; 131.1; 134.5; 141.8; 152.4 (ar,
C@CH2); 155.4; 156.6 (ar C–O); 169.6 (C@N); 29Si
NMR: ꢀ107.7 (79 MHz, CDCl3); ꢀ107.3 (CP/MAS).
2c: In 300 ml thf, 1c (10.0 g; 30.8 mmol) and triethyl-
amine (11.0 g; 109 mmol) were stirred at 50 ꢁC and phe-
nyltrichlorosilane (6.6 g; 31.2 mmol) was added
dropwise. The mixture was stirred at 50 ꢁC for 15 min
and then stored at room temperature overnight. The
precipitated triethylamine hydrochloride was filtered
off and washed with thf (50 ml). Upon removal of the
volatiles under vacuum, the filtrate yielded a yellow
oil. This was dissolved in chloroform (19 ml). Then, hex-
ane (10 ml) was added. The product was crystallized
from this solution at room temperature. These yellow
crystals were filtered off, washed with chloroform/hex-
ane mixture (1:4) and dried under vacuum. Further
addition of hexane to the filtrate yielded a second frac-
tion of the product within some weeks.
Anal. Calc. for C37H37N3O4Si (found, %): C: 72.17
(72.00); H: 6.06 (6.05); N: 6.82 (6.66).
1H NMR (400 MHz, CDCl3): 2.39 (s, 6, CH3); 3.65–
3.75; 4.30–4.40 (2 m, 4; N–CH2CH2–N); 6.40–7.40 (mm,
3
4
17, ar); 7.76 (dd, 1, ar, JHH = 8.0 Hz, JHH = 1.6 Hz);
8.55 (dd, 1, ar, JHH = 4.3 Hz, JHH = 1.6 Hz); 13C
NMR (101 MHz, CDCl3): 18.3 (CH3); 46.6 (N–
CH2CH2–N); 116.3; 117.0; 119.1; 120.3; 122.3; 125.5;
126.9; 127.5; 128.4; 128.9; 129.1; 132.8; 134.3; 135.7;
143.2; 146.9; 157.6 (ar); 160.4; 161.4 (ar C–O); 169.7
(C@N); 29Si NMR (79 MHz, CDCl3): ꢀ181.3.
3
4
The 1H, 13C and 29Si NMR spectra also show signals
of the enamine 2a as well as those of 8-hydroxyquinoline
due to partial dissociation of 3c in solution.1
1
Recently, we reported on the surprising dissociation of a Si–O bond
under formation of a Si–N enamine moiety [23]. In case of compounds
3c, 3i and 8a the same phenomenon was NMR spectroscopically
observed in CDCl3 solution.
Yield: 7.80 g (18.3 mmol; 59%), M.p. 163 ꢁC, yellow-
ish crystals.
Anal. Calc. for C26H26N2O2Si (found, %): C 73.21
(72.87); H 6.14 (6.26); N 6.57 (6.30).
In the solution of 3c in CDCl3, the coexistence of 3c, 2a and 8-
hydroxyquinoline (ratio 3c:2a = 3:1) was proven by 1H, 13C and 29Si
NMR spectroscopy. In case of a solution of 8a, the ligand 6 was 1H
and 13C NMR spectroscopically identified and signals of an enamine
moiety, which are different from those of the enamine complex 7, as
well as a 29Si NMR signal at ꢀ54.3 ppm indicate the formation of
complex 8aa. In the case of 3i, additional signals to those of 3i indicate
to the formation of an enamine different from 3b and an salen-type
ligand moiety, different from 1b. The 29Si NMR signal at ꢀ56.2 ppm
also points to the formation of a distinct tetracoordinate silicon
compound. In all three cases, a Si–O bond is cleaved and a Si–N bond
is formed. In those cases, OH groups are formed which posess an O–
H–N hydrogen bridge, thus indicating to that structural pattern as a
driving force for the elimination of an acid from hexacoordinate Si
complexes such as 3i. In case of complexes 3a, 3b and 3d, the
elimination of one half-side of the tetradentate salen-type ligand, as
reported for 3i, has not been observed. The tetradentate chelate in
these complexes seems to be stabilized by the five-membered diazasil-
oline ring system, thus, not easily providing for a non-coordinating
imine-N-atom to form the O–H–N hydrogen bridge.
1H NMR (400 MHz, CDCl3): 1.45–1.60; 1.80–1.90;
2.05–2.20 (3 m, 4, N–CH2–CH2–CH2–CH2–N); 2.40 (s,
3, CH3); 3.15–3.25; 3.55–3.65; 3.85–3.95 (3 m, 4; 4,57
N–CH2–CH2CH2–CH2–N); 4.08; 4.37 (2s, 2, C@CH2);
3
4
6.41 (dd, 1, ar, JH–H = 8.0 Hz, JH–H = 1.2 Hz); 6.80–
3
7.10 (mm, 9, ar); 7.36 (dd, 1, ar, JH–H =8.0 Hz, JH–H
4
=
1.2 Hz); 7.50 (dt, 1, ar); 7.59 (dd, 1, ar, JH–H = 7.6 Hz,
3
4JH–H = 1.6 Hz); 7.65 (dd, 1, ar, JH–H = 8.0 Hz,
3
4JH–H = 1.2 Hz); 13C NMR (101 MHz, CDCl3): 17.1
(CH3); 23.9; 29.0; 45.8; 49.9 (N–CH2–CH2–CH2–CH2–
N); 83.0 (C@CH2); 119.6; 120.6; 121.0; 122.8; 126.7;
126.8; 127.1; 127.3; 128.1; 128.6; 129.2; 129.4 134.4;
134.9; 142.7; 150.3 (ar, C@CH2); 154.7; 156.3 (ar C–O);

J. Wagler et al. / Inorganica Chimica Acta 358 (2005) 4270–4286
4273
3d: A solution of 2-hydroxy-1,4-naphthoquinone
(0.50 g; 2.9 mmol) in 8 ml thf was slowly added to a stir-
red solution of 2a (1.14 g; 2.86 mmol) in 40 ml thf. From
the orange colored clear product solution 3d precipi-
tated out after a few seconds. After one day the yellow
precipitate was filtered off, washed with thf (10 ml)
and dried under vacuum.
Yield: 1.25 g (2.19 mmol; 76%) yellow crystalline
powder, M.p. 179 ꢁC (decomposition).
Anal. Calc. for C34H28N2O5Si (found, %): C 71.31
Anal. Calc. for C24H23N2O2SiF (found, %): C 68.87
(69.18); H 5.54 (5.73); N 6.69 (6.35).
29Si NMR (CP/MAS 59.6 MHz; mspin = 4 kHz):
ꢀ179.9; ꢀ182.8.
3h: To a solution of 2a (1.80 g; 4.52 mmol) in thf (60
a solution of HBF4 (0.74 g; 4.57 mmol
ml),
HBF4 Æ diethylether) was added and the mixture was
stirred under reflux for 1 h to give a white precipitate.
It was cooled down to 20 ꢁC, filterted off, washed with
thf/diethylether 1:1 (5 ml) and diethylether (15 ml) and
dried under vacuum.
(71.27); H 4.93 (5.09); N 4.89 (4.92).
1H NMR (400 MHz, CDCl3): 2.57 (s, 6, CH3); 3.70–
3.80; 4.05–4.15 (2 m, 4; N–CH2–CH2–N); 6.55 (s, 1, ar);
6.60–7.90 (mm, 17, ar); 13C NMR (101 MHz, CDCl3):
18.4 (CH3); 46.3 (N–CH2CH2–N); 114.7; 118.3; 119.6;
122.9; 125.4; 125.5; 126.2; 127.1; 129.0; 131.2; 131.6;
132.9; 133.0; 133.2; 135.6; 157.1 (ar); 160.8; 161.0 (ar
C–O); 171.3 (C@N); 184.2; 186.1 (C@O); 29Si NMR
(79 MHz, CDCl3): ꢀ180.3; (CP/MAS): ꢀ180.9.
3e: A solution of p-tert-butylphenole (0.86 g; 5.73
mmol) in 6 ml thf was added to a stirred solution of
2a (2.28 g; 5.7 mmol) in 70 ml thf. Immediatedly, 3e pre-
cipitated out as a white solid. The mixture was stirred at
40 ꢁC for 10 min and stored at room temperature for 3
h. Then, the precipitate was filtered off, washed with thf
(6 ml) and dried under vacuum.
Yield: 1.40 g, nearly colorless (pale yellowish)
powder.
Anal. Calc. for C24H23N2O2SiF4B (found, %): C
59.30 (65.63); H 4.70 (5.60); N 5.80 (6.89).
29Si NMR (CP/MAS): see Section 3.
3i: A solution of benzoic acid (0.43 g; 3.5 mmol) in thf
(3 ml) was added to a stirred solution of 2b (1.43 g; 3.47
mmol) in thf (10 ml) at 20 ꢁC. The resulting clear solu-
tion was stored at 6 ꢁC for 2 days. The crystalline prod-
uct (3i Æ 0.5 thf) was then filtered off, washed with thf (3
ml) and dried under vacuum.
Yield: 1.70 g (2.98 mmol; 86%), nearly colorless (light
yellowish) crystals, M.p. 145 ꢁC.
Anal. Calc. for C34H34N2O4.5Si (found, %): C 71.55
(70.82); H 6.00 (5.99); N 4.91 (4.79).
Yield: 2.47 g (4.51 mmol; 79%) white powder, M.p.
240 ꢁC (decomposition).
Anal. Calc. for C34H36N2O3Si (found, %): C 74.42
1H NMR (400 MHz, CDCl3): 2.47 (s, 6, CH3); Due to
a dissociation equilibrium of 3i in solution1 and super-
position of many 1H NMR signals of two complexes, de-
1
(74.29); H 6.61 (7.00); N 5.11 (5.07).
tailed H chemical shifts for 3i cannot be provided.
29Si NMR (CP/MAS): ꢀ183.1.
13C NMR (101 MHz, CDCl3): 19.5 (CH3); 23.9 (N–
CH2CH2CH2–N); 52.5 (N–CH2CH2CH2–N); 117.7;
120.9; 122.6; 126.0; 126.9; 127.2; 128.2; 128.7; 130.4;
134.0; 134.5; 135.6; 156.1 (ar); 162.3 (ar C–O); 166.7
(COO); 173.9 (C@N); 29Si NMR: ꢀ184.2 (79 MHz,
CDCl3).
3f: A solution of phenyltetrazole (0.45 g; 3.1 mmol) in
10 ml thf was added dropwise to a stirred solution of 2a
(1.22 g; 3.06 mmol) in thf (30 ml). After a few seconds a
white precipitate formed. After 30 min it was filtered off,
washed with thf (15 ml) and dried under vacuum.
Yield: 1.25 g (2.30 mmol; 74%) white powder, M.p.:
>200 ꢁC (decomposition without melting).
4: In 20 ml chloroform, 1a (1.97 g; 6.46 mmol) and
triethylamine (2.50 g; 24.7 mmol) were stirred at 30
ꢁC. Then, phenyltrichlorogermane (1.70 g; 6.64 mmol)
was added dropwise. The resulting orange colored clear
solution was stored at room temperature for 1 week to
give yellow crystals of 4. They were filtered off, washed
with chloroform (3 ml) and dried under vacuum.
Yield: 1.73 g (3.91 mmol; 61%) orange-yellowish crys-
tals, M.p. decomposition (ca. 250 ꢁC).
Anal. Calc. for C31H28N6O2Si (found, %): C 68.36
(66.41); H 5.18 (5.15); N 15.43 (14.53).
1H NMR (400 MHz, DMSO-d6): 2.62 (s, 6, CH3);
3.90–4.10 (m, 4; N–CH2–CH2–N); 6.75–7.95 (mm, 18,
ar); 13C NMR (101 MHz, DMSO-d6): 18.9 (CH3); 45.9
(N–CH2CH2–N); 118.6; 119.2; 120.9; 126.0; 126.4;
127.1; 128.3; 128.5; 129.6; 130.1; 132.1; 133.0; 135.7;
151.9 (ar); 159.6 (ar C–O); 174.6 (C@N); 29Si NMR
(79 MHz, DMDO-d6): ꢀ175.6; (CP/MAS): ꢀ172.6.
3g: To a stirring solution of 2a (4.15 g; 10.4 mmol) in
200 ml thf, a solution of HF (0.29 g 70% HF in pyri-
dine = 203 mg, 10.15 mmol HF) in 5 ml thf was added.
Then the mixture was stirred under reflux for 2 h to give
a white precipitate. It was cooled down to 20 ꢁC, filtered
off, washed with thf (20 ml) plus pentane (40 ml) and
dried under vacuum.
Anal. Calc. for C24H22N2O2Ge (found, %): C 65.07
(64.85); H 5.00 (5.06); N 6.32 (6.14).
1H NMR (400 MHz, CDCl3): 2.49 (s, 3, CH3); 3.50–
4.20 (m, 4; N–CH2CH2–N); 4.14; 4.45 (2 s, 2, C@CH2);
6.65-7.50 (mm, 13, ar); 13C NMR (101 MHz, CDCl3):
18.0 (CH3); 45.6; 48.0 (N–CH2CH2–N); 86.0 (C@CH2);
118.6; 119.7; 120.1; 120.3; 123.4; 127.1; 127.2; 128.7;
128.8; 129.7; 132.4; 133.4; 134.7; 136.7; 153.1 (ar,
C@CH2); 156.8; 162.1 (ar C–O); 172.4 (C@N).
4a: In 20 ml dioxane, 4 (0.46 g; 1.04 mmol) was dis-
solved at 100 ꢁC and a solution of picric acid (0.24 g;
Yield: 3.70 g (8.85 mmol; 87%), beige powder, M.p.
230 ꢁC (decomposition).

4274
J. Wagler et al. / Inorganica Chimica Acta 358 (2005) 4270–4286
1.05 mmol) in dioxane (2 ml) was added. From the or-
ange colored clear solution 4a crystallized within some
days. After 2 weeks these crystals were filtered off and
dried under vacuum.
Yield: 0.45 g (0.67 mmol; 64%) orange colored crys-
tals; M.p. decomposition (ca. 200 ꢁC).
Anal. Calc. for C30H25N5O9Ge (found, %): C 53.61
4.58 (s, 2, N–CH2–Ph); 6.95-7.75 (mm, 23, ar); 13C
NMR (101 MHz, CDCl3): 19.6 (C–CH3); 47.7; 55.6
(N–CH2–Ph); 88.5 (C@CH2); 119.5; 120.0; 122.2;
122.7; 124.4; 126.5; 126.6; 126.7; 127.2; 127.8; 128.1;
128.2; 128.3; 128.7; 129.5; 129.8; 129.9; 131.5; 134.7;
135.0; 137.4; 140.2; 145.0 (ar, C@CH2); 150.6; 150.7
(ar C–O); 168.2 (C@N); 29Si NMR: ꢀ55.2 (79 MHz,
CDCl3, 20 ꢁC); ꢀ108.2 (CP/MAS).
(53.57); H 3.75 (3.80); N 10.42 (10.48).
1H NMR (400 MHz, DMSO-d6): 2.80 (s, 6, CH3); 4.1
(m, 4; N–CH2CH2–N); 6.90–7.05 (m, 4, ar); 7.20–7.35
(m, 3, ar); 7.40 (m, 2, ar); 7.50 (m, 2, ar); 7.88 (dd, 2,
8a: In 20 ml thf, 7 (3.10 g; 5.62 mmol) was dissolved
at room temperature and a solution of benzoic acid
(0.69 g; 5.66 mmol) in 4 ml thf was added dropwise.
The resulting solution was warmed up to 50 ꢁC and hex-
ane (15 ml) was added. The solution was stored at 6 ꢁC
for 2 weeks to give crystals of 8a Æ thf. They were filtered
off, washed with thf/hexane 1:1 (5 ml) and dried under
vacuum.
3
4
ar, JH–H = 8.4 Hz, JH–H = 1.2 Hz); 8.58 (s, 2, picrate);
13C NMR (101 MHz, DMSO-d6): 20.1 (CH3); 46.2 (N–
CH2CH2–N); 118.7; 119.5; 121.7; 124.1; 125.1; 128.3;
129.9; 131.2; 132.5; 136.3; 141.8 (ar); 160.7. 161.6 (ar
C–O); 179.5 (C@N).
5: In 40 ml chloroform, 1a (3.00 g; 10.1 mmol) and
N,N,N0,N0-tetramethylnaphthaline-1,8-diamine (6.54 g;
30.5 mmol) were stirred at room temperature and phe-
nyltrichlorostannane (2.88 g; 9.54 mmol) was added
dropwise. From the resulting brownish clear solution
the product crystallized within some days. After 4 days
it was filtered off, washed with chloroform (10 ml), stir-
red in chloroform (20 ml) at 30 ꢁC for 1 h, filtered off
and dried under vacuum.
Yield: 3.55 g (4.76 mmol; 85%), nearly colorless (light
yellowish) crystals.
Crystals of 8a Æ 0.5p-xylene were obtained by recrys-
tallization from p-xylene. M.p. 116 ꢁC.
Anal. Calc. for C47H43N2O4Si (found, %): C 77.55
(76.50); H 5.95 (6.00); N 3.85 (3.86).
1H NMR (400 MHz, CDCl3): 2.21 (s, 6, N@C–CH3);
2
4.99; 5.34 (2 d, 4, N–CH2-ar, JH–H = 16.4 Hz); 6.30-
7.90 (mm, 28, ar); 13C NMR (101 MHz, CDCl3): 20.1
(CH3); 55.7 (CH2); 163.8 (ar C–O); 167.6 (COO);
175.9 (C@N).
Yield: 2.86 g (5.45 mmol; 57%) colorless crystals;
M.p. 248 ꢁC (under decomposition).
Anal. Calc. for C24H23N2O2SnCl (found, %): C 54.84
Due to a partial dissociation equilibrium of 8a in
solution1 and superposition of many 13C NMR signals,
more detailed 13C chemical shift values cannot be
provided.
(54.87); H 4.41 (4.32); N 5.33 (5.17).
1H NMR (400 MHz, DMSO-d6): 2.59 (s, 6, CH3);
4.10; 4.60 (2 m (broad), 4; N–CH2CH2–N); 6.60–7.60
(mm, 13, ar); 13C NMR (101 MHz, DMSO-d6): 20.5
(CH3); 49.3 (broad) (N–CH2CH2–N); 117.0; 121.7
(broad); 123.1; 127.1; 127.9; 130.9; 134.0; 134.4 (ar);
147.3 (broad) (ar C–Sn); 164.8 (ar C–O); 177.6 (broad)
(C@N); 119Sn NMR: ꢀ570.3 (149 MHz, CDCl3);
ꢀ578.1 (149 MHz, DMSO-d6).
29Si NMR: ꢀ183.4 (79 MHz, CDCl3); ꢀ179.3 (CP/
MAS of the p-xylene solvate).
8b: In 25 ml thf, 7 (1.6 g; 2.9 mmol) was dissolved and
a solution of HF (0.083 g HF 70% in pyridine = 2.91
mmol HF) in 3 ml thf was slowly added at 20 ꢁC. The
mixture was filtered immediately and the volume of
the filtrate was reduced under vacuum until a white pre-
cipitate was formed. After 2 h, 5 ml of hexane was added
and the suspension was stirred for 15 min. Then, the
product (8b Æ THF) was filtered off and dried under
vacuum.
7: To a stirred mixture of 6 (16.0 g; 71.1 mmol) and
triethylamine (20.0 g; 198 mmol) in thf (180 ml), phenyl-
trichlorosilane (7.60 g; 35.9 mmol) was added at room
temperature. After 4 h the precipitated triethylamine
hydrochloride was filtered off and washed with thf (20
ml). The filtrate was collected in a Schlenk flask and
the volatiles were removed under vacuum to give a yel-
low oil. This was dissolved in chloroform (5 ml). After
warming up to 50 ꢁC, hexane (25 ml) was added. The
resulting clear solution was stored at room temperature
for 1 day. The crystallized product was filtered off and
dried under vacuum. Within some days more product
crystallized from the filtrate.
Yield: 1.14 g (1.82 mmol; 61%).
Crystals of 8b Æ 0.5p-xylene were obtained by recrys-
tallization from p-xylene. M.p. 138 ꢁC.
Anal. Calc. for C40H38N2O2SiF (found, %): C 76.77
(76.60); H 6.12 (6.30); N 4.48 (4.49).
1H NMR (400 MHz, CDCl3): 2.28 (s, 6, N@C–CH3);
4.28; 5.36 (2 d, 4, N–CH2-ar, JH–H = 15.6 Hz); 5.54 (d,
2
3
2, JHH = 8.4 Hz); 6.60–7.60 (mm, 21, ar); 13C NMR
Yield (first fraction): 5.80 g (10.5 mmol; 30%), orange
colored crystals, M.p. 114 ꢁC.
Anal. Calc. for C36H32N2O2Si (found, %): C 78.23
(78.12); H 5.84 (6.05); N 5.07 (4.92).
1H NMR (400 MHz, CDCl3): 2.18 (s, 3, C–CH3);
4.02; 4.73 (2 s, 2, C@CH2); 4.40 (m, 2, N–CH2–Ph);
(101 MHz, CDCl3): 19.5 (CH3); 55.5 (CH2); 117.6;
121.6; 122.4; 125.7; 126.3; 126.6 (2·); 128.6; 128.9;
133.8; 136.5; 138.7; 156.3 (ar); 162.6 (ar C–O); 175.2
(C@N); 29Si NMR: ꢀ180.9 (1JSiF = 175.7 Hz) (79
MHz, CDCl3); ꢀ181.0 (CP/MAS of the p-xylene
solvate).

J. Wagler et al. / Inorganica Chimica Acta 358 (2005) 4270–4286
4275
12: (Product of the reaction of 10a-BF4 with 2a, fol-
lowed by hydrolysis).
C17
C8
C7
A solution of 2a (330 mg; 0.828 mmol) and 10a-BF4
(440 mg; 0.748 mmol) in 50 ml of methylene chloride
was stirred for 6 h at room temperature. Concentrated
hydrochloric acid (2 ml) was added and the reaction
mixture was stirred for 2 d at room temperature. The or-
ganic layer was then washed with water (150 ml), con-
centrated NaHCO3-solution (150 ml), water (150 ml),
and was dried over CaCl2. After evaporation of the sol-
vent in vacuum, the residue was purified by column
chromatography (SiO2, CHCl3).
C16
N1
N2
Si1
O1
O2
C19
Yield: 390 mg (0.612 mmol; 82%), nearly colorless
crystals.
Fig. 1. Molecular structure of 2b in the crystal (ORTEP plot, 50%
probability ellipsoids, hydrogen atoms omitted).
Anal. Calc. for C45H36N2O2 (found, %): C 84.88
(84.58); H 5.70 (6.01); N 4.40 (4.19).
3
1H NMR (300 MHz, CDCl3): 3.69 (d, 2, JH–H = 7.5
3
Hz, Ar2CH–CH2–); 4.67 (t, 1, JH–H = 7.3 Hz, Ar2CH–
C17
CH2–); 6.88–7.79 (m, 32, ar).
13C NMR (76 MHz, CDCl3): 44.5 (Ar2CH-CH2–);
45.1 (Ar2CH–CH2–); 118.6; 118.9; 119.6; 122.7; 123.9;
124.2; 128.5; 129.2; 129.9; 136.4; 137.9; 146.2; 147.8;
162.5 (ar); 204.5 (C@O).
C16
N2
C8
C7
N1
Si1
O1
O2
3. Results and discussion
C19
Three tetradentate salen-type ligands (1a–1c) with a
di-, tri- and tetramethylene-spacer between the 2-
oxyacetophenoneimine moieties, respectively, were com-
bined with phenyltrichlorosilane. Triethylamine was
used as a supporting base to trap HCl (Scheme 2). The
formation of 2a was previously reported [13,14]. Com-
pounds 1b and 1c reacted in a similar manner to give
the pentacoordinate silicon complexes 2b and 2c with
enamine functionalized salen-type ligands, respectively.
Both complexes were characterized by 1H, 13C and
29Si NMR spectroscopy as well as by X-ray crystal
structure analysis (Figs. 1 and 2, Table 1).
While the trigonal bipyramidal coordination sphere
around the Si atom in 2a is considerably distorted
(60% TBP); {%TBP = 100% · [(axial angle ꢀ equatorial
angle)/60]} [8a], the donor atoms in 2b and 2c are rather
well trigonal bipyramidally arranged (93% TBP in 2b,
96% TBP in 2c). The 29Si NMR spectra of 2b and 2c also
Fig. 2. Molecular structure of 2c in the crystal (ORTEP plot, 50%
probability ellipsoids, hydrogen atoms omitted).
Table 1
Selected bond lengths (A) and bond angles (ꢁ) as well as 29Si NMR
shift values (ppm) for 2a, 2b and 2c
˚
Compound
2a
2b
2c
Si1–N1
Si1–N2
Si1–O1
Si1–O2
Si1–C19
C16@C17
N1@C7
2.010(3)
1.756(3)
1.684(2)
1.720(2)
1.871(3)
1.340(6)
1.289(4)
2.077(2)
1.749(2)
1.665(1)
1.727(2)
1.876(2)
1.333(3)
1.292(3)
2.055(1)
1.756(1)
1.664(1)
1.733(1)
1.888(1)
1.351(2)
1.294(2)
N1–Si1–O2
C19–Si1–N2
C19–Si1–O1
N2–Si1–O1
167.9(1)
116.0(2)
111.3(1)
131.8(1)
175.95(8)
124.52(9)
115.03(8)
119.89(8)
176.60(5)
124.76(6)
115.44(6)
119.16(5)
PhSiCl3
3 Et3N
d29Si (CDCl3)
ꢀ115.8
ꢀ116.1
ꢀ107.7
ꢀ107.3
ꢀ106.4
ꢀ109.6
n
n
N
d29Si (CP/MAS)
N
N
N
O
Si
- 3 Et3NHCl
OH HO
O
Ph
1a: n = 2
1b: n = 3
1c: n = 4
reveal pentacoordination of the silicon atoms, in solid
state as well as in solution. However, the Si nuclei in
2b and 2c are much less shielded than in 2a, probably
due to the longer Si–N dative bond (Si1–N1) in 2b
2a: n = 2
2b: n = 3
2c: n = 4
Scheme 2.

4276
J. Wagler et al. / Inorganica Chimica Acta 358 (2005) 4270–4286
and 2c. Because of the more flexible tri- and tetrameth-
ylene spacers (compared with the ethylene group in the
ligand system of 2a), a trigonal bipyramidal arrange-
ment of the five donor atoms is more easily accessible
in 2b and 2c, and the donor atom N1 is not forced to
coordinate with the silicon atom in such a short distance
as in 2a. However, the Si1–N1 dative bonds in 2b and 2c
have similar lengths, indicating an energetic optimum
for this dative bonding situation.
about 14 ppm is covered by those compounds. Some
influences on the shielding of the Si nucleus can be de-
rived from the molecular structures of 3a–3c (Figs. 3–
5, Table 2).
As previously reported, the enamine complex 2a eas-
ily reacts with Brønsted acids in a 1,4-addition reaction
to yield hexacoordinate silicon complexes (Scheme 3)
[13–15]. Several novel complexes of this type have now
been found (3c–3h). The hexacoordinate Si compounds
were mainly characterized by multinuclear NMR spec-
troscopy in chloroform solution. In some cases
(3f–3h), because of the poor solubility of the complexes,
29Si solid-state NMR spectroscopy was applied
(Table 2). All complexes 3a–3h involve a hexacoordinate
silicon atom. Although only one monodentate donor li-
gand was varied, a large 29Si chemical shift range of
C25
O3
N2
N1
Si1
O1
O2
C19
N
O
N
O
Si
3a: X = benzoate
Ph
2a
Fig. 3. Molecular structure of 3a in the crystal (ORTEP plot, 50%
probability ellipsoids, hydrogen atoms and p-xylene molecule omitted).
3b: X = picrate
3c: X = 8-oxyquinolinate
3d: X = 2-oxy-1,4-naphthoquinonate
3e: X = p-tert-butylphenolate
3f: X = 5-phenyl-2-tetrazolide
3g: X = F-
HX
X
N
O
N
O
-
Si
3h: X = BF4
Ph
3a - 3h
Scheme 3.
C25
O3
Table 2
Selected bond lengths (A) and bond angles (ꢁ) as well as 29Si NMR
shifts of 3a, 3b, 3c and 3i
˚
O2
O1
Compound
3a
3b
3c
3i
Si1
Si1–N1
Si1–N2
Si1–O1
Si1–O2
Si1–O3
Si1–C19
1.932(2)
1.928(2)
1.733(1)
1.735(1)
1.821(1)
1.948(2)
1.925(2)
1.919(2)
1.729(2)
1.727(2)
1.900(2)
1.911(3)
1.969(2)
1.943(2)
1.722(1)
1.735(1)
1.779(1)
1.949(2)
2.001(2)
2.005(2)
1.754(2)
1.746(2)
1.792(2)
1.930(3)
N2
N1
C19
Si1–O3–C25
N1–Si1–N2
N1–Si1–O1
N2–Si1–O2
O3–Si1–C19
133.6(1)
83.85(7)
92.28(7)
91.60(7)
179.55(8)
124.8(1)
83.47(9)
91.77(8)
92.05(8)
174.1(1)
129.9(1)
84.93(7)
92.48(7)
92.37(7)
170.48(7)
140.3(2)
91.4(1)
88.18(9)
87.74(9)
177.9(1)
Fig. 4. Molecular structure of 3b in the crystal (ORTEP plot, 50%
probability ellipsoids, hydrogen atoms and dioxane molecules
omitted).
d29Si (solid)
ꢀ186.0
ꢀ172.0
d29Si (CDCl3)
ꢀ181.3
ꢀ184.2

J. Wagler et al. / Inorganica Chimica Acta 358 (2005) 4270–4286
4277
oxygen atom of the added anion acts as the donor atom
since there are two resonance structures of the ligand
giving rise to possible isomers (Scheme 4). Unfortu-
nately, we were not able to determine the crystal struc-
ture of 3d up to now.
N3
Compound 3e is not soluble in most organic solvents
and was, therefore, characterized by 29Si solid-state
NMR spectroscopy. The chemical shift value of
ꢀ183.1 ppm clearly indicates hexacoordination of the
Si atom. This value differs only slightly from those of
3c and 3d. Similar coordination spheres around the sili-
con atoms (salen-type ligand + phenyl group + phenolic
oxygen atom) can be considered as the main reason.
In addition to the Si–O bond formation, the proton-
ation of the terminal enamine carbon atom by the
Brønsted acid contributes to the driving force of the addi-
tion process of Brønsted acids to 2a. Thus, the addition of
5-phenyltetrazole to 2a also yields a product with hexaco-
ordinate Si atom, 3f. Single crystals of 3f were obtained
from dioxane. Its molecular structure is given in Fig. 6.
The silicon atom of 3f is almost octahedrally sur-
rounded by the six donor atoms. The Si–N bond Si1–
C25
O3
N2
Si1
N1
O2
O1
C19
Fig. 5. Molecular structure of 3c in the crystal (ORTEP plot, 50%
probability ellipsoids, hydrogen atoms and thf molecule omitted).
˚
Due to the restricted coordination sphere within the
tetradentate ligand system, bond lengths from the silicon
atom to the monodentate donor groups (Si1–O3, Si1–
C19) vary much more than those to the donor atoms
of the tetradentate ligand. As mentioned for 2a–2c,
bond lengths have an important influence on the shield-
ing of the Si nucleus. Thus, the weaker shielding in 3b
compared to 3a might be explained by the elongation
N4 (2.00 A) is significantly longer than the Si–N dative
bonds of the tetradentate salen-type ligand. Thus, the
resonance structures of Scheme 5, top, demonstrate
the bonding situation in complex 3f.
Up to now, only one molecule (I, Scheme 5, bottom)
with a Si bonded tetrazole nitrogen atom has been char-
acterized by X-ray crystal structure analysis [17].
In this molecule (I), the Si–N bond is remarkably
˚
˚
of the Si–O bond (Si1–O3 = 1.90 A). However, other ef-
fects must be taken into account to explain the stronger
shielding in 3a than in 3c, although 3c exhibits a much
long (1.98 A) although the Si atom is tetracoordinate.
A significant stretching of Si–N bonds (N = donor atom
from N-methylimidazole) is usually observed for
increasing coordination number of the Si atom [18].
Therefore, it is surprising that the Si–N(tetrazole) bond
in 3f (with hexacoordinate Si atom) is not much longer
than in the tetracoordinate silicon compound (I). In
comparison with the complexes 3a and 3c, however,
the poor donor action of the phenyltetrazole becomes
obvious. In 3f the Si–C bond is significantly shorter,
similar to that in 3b, which also incorporates a poor do-
nor anion.
The formation of a Si–F bond is highly favorable,
thus 3g was readily obtained by addition of HF to 2a.
Even the addition of HBF4 to 2a and the coordination
of the tetrafluoroborate anion to yield 3h was successful.
Unfortunately, both complexes 3g and 3h are insoluble
˚
shorter Si–O bond (Si1–O3 = 1.78 A) than 3a (Si1–
O3 = 1.82 A). In all three cases, the electronic situation
˚
of the donor atoms O3 can be interpreted to be sp2-
hybridized, referring to the O atom attachment at p-
electron systems as well as to the measured bond angles.
Silicon atoms in electronegatively substituted tetra- and
hypercoordinate silicon compounds are reported to
interact with additional lone pairs of the donor atoms
connected to them (p-donation) [16]. Taking this into
account, an effective O ! Si p-donation is possible in
3a, but not in 3b and 3c. In those compounds, the p-
orbital at the sp2-hybridized atom O3 could be either
in conjugation with the aromatic p-electron system of
the monodentate ligand or in p-interaction with the sil-
icon atom. In all three cases, the anion of the added
Brønsted acid coordinates at the Si atom as a monoden-
tate ligand although further donor atoms are still avail-
able (carbonyl oxygen atom of the benzoate, oxygen
atoms of the picrate anion’s nitro groups, nitrogen
atom of the 8-oxyquinolinate). Even in compound 3d,
the 2-oxy-1,4-naphthoquinonate anion may act as a che-
lating ligand, but the silicon atom remains hexacoordi-
nate. In the case of complex 3d, it is not clear which
O
O
O
O
O
O
[Si]
[Si]
Scheme 4.

4278
J. Wagler et al. / Inorganica Chimica Acta 358 (2005) 4270–4286
N1
N6
N5
O1
Si1
O2
C19
N4
N2
N3
-170
-180
-190
(ppm)
Fig. 7. 29Si CP/MAS NMR spectrum (isotropic chemical shift signals)
of 3g (top left), 3h (crude mixture of 3h and 3g, top right) and 3h
(bottom, calculated from the crude spectrum by substracting the signal
of 3g, middle).
Fig. 6. Molecular structure of 3f in the crystal (ORTEP plot, 50%
probability ellipsoids, hydrogen atoms omitted). Selected bond lengths
˚
(A) and bond angles (ꢁ): Si1–O1: 1.710(2), Si1–O2: 1.724(2), Si1–N1:
1.916(2), Si1–N2: 1.928(2), Si1–N4: 1.995(2), Si1–C19: 1.929(2); C19–
Si1–N4: 176.31(8).
peaks: 160 Hz). Both effects result from the poor donor
ꢀ
ability of the BF4 anion (compared with fluoride).
Ph
Ph
Unfortunately, 3h was not obtained as a pure substance.
The used solvent (thf) is a competitive donor for the BF3
moiety. Thus, BF3 can be eliminated from the silicon-
coordinated fluoride anion.
Analogous acid additions were realized with the
enamines 2b and 2c. The preparation of 3i by adding
benzoic acid to 2b is a representative example. The
molecular structure of 3i (Fig. 8) is similar to the one
of 3a; the Si atom is situated in a distorted octahedral
coordination sphere.
The dative Si–N bonds in 3i are significantly longer
than in 3a (Table 2). As mentioned above (comparison
between 2a, 2b, 2c), it can be concluded that the imine
nitrogen atoms and even the phenolic oxygen donor
atoms of the tetradentate salen-type ligand are forced
to coordinate the Si atom in a very short distance in case
N
N
N
N
N
N
N
N
N
N
O
N
N
O
Si
Si
O
O
Ph
Ph
3f
N
N
N
Me
Me
Me
N
Si
PhMe2Si
SiMe2Ph
SiMe2Ph
(I)
Scheme 5.
in almost all organic solvents, and the preparation of
crystals for structure analysis failed. The coordination
of one fluorine atom at the hexacoordinate Si atom
was only proven by 29Si solid-state NMR spectroscopy
(Fig. 7).
1
In both spectra, doublets account for the JSi–F cou-
N1
pling. The spectrum of 3h contains the signal of 3g
which occurs as a shoulder. The spectrum of pure 3h
was obtained by fitting and substracting the signal of
3g. Although the difference between the two peaks of
each signal does not exactly represent the 1JSi–F coupling
constant, a rough estimate can be made. Compared with
the 1JSi–F coupling constant of a related complex (silicon
atom hexacoordinate by a salen-type ligand plus two F
atoms), the difference (174 Hz) is close to the expected
1JSi–F value (179.2 Hz in the difluorosilicon complex
with the same ligand [19]). The spectrum of 3h indicates
a less shielded hexacoordinate Si nucleus as well as a
O1
C19
C25
Si1
O2
O3
N2
Fig. 8. Molecular structure of 3i in the crystal (ORTEP plot, 50%
probability ellipsoids, hydrogen atoms and thf molecule omitted).
1
smaller JSi–F coupling constant (difference between

J. Wagler et al. / Inorganica Chimica Acta 358 (2005) 4270–4286
4279
of ligand 1a. Accompanied with the lengthening of the
bonds to the salen-type ligand, the bonds Si1–C19 and
Si1–O3 to the monodentate donor ligands are signifi-
cantly shortened. This effect was not observed in case
of the pentacoordinate silicon complexes 2a–2c.
Phenyltrichlorogermane and 1a react in a similar
manner to yield an enamine functionalized pentacoordi-
nate germanium complex 4. The molecular structure of 4
(Fig. 9) is similar to that of 2a. Compound 4 also crys-
tallizes in the non-centrosymmetric monoclinic space
group P21.
O5
C19
C17
N2
C16
C8
N1
C7
Due to the bigger germanium atom, the trigonal bipy-
ramidal coordination sphere of 4 is more distorted than
that of 2a (60% TBP in 2a, 51% TBP in 4). In contrast to
the behavior of 2a, the reaction of 4 with picric acid
yields the ionic pentacoordinate germonium picrate 4a
(Scheme 6, Fig. 10).
Ge1
O1
O2
Fig. 10. Molecular structure of 4a in the crystal (ORTEP plot, 50%
probability ellipsoids, hydrogen atoms omitted).
The addition of acids to the silicon enamine complex
2a was already proposed to occur via a pentacoordinate
siliconium ion, which was detected by ESI-MS [13,14].
In case of compound 4a, however, the onium ion formed
by protonation of the terminal enamine carbon atom is
stable enough to yield an isolable germonium salt. The
Ge atom of 4a is situated in a distorted square pyrami-
dal coordination sphere (25% TBP). Unlike in 4, the
Si–N and Si–O bonds in 4a have similar lengths. Be-
cause of the symmetry of the tetradentate salen-type li-
gand in 4a, the square pyramidal arrangement of the
donor atoms around the central atom should be pre-
ferred. However, in case of pentacoordinate siliconium
cations with a tetradentate salen-type ligand the Si atom
can be situated in a trigonal bipyramidal coordination
sphere with Si–N dative bonds of different lengths [20].
The coordination of the picrate ion at the Si atom in
3b and the ionic character of 4a result in a significantly
different molecular structure of the picrate ion in these
two compounds. Selected bond lengths of the picrate
fragments of 3b and 4a, respectively, are given in
Scheme 7.
C17
C8
N2
O2
N1
C16
C7
Ge1
O1
C19
Fig. 9. Molecular structure of 4 in the crystal (ORTEP plot, 50%
probability ellipsoids, hydrogen atoms omitted).
The picrate ion in 4a has a much shorter C–O bond to
the phenolic oxygen atom than in 3b. The C–C bonds of
N
N
O
2N
C
A: 1.289
B: 1.418
C: 1.375
D: 1.373
E: 1.463
B
A
D
OH HO
O
E NO2
1a
PhGeCl3
3 Et3N
O2N
- 3 Et3NHCl
O2N
2b
O
NO2
O2N
picric acid
C
N
O
N
O
A: 1.237
B: 1.458
C: 1.370
D: 1.388
E: 1.432
B
A
D
O2N
N
Ge
O
E NO2
Ph
4
N
O
Ge
O2N
O
Ph
4a
4a
Scheme 6.
Scheme 7.

4280
J. Wagler et al. / Inorganica Chimica Acta 358 (2005) 4270–4286
its aromatic ring alternate in bond length. Thus, the
quinoid resonance structure can be drawn. Due to the
coordination of the phenolic oxygen atom of the picrate
ion at the Si atom in 3b, the C–O bond is stretched and
the alternation of the C–C bond lengths in the aromatic
ring is reduced. Therefore, the aromatic resonance struc-
tures of the picrate ion in 3b are more significant than in
4a.
The molecular structures of 3b and 4a clearly differen-
tiate the reactivities of the onium ions formed by pro-
tonation of the enamines 2a and 4, respectively.
Contrasting the formation of enamine complexes with
phenyltrichlorosilane and -germane, the reaction of phe-
nyltrichlorostannane with 1a yields a hexacoordinate tin
complex 5. Even with a proton sponge, a very strong
supporting base, the remaining chlorine atom could
not be eliminated from the tin atom to give rise to an en-
amine functionalized Sn complex. The molecular struc-
ture of 5 is given in Fig. 11.
The tin atom in 5 is octahedrally surrounded by the
donor atoms. Surprisingly, the salen-type ligand in this
compound is mer–fac-arranged. This results in a cis-con-
figuration of the two monodentate donors – the phenyl
group and the chlorine atom. Previously, several molec-
ular structures of salen-Sn-complexes with two monod-
entate ligands have been published, but in all cases
there was an equatorial arrangement of the salen-type
ligand and trans-configuration of the monodentate li-
gands [21]. Up to now, only one hexacoordinate salen-
Sn-complex – with a bidentate chelating ligand – was
reported to have a mer–fac-configuration of its tetraden-
tate salen ligand moiety [22]. In solution the Sn atom is
also hexacoordinate (d119Sn = ꢀ 570.3 ppm in CDCl3,
d119Sn = ꢀ 578.1 ppm in DMSO-d6). Unlike the mer–
fac arrangement of the tetradentate ligand sytem in solid
state, the 1H NMR-spectra of 5 in CDCl3 show only one
broad singlet for the –CH3– and –CH2– groups of the
salen-type ligand, respectively. This indicates a configu-
rational fluctuation of molecule 5 in solution.
Instead of tetradentate salen-type ligands, two biden-
tate 2-iminomethylphenolate ligands (6) can also be ap-
plied for preparing enamine functionalized silicon
complexes (Scheme 8).
As proven by 29Si CP/MAS NMR spectroscopy,
compound 7 has a pentacoordinate Si atom in solid state
(d29Si = ꢀ 108.2 ppm). In contrast, a solution of 7 in
CDCl3 gives rise to a 29Si NMR signal at ꢀ55.2 ppm.
Its location is typical for a tetracoordinate silicon com-
pound. The dative Si–N bond in 7 is not maintained in
CDCl3 solution at room temperature. It should be men-
tioned that a strong temperature dependence of this
Si N coordination has recently been proven for a
PhSiCl3
3 Et3N
Bz
Bz
N
N
O
O
2
Si
- 3 Et3NHCl
OH
N
Ph
Bz
6
7
HX
Bz
X
O
N
N
O
8a: X = benzoate
8b: X = F
Si
Ph
Bz
Scheme 8.
C17
C8
C16
C7
N1
N2
O3
O2
O4
Cl1
N1
Si1
O2
N2
O1
Sn1
C81
C19
O1
Fig. 11. Molecular structure of 5 in the crystal (ORTEP plot, 50%
probability ellipsoids, hydrogen atoms omitted).
Fig. 12. Molecular structure of 8a in the crystal (ORTEP plot, 50%
probability ellipsoids, hydrogen atoms and p-xylene molecule omitted).

J. Wagler et al. / Inorganica Chimica Acta 358 (2005) 4270–4286
4281
Table 3
Selected bond lengths (A) and bond angles (ꢁ) of 8a and 8b
˚
Compound
8a
8b
Si1–N1
Si1–N2
1.970(2)
1.933(2)
1.748(2)
1.748(2)
1.833(2)
1.964(1)
1.964(1)
1.7524(9)
1.7558(9)
Si1–O1
Si1–O2
F1
N2
O1
O2
Si1–O3
Si1–F1
Si1–C71 or –C81
1.6767(9)
1.939(2)
Si1
C71
1.925(2)
N1
N1–Si1–N2
O1–Si1–O2
C81–Si1–O3
C71–Si1–F1
177.23(8)
175.86(8)
173.80(9)
177.26(5)
178.77(5)
179.48(5)
Fig. 13. Molecular structure of 8b in the crystal (ORTEP plot, 50%
probability ellipsoids, hydrogen atoms and p-xylene molecule omitted).
Table 4
Rate constants for the reactions of enamine complexes 2a, 4 and
enamine 9 with benzhydrylium ions 10a, b
M
NR2
k2 (Mꢀ1 sꢀ1
)
krel
175
Si
Ge
2a
4
NPh2
10a
10b
2.38 · 106
7.36 · 104
N
Ge
4
9
NPh2
NPh2
10a
10a
1.75 · 107a
1.36 · 104b
1287
1
O
N
Ph
a
Calculated.
From [30].
b
BF4
R2N
NR2
NR2
H
R2N
10a: NR2 = NPh2
10b: NR2 =
N
k
BF4
N
N
O
CH2Cl2, 20˚C
M
O
Ph
N
O
N
O
-80
-120
-160
-200
-240
-320
(ppm)
M
11
Fig. 14. Experimental (top) and calculated (bottom) 29Si CP/MAS
HCl / H2O
Ph
for M = Si, R = Ph
NMR spectrum of 8b at a spinning rate of mspin = 600 Hz.
2a, 4
NR2
similar compound [23]. Complex 7 also reacts with
Brønsted acids (Scheme 8). The addition of benzoic acid
or hydrogen fluoride yielded 8a and 8b, respectively.
Both complexes were characterized by single crystal
X-ray structure analysis (Figs. 12 and 13, Table 3).
In both complexes, the Si atom is octahedrally
coordinated by the donor atoms of the bidentate che-
lating ligand moieties in equatorial positions and the
R2N
[PhSiO1.5
]
O
O
NH3
HO
HO
H3N
12
Scheme 9.

4282
J. Wagler et al. / Inorganica Chimica Acta 358 (2005) 4270–4286
two monodentate substituents in axial positions. Un-
like in complex 3a and related compounds, the oxygen
atoms as well as the nitrogen atoms of the chelating
ligands are trans situated in 8a and 8b. As described
for 3a and 3i, the carbonyl oxygen atom does not
coordinate at the Si atom in 8a. Compounds 8a and
8b have hexacoordinate Si atoms in the solid state
and in CDCl3 solution. The 29Si NMR shifts of solid
state and solution NMR spectra are comparable and
therefore indicate similar geometries of the coordina-
tion sphere around the silicon atom in solution and
in the solid state. The 29Si isotropic chemical shift va-
lue for 8b as well as the values of the principal compo-
nents of the chemical shift tensor were determined by
modeling the solid-state NMR spectrum with CSO-
The broad signal shape mainly results from the direct
dipolar interaction with the Si–F bond. The principal
components of the 29Si chemical shift tensor
(d11 = ꢀ 148, d22 = ꢀ 160, d33 = ꢀ 226) indicate a small
anisotropy of this solid-state NMR signal (X = 78
ppm). Silicon atoms of divalent silylenes [25] and triva-
lent silenes [26] as well as Si atoms in pentacoordinate
silicon compounds such as silatranes [27], pentacoordi-
nate zwitterionic silicon bis-chelates [28] or siliconium
ions with salen-type ligand [20] have 29Si solid-state
NMR signals with a significantly larger span.
For comparison with hexacoordinate silicon com-
pounds, the 29Si chemical shift tensors were also deter-
mined for 3a (X = 33 ppm) and 8a (X = 42 ppm). They
indicate an even smaller anisotropy of the 29Si solid-
state NMR signal in these two complexes than in 8b.
The modeling of the 29Si CP/MAS NMR spectrum of
8b was successful only for coaxially featured tensors of
both the shielding and the dipolar coupling. Because
1
LIDS (Fig. 14) [24]. The JSiF coupling constant (182
Hz) was available from solution NMR and the dipolar
coupling (D = ꢀ 4801 Hz) was calculated from the
Si–F bond length.
Table 5
Crystal data and structure refinement for 2b, 2c and 3a
Compound
2b
2c
3a Æ 5/8xylene
Empirical formula
Formula weight
T (K)
C
25H24N2O2Si
C26H26N2O2Si
426.58
198(2)
C36.5H34.875N2O4Si
593.90
198(2)
412.55
198(2)
0.71073
monoclinic
P21
˚
k (A)
0.71073
triclinic
0.71073
monoclinic
C2/c
Crystal system
Space group
Unit cell dimensions
ꢀ
P1
˚
a (A)
9.3330(7)
11.7480(7)
8.3260(7)
8.6820(7)
16.5090(10)
77.871(7)
84.723(5)
66.616(6)
1070.86(14)
2
24.473(5)
12.1703(12)
˚
b (A)
˚
c (A)
10.1150(12)
90
21.020(2)
90
a (ꢁ)
b (ꢁ)
c (ꢁ)
112.354(7)
90
1025.71(16)
99.892(12)
90
6167.7(15)
3
˚
V (A )
Z
2
1.336
8
1.279
Dcalc (Mg mꢀ3
)
1.323
0.136
452
Absorption coefficient (mmꢀ1
F(000)
)
0.140
436
0.119
2511
Crystal size (mm3)
0.40 · 0.21 · 0.15
5.03–24.99
0.48 · 0.32 · 0.20
4.83–27.50
ꢀ10 6 h 6 10,
ꢀ11 6 k 6 11, ꢀ21 6 l 6 21
33965
0.35 · 0.30 · 0.20
3.61–25.00
h Range for data collection (ꢁ)
Index ranges
ꢀ11 6 h 6 11,
ꢀ13 6 k 6 13, ꢀ12 6 l 6 11
17671
ꢀ29 6 h 6 29,
ꢀ14 6 k 6 14, ꢀ24 6 l 6 24
47755
Number of reflections
collected
Number of independent
3561 (0.1076)
4812 (0.0519)
5411 (0.0497)
reflections (Rint
)
Completeness to hmax (%)
Absorption correction
Maximum and minimum
transmissions
99.0
none
98.0
none
99.7
semi-empirical
0.9765 and 0.9594
Refinement method
full-matrix least-squares
on F2
3561/1/272
full-matrix least-squares on F2
full-matrix least-squares on F2
Data/restraints/parameters
Final R indices [I > 2r(I)]
R indices (all data)
Goodness-of-fit on F2
4812/0/281
5411/0/441
R1 = 0.0326, wR2 = 0.0672
R1 = 0.0480, wR2 = 0.0710
1.010
R1 = 0.0353, wR2 = 0.0855
R1 = 0.0523, wR2 = 0.0928
1.056
R1 = 0.0428, wR2 = 0.0984
R1 = 0.0675, wR2 = 0.1070
1.033
Largest difference in
peak and hole (e A
0.139 and ꢀ0.184
0.302 and ꢀ0.310
0.314 and ꢀ0.279
ꢀ3
˚
)

J. Wagler et al. / Inorganica Chimica Acta 358 (2005) 4270–4286
4283
of the Si–F bond one principal component of the dipolar
coupling tensor should be situated on this bond axis.
The strong axiality of the chemical shift tensor leads
to the assumption that the principal component of the
chemical shift tensor d33 (ꢀ226 ppm) is also aligned
along the Si–F bond and the components d11 and d22
are situated in the equatorial plane between the Si–O
and Si–N bonds. This assumption is supported by the
fact that the principal components of the 29Si chemical
shift tensor of the similarly shaped complex 8a
(d11 = ꢀ 160.4, d22 = ꢀ 175.1, d33 = ꢀ 202.4) are similar
to those of 8b except d33. This value differs by about
23 ppm between both complexes and should be caused
by the main structural difference between both com-
pounds – the bonds to the different monodentate donor
moieties Si1–F1 and Si1–O3.
For that purpose, we studied the rates of the reac-
tions of compounds 2a and 4 with benzhydrylium ions
10a, b using the same photometric method previously
employed for determining the nucleophilic reactivities
of a large variety of enamines [29].
Since the reaction of 4 with the N,N-diphenylamino
substituted benzhydrylium ion 10a (E = ꢀ4.72) [30]
was too fast for direct measurements, we have deter-
mined the rate constant of the reaction of 4 with the less
electrophilic bis-4(N-pyrrolidino)benzhydrylium ion 10b
(E = ꢀ 7.69) [30]. This value can easily be converted into
a rate constant for the reaction of 4 with 10a, employing
the equation [logk = s(E + N)] and s = 0.8, the typical
value for enamines [29].
The last column in Table 4 shows that the enamine
complexes 2a and 4 are considerably more nucleophilic
than the non-coordinated model compound 9. One can
assume that this reactivity difference is predominantly
due to complexation and not due to the extra oxygen
in ortho-position of the phenyl ring in 2a and 4, because
we have previously shown that the nucleophilicities of
Compounds 2a and 4 are structurally analogous to a-
(N-morpholino)styrene (enamine 9), and the question ar-
ose as to how substitution of the enamine nitrogen by the
pentacoordinate silicon or germanium group is affecting
the nucleophilicity of the enamine double bond.
Table 6
Crystal data and structure refinement for 3b, 3c and 3f
Compound
3b Æ 2dioxane
38H41N5O13Si
3c Æ thf
3f
Empirical formula
Formula weight
T (K)
C
C37H37N3O4Si
615.79
198(2)
C31H28N6O2Si
544.68
93(2)
803.85
198(2)
˚
k (A)
0.71073
triclinic
0.71073
monoclinic
P21/n
0.71073
monoclinic
P21/n
Crystal system
Space group
Unit cell dimensions
ꢀ
P1
˚
a (A)
12.1140(10)
12.3400(10)
13.3640(10)
107.750(10)
97.480(10)
90.700(10)
1883.6(3)
12.219(2)
16.685(3)
13.3869(12)
11.8375(10)
˚
b (A)
˚
c (A)
15.700(3)
90
17.4122(15)
90
a (ꢁ)
b (ꢁ)
c (ꢁ)
104.19(3)
90
3103.1(11)
91.756(4)
90
2758.0(4)
3
˚
V (A )
Z
2
1.417
4
1.318
4
1.312
Dcalc (Mg mꢀ3
)
Absorption coefficient (mmꢀ1
F(000)
Crystal size (mm3)
)
0.137
844
0.122
1304
0.126
1144
0.55 · 0.24 · 0.23
3.55–25.00
ꢀ14 6 h 6 14,
ꢀ14 6 k 6 14, ꢀ15 6 l 6 15
44074
0.56 · 0.21 · 0.18
3.59–25.00
0.25 · 0.15 · 0.05
1.89–25.00
h Range for data collection (ꢁ)
Index ranges
ꢀ14 6 h 6 14,
ꢀ19 6 k 6 19, ꢀ18 6 l 6 18
54787
ꢀ15 6 h 6 15,
ꢀ14 6 k 6 13, ꢀ20 6 l 6 20
23183
Number of reflections collected
Number of independent
6611 (0.0461)
5456 (0.0889)
4840 (0.0711)
reflections (Rint
)
Completeness to hmax (%)
Absorption correction
Maximum and minimum
transmissions
99.7
semi-empirical
0.9688 and 0.9286
99.7
none
100.0
none
Refinement method
full-matrix least-squares on F2
6611/40/571
full-matrix least-squares on F2
5456/0/454
full-matrix least-squares on F2
4840/0/361
Data/restraints/parameters
Final R indices [I > 2r(I)]
R indices (all data)
Goodness-of-fit on F2
R1 = 0.0494, wR2 = 0.1229
R1 = 0.0799, wR2 = 0.1361
1.052
R1 = 0.0400, wR2 = 0.0932
R1 = 0.0617, wR2 = 0.1022
1.043
R1 = 0.0449, wR2 = 0.0993
R1 = 0.0815, wR2 = 0.1094
1.015
Largest difference in
peak and hole (e A
0.300 and ꢀ0.325
0.272 and ꢀ0.262
0.244 and ꢀ0.267
ꢀ3
˚
)

4284
J. Wagler et al. / Inorganica Chimica Acta 358 (2005) 4270–4286
1,1-disubstituted ethylenes, which carry a weak and
strong electron donor, are hardly affected when the elec-
tron-donating ability of the weakly donating substituent
is modified [31,32].
In accord with this interpretation, a remarkable
change of nucleophilicity is observed when the central
atom is exchanged, and the germanium complex 4 is 7
times more reactive than the analogous silicon complex
2a.
It should be noted that the intermediates 11, which
are produced from the reactions of 2a and 4 with 10-
BF4 (Scheme 9), have not been isolated. In order to cor-
roborate the course of the reaction, we have isolated
compound 12, however, which was obtained by hydroly-
sis of 11.
and an amine base to yield pentacoordinate enamine
functionalized silicon complexes. Such enamine com-
plexes are established with this work as starting points
for switching of coordination mode from penta- to hexa-
coordinate silicon complexes. The influence of the
substituent R (oligomethylene bridge providing for tet-
radentate ligands, alkyl group giving rise to bidentate
chelating ligands) on the coordination behavior of the li-
gand in the Si coordination sphere was shown with
examples of X-ray structures as well as NMR spectro-
scopic results pointing out selected coordination–
reactivity-relationships.
The investigation of the different reactivities of
silicon, germanium and tin derivatives on this field re-
sulted in the first X-ray structurally confirmed hexa-
coordinate salen-tin-complex with two cis-featured
monodentate donors as well as the first X-ray structure
of a cationic pentacoordinate germonium salt with
salen-type ligand. The enamine moieties of herein pre-
sented pentacoordinate silicon and germanium enamine
complexes were found to have remarkably upscaled
4. Conclusions
2-Iminomethylphenole-type ligands o-HO-C6H4–
C(CH3)@N–R can be reacted with phenyltrichlorosilane
Table 7
Crystal data and structure refinement for 3i, 4 and 4a
Compound
3i Æ 1/2thf
34H34N2O4.5Si
570.72
4
4a
Empirical formula
Formula weight
T (K)
C
C24H22GeN2O2
443.03
293(2)
C30H25GeN5O9
672.14
93(2)
198(2)
˚
k (A)
0.71073
monoclinic
P21/c
1.54178
monoclinic
P21
0.71073
monoclinic
P21/n
Crystal system
Space group
Unit cell dimensions
˚
a (A)
13.8630(10)
12.5700(10)
8.8810(10)
11.4520(10)
14.0138(3)
10.8914(2)
˚
b (A)
˚
c (A)
17.840(3)
90
10.7840(10)
90
18.3482(4)
90
a (ꢁ)
b (ꢁ)
c (ꢁ)
112.329(10)
90
2875.7(6)
112.240(10)
90
1015.19(17)
97.3400(10)
90
2777.54(10)
3
˚
V (A )
Z
4
1.318
2
1.449
4
1.607
Dcalc (Mg mꢀ3
)
Absorption coefficient (mmꢀ1
F(000)
Crystal size (mm3)
)
0.126
1208
2.229
456
1.171
1376
0.34 · 0.29 · 0.10
3.57–25.00
0.30 · 0.30 · 0.20
4.43–74.91
0.40 · 0.30 · 0.20
1.73–33.00
h Range for data collection (ꢁ)
Index ranges
ꢀ16 6 h 6 15,
ꢀ14 6 k 6 14, ꢀ21 6 l 6 21
23117
ꢀ11 6 h 6 8,
ꢀ14 6 k 6 14, ꢀ12 6 l 6 13
5836
ꢀ21 6 h 6 21,
ꢀ16 6 k 6 16, ꢀ28 6 l 6 27
56605
Number of reflections collected
Number of independent
5007 (0.0912)
4154 (0.0474)
10477 (0.0326)
reflections (Rint
)
Completeness to hmax (%)
Absorption correction
Maximum and minimum
transmissions
98.9
semi-empirical
0.9875 and 0.9583
99.9
psi-scan
0.9942 and 0.6793
100.0
semi-empirical
0.7995 and 0.6602
Refinement method
full-matrix least-squares on F2
5007/0/400
full-matrix least-squares on F2
4154/1/263
full-matrix least-squares on F2
10477/12/424
Data/restraints/parameters
Final R indices [I > 2r(I)]
R indices (all data)
Goodness-of-fit on F2
R1 = 0.0504, wR2 = 0.0943
R1 = 0.1056, wR2 = 0.1093
1.041
R1 = 0.0477, wR2 = 0.1065
R1 = 0.0586, wR2 = 0.1118
1.181
R1 = 0.0283, wR2 = 0.0706
R1 = 0.0432, wR2 = 0.0764
1.067
ꢀ3
˚
Largest difference in peak and hole (e A
)
0.292 and ꢀ0.278
0.743 and ꢀ0.410
0.493 and ꢀ0.341

J. Wagler et al. / Inorganica Chimica Acta 358 (2005) 4270–4286
4285
Table 8
Crystal data and structure refinement for 5, 8a and 8b
Compound
5
8a Æ 1/2xylene
8b Æ 1/2xylene
Empirical formula
Formula weight
T (K)
C24H23ClN2O2Sn
525.58
203(2)
C47H43N2O4Si
727.92
93(2)
C40H38FN2O2Si
625.81
93(2)
˚
k (A)
0.71073
monoclinic
Cc
0.71073
triclinic
0.71073
triclinic
Crystal system
Space group
Unit cell dimensions
ꢀ
ꢀ
P1
P1
˚
a (A)
12.1162(5)
20.1122(8)
11.132(2)
13.256(3)
14.160(3)
9.6551(3)
11.1618(3)
15.0475(4)
87.1770(10)
88.1990(10)
81.6860(10)
1602.17(8)
2
˚
b (A)
˚
c (A)
9.5878(4)
90
a (ꢁ)
b (ꢁ)
c (ꢁ)
111.473(10)
96.678(10)
100.238(10)
1871.7(7)
108.738(2)
90
2212.55(16)
3
˚
V (A )
Z
4
1.578
2
1.292
Dcalc (Mg mꢀ3
)
1.297
0.119
662
Absorption coefficient (mmꢀ1
F(000)
)
1.299
1056
0.112
770
Crystal size (mm3)
0.70 · 0.40 · 0.25
2.03–39.00
0.14 · 0.12 · 0.04
1.70 to 25.00
ꢀ13 6 h 6 12,
ꢀ14 6 k 6 15, ꢀ16 6 l 6 16
22312
0.30 · 0.20 · 0.10
1.85 to 25.50
ꢀ11 6 h 6 11,
h Range for data collection (ꢁ)
Index ranges
ꢀ21 6 h 6 21,
ꢀ35 6 k 6 33, ꢀ16 6 l 6 14
38494
ꢀ13 6 k 6 13, ꢀ18 6 l 6 18
25989
5926 (0.0271)
Number of reflections collected
Number of independent
11795 (0.0279)
6551 (0.0627)
reflections (Rint
)
Completeness to hmax (%)
Absorption correction
Maximum and minimum
transmissions
100.0
semi-empirical
0.7225 and 0.5126
99.2
semi-empirical
0.9955 and 0.8965
99.3
semi-empirical
0.9882 and 0.8908
Refinement method
full-matrix least-squares on F2
11795/2/271
full-matrix least-squares on F2
6551/0/488
full-matrix least-squares on F2
5926/0/415
Data/restraints/parameters
Final R indices [I > 2r(I)]
R indices (all data)
Goodness-of-fit on F2
R1 = 0.0226, wR2 = 0.0534
R1 = 0.0303, wR2 = 0.0554
1.067
R1 = 0.0473, wR2 = 0.0869
R1 = 0.0969, wR2 = 0.0978
0.956
R1 = 0.0340, wR2 = 0.0823
R1 = 0.0456, wR2 = 0.0869
1.061
Largest difference in peak
ꢀ3
0.797 and ꢀ0.354
0.329 and ꢀ0.332
0.289 and ꢀ0.336
˚
and hole (e A
)
nucleophilicities compared with structurally related me-
tal-free enamines.
the data can be obtained free of charge on application
to CCDC, 12 Union Road, Cambridge CB2 1EZ, UK
(fax: (internat.) +44 1223 336 033; e-mail: deposit@ccdc.
cam.ac.uk).
5. Supplementary material
Structure solution and refinement of F2 against all
reflections were performed with the softwares SHELXS-
97 and SHELXL-97, G.M. Sheldrick, Universita¨t Go¨ttin-
gen (1986–1997). Data of structure determination and
refinement for the herein presented crystal structures
are summarized in Tables 5–8. Crystallographic data
(excluding structure factors) for the structures reported
in this paper have been deposited with the Cambridge
Crystallographic Data Centre as supplementary publica-
tion nos. CCDC-246313 (2b), CCDC-246314 (2c),
CCDC-246310 (3a), CCDC-246317 (3b), CCDC-
246309 (3c), CCDC-262715 (3f), CCDC-246315 (3i),
CCDC-194702 (4), CCDC-246316 (4a), CCDC-199261
(5), CCDC-246311 (8a), CCDC-246312 (8b). Copies of
Acknowledgments
This work was financially supported by the German
Science Foundation (DFG) and the German Chemical
Industries Fund.
References
[1] R.J.P. Corriu, J.C. Young, in: S. Patai, Z. Rappoport (Eds.), The
Chemistry of Organic Silicon Compounds, vol. 1, Wiley, New
York, 1989, pp. 1241–1288.
[2] D. Kost, I. Kalikhman, in: Z. Rappoport, Y. Apeloig (Eds.), The
Chemistry of Organic Silicon Compounds, vol. 2, Wiley, New
York, 1998, pp. 1339–1446.

4286
J. Wagler et al. / Inorganica Chimica Acta 358 (2005) 4270–4286
[3] C. Chuit, R.J.P. Corriu, C. Reye, J.C. Young, Chem. Rev. 93
(1993) 1371.
[14] J. Wagler, U. Bo¨hme, G. Roewer, Angew. Chem., Int. Ed. 41
(2002) 1732.
[4] (a) M. Golam Mortuza, M. Rafiqul Ahsan, J.A. Chudek, G.
Hunter, Chem. Commun. (2000) 2055;
[15] J. Wagler, U. Bo¨hme, G. Roewer, in: N. Auner, J. Weis (Eds.),
Organosilicon Chemistry V: from Molecules to Materials, Wiley–
VCH, Weinheim, 2003, p. 317.
(b) S. Yamaguchi, S. Akiyama, K. Tamao, J. Organomet. Chem.
652 (2002) 3;
[16] E. Williams, in: S. Patai, Z. Rappoport (Eds.), The Chemistry of
Organic Silicon Compounds, vol. 1, Wiley, New York, 1989, pp.
511–554.
(c) C. Muguruma, N. Koga, Y. Hatanaka, I. El-Sayed, M.
Mikami, M. Tanaka, J. Phys. Chem. A 104 (2000) 4928;
(d) I. El-Sayed, Y. Hatanaka, S. Onozawa, M. Tanaka, J. Am.
Chem. Soc. 123 (2001) 3597.
[17] A. Alvanipour, N.H. Buttrus, C. Eaborn, P.B. Hitchcock, A.L.
Mansour, A.K. Saxena, J. Organomet. Chem. 349 (1988) 29.
[18] (a) H. Burger, K. Hensen, P. Pickel, Z. Anorg. Allg. Chem. 617
(1992) 93;
[5] A.R. Bassindale, D.J. Parker, P.G. Taylor, J. Chem. Soc., Perkin
Trans. 2 (2000) 1059.
[6] (a) S. Riggleman, P. DeShong, J. Org. Chem. 68 (2003) 8106;
(b) W.M. Seganish, P. DeShong, J. Org. Chem. 69 (2004) 1137;
(c) S.E. Denmark, D. Wehrli, J.Y. Choi, Org. Lett. 2 (2000) 2491;
(d) S.E. Denmark, R.F. Sweis, Acc. Chem. Res. 35 (2002) 835;
(e) C. Wolf, R. Lerebours, Org. Lett. 6 (2004) 1147.
[7] (a) D. Kummer, S.H. Abdel Halim, Z. Anorg. Allg. Chem. 622
(1996) 57;
(b) K. Hensen, M. Kettner, P. Pickel, M. Bolte, Z. Naturforsch.
B 54 (1999) 200;
(c) H. Hensen, B. Spangenberg, M. Bolte, Acta Crystallogr., Sect.
C 56 (2000) 1245.
[19] (a) F. Mucha, U. Bo¨hme, G. Roewer, Chem. Commun. (1998)
1289;
(b) F. Mucha, J. Haberecht, U. Bo¨hme, G. Roewer, Monatsh.
Chem. 130 (1999) 117.
(b) A.R. Bassindale, M. Borbaruah, S.J. Glynn, D.J. Parker,
P.G. Taylor, J. Chem. Soc., Perkin Trans. 2 (1999) 2099.
[8] (a) D. Kost, V. Kingston, B. Gostevskii, A. Ellern, D. Stalke, B.
Walfort, I. Kalikhman, Organometallics 21 (2002) 2293;
(b) I. Kalikhman, O. Girshberg, L. Lameyer, D. Stalke, D. Kost,
J. Am. Chem. Soc. 123 (2001) 4709;
[20] J. Wagler, U. Bo¨hme, E. Brendler, G. Roewer, Z. Naturforsch. B
59 (2004) 1348.
[21] (a) S.-G. Teoh, G.-Y. Yeap, C.-C. Loh, L.-W. Foong, S.-B. Teo,
H.-K. Fun, Polyhedron 16 (1997) 2213;
(b) D.K. Dey, M.K. Das, H. No¨th, Z. Naturforsch. B 54 (1999)
145;
(c) I. Kalikhman, B. Gostevskii, O. Girshberg, S. Krivonos, D.
Kost, Organometallics 21 (2002) 2551;
(c) D.K. Dey, M.K. Saha, M.K. Das, N. Bhartiya, R.K. Bansal,
G. Rosair, S. Mitra, Polyhedron 18 (1999) 2687;
(d) M. Calligaris, G. Nardin, L. Randaccio, Dalton Trans. (1972)
2003.
(d) D. Kost, I. Kalikhman, Adv. Organomet. Chem. 50 (2004) 1;
(e) M. Nakash, M. Goldvaser, J. Am. Chem. Soc. 126 (2004)
3436.
[9] N. Kano, F. Komatsu, T. Kawashima, J. Am. Chem. Soc. 123
(2001) 10778.
[22] D. Agustin, G. Rima, H. Gornitzka, J. Barrau, J. Organomet.
Chem. 592 (1999) 1.
[10] (a) D.E. Morse, in: N. Auner, J. Weis (Eds.), Organosilicon
Chemistry IV: from Molecules to Materials, Wiley–VCH, Wein-
heim, 2000, p. 5;
[23] J. Wagler, U. Bo¨hme, E. Brendler, G. Roewer, Organometallics
24 (2005) 1348.
[24] K. Eichele, R.E. Wasylishen, CSOLIDS, University of Tubingen,
¨
(b) R. Tacke, C. Burschka, I. Richter, B. Wagner, R. Willeke, J.
Am. Chem. Soc. 122 (2000) 8480;
2001. Available from: <http://casgm3.anorg.chemie.uni-tuebin-
gen.de/klaus/soft/index.html>.
[25] R. West, J.J. Buffy, M. Haaf, T. Muller, B. Gerhus, M.F.
¨
(c) R. Tacke, A. Stewart, J. Becht, C. Burschka, I. Richter, Can.
J. Chem. 78 (2000) 1380;
(d) K. Benner, P. Klufers, M. Vogt, Angew. Chem. 115 (2003) 1088;
¨
Lappert, Y. Apeloig, J. Am. Chem. Soc. 120 (1998) 1639.
[26] D. Auer, C. Strohmann, A.V. Arbuznikov, M. Kaupp, Organo-
metallics 22 (2003) 2442.
(e) K. Benner, P. Klu¨fers, M. Vogt, Angew. Chem., Int. Ed. 42
(2003) 1058;
[27] J.H. Iwamiya, G.E. Maciel, J. Am. Chem. Soc. 115 (1993) 6835.
[28] R. Bertermann, A. Biller, M. Kaupp, M. Penka, O. Seiler, R.
Tacke, Organometallics 22 (2003) 4104.
(f) S.D. Kinrade, A.S. Schach, R.J. Hamilton, C.T.G. Knight,
Chem. Commun. (2001) 1564;
(g) S.D. Kinrade, R.J. Hamilton, A.S. Schach, C.T.G. Knight, J.
Chem. Soc., Dalton Trans. (2001) 961.
[29] B. Kempf, N. Hampel, A.R. Ofial, H. Mayr, Chem. Eur. J. 9
(2003) 2209.
[11] (a) R.W. Hay, Bio-inorganic Chemistry, Ellis Horwood, Chich-
ester, 1984;
[30] H. Mayr, T. Bug, M.F. Gotta, N. Hering, B. Irrgang, B. Janker,
B. Kempf, R. Loos, A.R. Ofial, G. Remennikov, H. Schimmel, J.
Am. Chem. Soc. 123 (2001) 9500, E = electrophilicity parameter
of the benzhydrilium cation; N = nucleophilicity parameter of the
enamine.
(b) J.H. Greaves, R. Redfern, H. Tinworth, J. Hyg. 73 (1974) 39;
(c) B.D. Rennison, J. Hyg. 73 (1974) 45;
(d) F.P. Rowe, T. Swinney, A. Bradfield, J. Hyg. 73 (1974) 49.
[12] (a) K. Kalka, N. Ahmad, T. Criswell, D. Boothman, H. Mukhtar,
Cancer Res. 60 (2000) 5984;
[31] H. Mayr, in: J.E. Puskas, A. Michel, S. Barghi, C. Paulo (Eds.),
Ionic Polymerizations and Related Processes, NATO Science
Series: E Applied Sciences, vol. 359, Kluwer Academic Publishers,
Dordrecht, 1999, pp. 99–115.
(b) C.M. Whitacre, D.K. Feyes, T. Satoh, J. Grossmann, J.W.
Mulvihill, H. Mukhtar, N.L. Oleinick, Clin. Cancer Res. 6 (2000)
2021.
[32] H. Mayr, B. Kempf, A.R. Ofial, Acc. Chem. Res. 36 (2003) 66.
[33] P. Pfeiffer, E. Breit, E. Lubbe, T. Tsumaki, Liebigs Ann. Chem.
¨
[13] J. Wagler, U. Bo¨hme, G. Roewer, Angew. Chem. 114 (2002)
1825.
503 (1933) 127.