Neutral pentacoordinate halogeno- and pseudohalogenosilicon(IV) complexes with a tridentate dianionic O,N,O or N,N,O ligand: Synthesis and structural characterization in the solid state and in solution

Article abstract of DOI:10.1002/ejic.201200308

A series of neutral pentacoordinate silicon(IV) complexes with an SiFO ₂NC, SiBrO₂NC, SiO₂NC₂, SiO ₂N₂C, SiFON₂C, SiBrON₂C, SiON ₂C₂, or SiON₃C skeleton was synthesized and structurally characterized by multinuclear NMR spectroscopy in the solid state and in solution and by single-crystal X-ray diffraction. The compounds studied contain a tridentate dianionic O,N,O or N,N,O ligand, a phenyl ligand, and a (pseudo)halogeno ligand (F, Br, CN, N₃, NCS). The structures, NMR spectroscopic parameters, and chemical properties of these silicon(IV) complexes were compared with those of related compounds that contain an analogous tridentate dianionic S,N,O ligand instead of the O,N,O or N,N,O ligand (comparison of S/O/NMe analogues). In addition, two cationic pentacoordinate silicon(IV) complexes with an SiO₂N₂C and SiON ₃C skeleton, respectively, were synthesized (isolated as iodides) and structurally characterized. These compounds contain a tridentate O,N,O or N,N,O ligand, a phenyl ligand, and an acetonitrile ligand. The experimental investigations reported in this article were complemented by computational studies to better understand the different properties of some of the S/O/NMe analogues studied.

Full text of DOI:10.1002/ejic.201200308

FULL PAPER
DOI: 10.1002/ejic.201200308
Neutral Pentacoordinate Halogeno- and Pseudohalogenosilicon(IV) Complexes
with a Tridentate Dianionic O,N,O or N,N,O Ligand: Synthesis and Structural
Characterization in the Solid State and in Solution
Jörg Weiß,[a] Bastian Theis,[a] Stefan Metz,[a] Christian Burschka,[a] Célia Fonseca Guerra,[b]
F. Matthias Bickelhaupt,*[b] and Reinhold Tacke*[a]
Keywords: Silicon / Coordination chemistry / Pentacoordination / Tridentate ligands / Halogeno ligands / Pseudohalogeno
ligands
A series of neutral pentacoordinate silicon(IV) complexes
with an SiFO2NC, SiBrO2NC, SiO2NC2, SiO2N2C, SiFON2C,
SiBrON2C, SiON2C2, or SiON3C skeleton was synthesized
and structurally characterized by multinuclear NMR spec-
troscopy in the solid state and in solution and by single-crys-
tal X-ray diffraction. The compounds studied contain a tri-
dentate dianionic O,N,O or N,N,O ligand, a phenyl ligand,
and a (pseudo)halogeno ligand (F, Br, CN, N3, NCS). The
structures, NMR spectroscopic parameters, and chemical
properties of these silicon(IV) complexes were compared
with those of related compounds that contain an analogous
tridentate dianionic S,N,O ligand instead of the O,N,O or
N,N,O ligand (comparison of S/O/NMe analogues). In ad-
dition, two cationic pentacoordinate silicon(IV) complexes
with an SiO2N2C and SiON3C skeleton, respectively, were
synthesized (isolated as iodides) and structurally charac-
terized. These compounds contain a tridentate O,N,O or
N,N,O ligand, a phenyl ligand, and an acetonitrile ligand.
The experimental investigations reported in this article were
complemented by computational studies to better under-
stand the different properties of some of the S/O/NMe ana-
logues studied.
get information about the effect of the soft sulfur ligand
atom on the structure and properties of these compounds,
we have been interested in the synthesis and characteriza-
tion of the corresponding analogues 2a–2g and 3a–3g that
contain a hard oxygen or nitrogen atom (NMe group) in-
stead of the sulfur ligand atom. In a preliminary study, we
have already demonstrated that the chlorosilicon(IV) com-
plexes 2b and 3b (analogues of 1b) can be synthesized.[4] In
extension of these investigations, we have now succeeded
in synthesizing compounds 2a, 2c, 2e–2g, 3a, and 3c–3g.
Interestingly, the synthesis of the iodosilicon(IV) complex
2d failed; instead, the ionic binuclear tetracoordinate sili-
con(IV) compound 2i was obtained. In addition, we suc-
ceeded in synthesizing the ionic pentacoordinate silicon(IV)
complexes 2h and 3h, whereas the corresponding sulfur an-
alogue 1h does not exist.
We report here on (i) the syntheses of compounds 2a, 2c,
2e–2i, 3a, and 3c–3h, (ii) their structural characterization in
the solid state (crystal structure analysis, except for 3d;
NMR spectroscopy, except for 2h, 3d, and 3h) and in solu-
tion (NMR spectroscopy), and (iii) a systematic compari-
son of the respective S/O/NMe analogues in question. The
experimental investigations described herein were comple-
mented by quantum-chemical studies to better understand
the different properties of some of the S/O/NMe analogues
studied.
Introduction
Higher-coordinate silicon compounds, with the first ex-
amples dating back to the beginning of the 19th century,
have received considerable attention in recent years.[1,2]
Most of the penta- and hexacoordinate silicon(IV) com-
pounds reported in the literature contain hard ligand atoms
such as fluorine, chlorine, oxygen, nitrogen, and carbon
atoms.[1] Recently, quite new perspectives for the chemistry
of higher-coordinate silicon have been generated by intro-
ducing soft chalcogen ligand atoms (sulfur, selenium, tel-
lurium) into the coordination sphere of the hard silicon(IV)
coordination center.[2b,2f,2m,3] In this context, we have re-
cently reported on the syntheses, structures, and properties
of the neutral pentacoordinate silicon(IV) complexes 1a–
1g.[2f,3f] These compounds contain a tridentate S,N,O li-
gand, a phenyl ligand, and a (pseudo)halogeno ligand. To
[a] Universität Würzburg, Institut für Anorganische Chemie,
Am Hubland, 97074 Würzburg, Germany
Fax: +49-931-31-84609
E-mail: r.tacke@uni-wuerzburg.de
[b] Department of Theoretical Chemistry, Amsterdam Center for
Multiscale Modeling, VU University,
De Boelelaan 1083, 1081 HV Amsterdam, The Netherlands
Fax: +31-20-59-87629
E-mail: f.m.bickelhaupt@vu.nl
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201200308.
3216
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2012, 3216–3228

Neutral Pentacoordinate Silicon(IV) Complexes
3f, X = N3; 3g, X = NCS), in acetonitrile or benzene (3c
only) (Scheme 2). Attempts to synthesize the corresponding
iodosilicon(IV) complex 3d analogously to the synthesis of
3e–3g by reaction of 3b with iodotrimethylsilane in aceto-
nitrile failed (see below); however, compound 3d could be
prepared by treatment of 3b with iodotrimethylsilane in
dichloromethane at –80 °C in an NMR tube, but upon
warming of the solution to –20 °C, compound 3d decom-
posed. All attempts to isolate 3d below –20 °C failed.
Results and Discussion
Scheme 2. Synthesis of compounds 3a and 3c–3g.
Syntheses
Interestingly, treatment of the chlorosilicon(IV) com-
plexes 2b and 3b with iodotrimethylsilane in acetonitrile did
not yield the expected iodosilicon(IV) complexes 2d and 3d.
Instead, the ionic complexes 2h (isolated as the solvate
2h·CH3CN) and 3h were obtained (Scheme 3), in which a
neutral acetonitrile molecule is coordinated to the silicon
coordination center; i.e., the resulting pentacoordinate sili-
con(IV) complexes are positively charged, with an iodide
ion as the counterion. In this context, it is interesting to
note that the analogous iodosilicon(IV) complex 1d (triden-
tate S,N,O ligand) can be easily synthesized by reaction of
the corresponding chlorosilicon(IV) complex 1b with iodo-
trimethylsilane in acetonitrile.[2f] Thus, replacement of the
soft sulfur ligand atom by the hard oxygen or nitrogen li-
gand atom (1b/1d Ǟ 2b/2d, 3b/3d) affects the chemical
properties of these analogues dramatically (in this context,
see the section Computational Studies).
The fluorosilicon(IV) complex 2a was synthesized by
treatment of the corresponding chlorosilicon(IV) complex
2b with ammonium fluoride in tetrahydrofuran (Scheme 1).
The related (pseudo)halogenosilicon(IV) complexes 2c and
2e–2g were obtained by reaction of 2b with the correspond-
ing (pseudo)halogenotrimethylsilane, Me3SiX (2c, X = Br;
2e, X = CN; 2f, X = N3; 2g, X = NCS), in acetonitrile
(Scheme 1).
As already mentioned above, treatment of the pentacoor-
dinate chlorosilicon(IV) complex 3b (N,N,O ligand) with
iodotrimethylsilane in dichloromethane at –80 °C yielded
3d; however, all attempts to isolate 3d failed so far. In con-
trast, reaction of the analogous chlorosilicon(IV) complex
2b (O,N,O) with iodotrimethylsilane in dichloromethane at
20 °C afforded the ionic binuclear tetracoordinate sili-
Scheme 1. Synthesis of compounds 2a, 2c, and 2e–2g.
The fluorosilicon(IV) complex 3a was synthesized by con(IV) complex 2i (Scheme 3), whereas this synthesis in
treatment of the corresponding chlorosilicon(IV) complex dichloromethane at –80 °C failed. Neither the formation of
3b with ammonium fluoride in tetrahydrofuran (Scheme 2). 2d nor the conversion of 2d to 2i could be observed.
Alternatively, compound 3a was obtained by reaction of 3b
Compounds 2a, 2c, 2e–2g, 2h·CH3CN, 2i, 3a,
with silver tetrafluoroborate/triethylamine in acetonitrile 3c·0.5C6H6, 3e–3g, and 3h were isolated as yellow or
(Scheme 2). The related (pseudo)halogenosilicon(IV) com- orange-colored solids (yields: 2a, 57%; 2c, 61%; 2e, 64%;
plexes 3c (isolated as 3c·0.5C6H6) and 3e–3g were synthe- 2f, 79%; 2g, 88%; 2h·CH3CN, 44%; 2i, 58%; 3a, 49%/61%;
sized by treatment of 3b with the corresponding (pseudo)- 3c·0.5C6H6, 84%; 3e, 81%; 3f, 88%; 3g, 70%; 3h, 64%).
halogenotrimethylsilane, Me3SiX (3c, X = Br; 3e, X = CN; Their identities were established by elemental analyses
Eur. J. Inorg. Chem. 2012, 3216–3228
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3217

F. M. Bickelhaupt, R. Tacke et al.
FULL PAPER
Scheme 3. Synthesis of compounds 2h, 2i, and 3h.
Figure 2. Molecular structure of the cation of 2h in the crystal of
2h·CH3CN (probability level of displacement ellipsoids 50%). Se-
lected bond lengths [Å] and angles [°]: Si–O1 1.6761(16), Si–O2
1.6899(16), Si–N1 1.8969(19), Si–N2 1.934(2), Si–C1 1.858(2), N2–
C2 1.136(3), C2–C3 1.442(3); O1–Si–O2 132.55(8), O1–Si–N1
87.05(8), O1–Si–N2 84.12(8), O1–Si–C1 115.26(9), O2–Si–N1
91.78(8), O2–Si–N2 83.71(8), O2–Si–C1 111.54(9), N1–Si–N2
163.12(8), N1–Si–C1 100.81(9), N2–Si–C1 95.97(9), Si–N2–C2
177.96(19), N2–C2–C3 178.7(3).
(C,H,N), multinuclear NMR spectroscopic studies in the
solid state (except for 2h·CH3CN and 3h) and in solution,
and crystal structure analyses (the crystal structure analysis
of 3h was performed with the solvate 3h·2CH3CN). Com-
pound 3d was only characterized by NMR spectroscopy in
solution.
Crystal Structure Analyses
Compounds 2a, 2c, 2e–2g, 2h·CH3CN, 2i, 3a,
3c·0.5C6H6, 3e–3g, and 3h·2CH3CN were structurally char-
acterized by single-crystal X-ray diffraction. The crystal
data, the experimental parameters used for these studies,
and the molecular structures of 2a, 2e–2g, 2i (cation), 3c,
and 3e–3g are given in the Supporting Information. The
molecular structures of compounds 2c, 2h (cation), 3a, and
3h (cation) are shown in Figures 1, 2, 3, and 4; selected
bond lengths and angles are given in the respective captions.
Figure 3. Molecular structure of 3a in the crystal (probability level
of displacement ellipsoids 50%). Selected bond lengths [Å] and
angles [°]: Si–F 1.6526(7), Si–O 1.7225(7), Si–N1 1.9603(8), Si–N2
1.7549(9), Si–C1 1.8860(9); F–Si–O 87.76(3), F–Si–N1 171.25(4),
F–Si–N2 92.71(4), F–Si–C1 96.07(4), O–Si–N1 88.27(3), O–Si–N2
128.23(4), O–Si–C1 112.73(4), N1–Si–N2 83.64(4), N1–Si–C1
92.66(4), N2–Si–C1 118.66(4).
The silicon coordination polyhedra of the neutral penta-
coordinate (pseudo)halogenosilicon(IV) complexes studied
are strongly distorted trigonal bipyramids, with the imino
nitrogen atom of the tridentate O,N,O and N,N,O ligand,
respectively, and the monodentate (pseudo)halogeno ligand
in the two axial positions. Analogous structures were also
observed for the related silicon(IV) complexes 1a–1g (S,N,O
ligand),[2f,3f] 2b (O,N,O),[4] and 3b (N,N,O).[4] In the case of
Figure 1. Molecular structure of 2c in the crystal (probability level
of displacement ellipsoids 50%). Selected bond lengths [Å] and
angles [°]: Si–Br 2.3870(9), Si–O1 1.6843(19), Si–O2 1.694(2), Si–N
1.974(2), Si–C1 1.859(3); Br–Si–O1 87.76(7), Br–Si–O2 85.91(8),
Br–Si–N 165.41(7), Br–Si–C1 98.34(8), O1–Si–O2 134.02(11), O1–
Si–N 91.04(10), O1–Si–C1 112.69(11), O2–Si–N 84.49(10), O2–Si–
C1 113.29(11), N–Si–C1 95.53(10).
3218
www.eurjic.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2012, 3216–3228

Neutral Pentacoordinate Silicon(IV) Complexes
1a–1g,[2f,3f] 2b,[4] and 3b[4] are also summarized in Table 1.
Comparison of all these data demonstrates that the degree
of distortion of the trigonal-bipyramidal silicon coordina-
tion polyhedra of the respective S/O/NMe analogues differs
significantly. The highest degree of distortion was observed
for the series of compounds with the O,N,O ligand, fol-
lowed by those with the S,N,O and N,N,O ligands; only 1a
and 1d deviate from this rank order. With Berry distortions
of Ͼ50%, the silicon coordination polyhedra of 1e (57.3%),
2e (52.1%), and 2f (57.3%) can be alternatively described
as strongly distorted square pyramids.
Contrary to the significantly different degrees of distor-
tion (Table 1) observed for the respective S/O/NMe ana-
logues, replacement of the sulfur ligand atom of 1a–1g by
an oxygen or a methyl-bound nitrogen atom only slightly
affects the analogous silicon–ligand bond lengths. As can
be seen from Tables 2 and 3, most of the analogous Si–X
(X = F, Cl, Br, CN, N3, NCS), Si–O(enolato), Si–N(imino),
and Si–C bond lengths of the respective S/O/NMe ana-
logues are very similar. The greatest differences were ob-
served for the Si–CN bond lengths of the S/O/NMe triple
1e/2e/3e (Si–CN bond, Δmax = 0.166 Å); all the other Δmax
values are in the range 0.006–0.052 Å (Table 2 and Table 3).
The Si–O(phenolato) bond lengths of 2a–2c and 2e–2g
Figure 4. Molecular structure of the cation of 3h in the crystal of
3h·2CH3CN (probability level of displacement ellipsoids 50%). Se-
lected bond lengths [Å] and angles [°]: Si–O 1.6770(19), Si–N1
1.9227(19), Si–N2 1.741(2), Si–N3 1.996(2), Si–C1 1.873(2), N3–
C2 1.136(3), C2–C3 1.460(4); O–Si–N1 93.08(9), O–Si–N2
126.63(10), O–Si–N3 83.09(9), O–Si–C1 112.49(10), N1–Si–N2
86.18(9), N1–Si–N3 172.58(10), N1–Si–C1 96.83(9), N2–Si–N3
90.99(9), N2–Si–C1 120.58(11), N3–Si–C1 90.52(10), Si–N3–C2
168.6(2), N3–C2–C3 179.9(3).
(O,N,O ligand) amount to 1.6788(10)–1.7013(12) Å (Δmax
,
0.023 Å), and the Si–N(amido) bond lengths of 3a–3c and
3e–3g (N,N,O ligand) are in the range 1.7432(13)–
1.7591(12) Å (Δmax, 0.016 Å); i.e., the different (pseudo)-
halogeno ligands do not significantly affect the Si–O(phen-
olato) and Si–N(amido) bond lengths.
1e, 2e, and 2f, the silicon coordination polyhedra are even
better described as strongly distorted square pyramids (see
below).
Table 1 shows the Berry distortions (transition trigonal
bipyramidǞsquare pyramid) and the axial N–Si–X (X =
F, Cl, Br, CN, N3, NCS) bond angles of the new com-
The cationic pentacoordinate silicon(IV) complexes
pounds studied. The Berry distortions range from 29.8% to 2h·CH3CN (O,N,O ligand) and 3h·2CH3CN (N,N,O), with
57.3% (O,N,O ligand) and from 16.2% to 32.5% (N,N,O), their axial acetonitrile ligand, have similar structures as ob-
respectively. The axial N–Si–X angles are in the range served for the related neutral (pseudo)halogenosilicon(IV)
161.14(5)–169.55(5)° (O,N,O) and 170.25(5)–175.27(5)° complexes discussed above. The silicon coordination poly-
(N,N,O). The corresponding data of the related compounds hedra are strongly distorted trigonal bipyramids (Berry dis-
Table 1. Berry distortions [%] and axial N–Si–X (X = F, Cl, Br, I, CN, N3, NCS) bond angles [°] of compounds 1a–1g, 2a–2c, 2e–2g,
3a–3c, and 3e–3g.
Compounds
(ligands)
Berry
distortions
N–Si–X
Compounds
(ligands)
Berry
distortions
N–Si–X
Compounds
(ligands)
Berry
distortions
N–Si–X
1a (S,N,O, F)
1b (S,N,O, Cl)
1c (S,N,O, Br)
1d (S,N,O, I)
1e (S,N,O, CN)
1f (S,N,O, N3)
1g (S,N,O, NCS)
19.0[a]
28.4[b]
27.9[a]
22.5[b]
57.3[a]
42.6[a]
29.8[a]
168.46(3)[a] 2a (O,N,O, F)
167.66(3)[b] 2b (O,N,O, Cl)
167.41(3)[a] 2c (O,N,O, Br)
167.96(7)[b] 2d (O,N,O, I)
161.73(5)[a] 2e (O,N,O, CN)
164.41(6)[a] 2f (O,N,O, N3)
29.8
169.55(5)
3a (N,N,O, F)
26.1
171.25(4)
173.04(5)[c]
175.27(5)
41.2[c]
42.6
166.02(2)[c] 3b (N,N,O, Cl)
23.8[c]
16.2
165.41(7)
3c (N,N,O, Br)
3d (N,N,O, I)
3e (N,N,O, CN)
3f (N,N,O, N3)
3g (N,N,O, NCS)
52.1
57.3
163.23(6)
161.14(5)
165.80(5)
27.0
32.5
25.6
171.73(5)
170.25(5)
171.98(6)
168.30(7)[a] 2g (O,N,O, NCS) 42.6
[a] Data taken from ref.[2f] [b] Data taken from ref.[3f] [c] Data taken from ref.[4]
Table 2. Comparison of the Si–X (X = F, Cl, Br, I, CN, N3, NCS) bond lengths [Å] of 1a–1g, 2a–2c, 2e–2g, 3a–3c, and 3e–3g and
maximum bond lengths differences Δmax [Å].
Compounds (ligands) and Δmax Si–F
Si–Cl
Si–Br
Si–I
Si–CN
Si–N3
Si–NCS
1a–1g (S,N,O)
2a–2c, 2e–2g (O,N,O)
3a–3c, 3e–3g (N,N,O)
Δmax
1.6687(5)[a]
1.6462(11)
1.6526(7)
0.023
2.1954(4)[b]
2.1913(3)[c]
2.2098(6)[c]
0.019
2.4051(4)[a]
2.3870(9)
2.4298(7)
0.043
2.7396(8)[b]
1.9563(16)[a]
2.0383(15)
1.8726(13)
0.166
1.8573(14)[a]
1.8378(12)
1.8894(13)
0.052
1.8729(17)[a]
1.8470(12)
1.8502(13)
0.026
[a] Data taken from ref.[2f] [b] Data taken from ref.[3f] [c] Data taken from ref.[4]
Eur. J. Inorg. Chem. 2012, 3216–3228
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3219

F. M. Bickelhaupt, R. Tacke et al.
FULL PAPER
Table 3. Comparison of the Si–O(enolato), Si–N(imino), and Si–C NMR Studies
bond lengths [Å] of 1a–1g, 2a–2c, 2e–2g, 3a–3c, and 3e–3g and
maximum bond lengths differences Δmax [Å].
Compounds 2a, 2c, 2e–2i, 3a, and 3c–3h were charac-
terized by multinuclear NMR spectroscopy in the solid
state (except for 2h·CH3CN, 3d, and 3h) and in solution
(except for 2i). The NMR spectroscopic data obtained
(Table 4 and Table 5) are in accordance with the experimen-
tally established crystal structures of these compounds.
Analysis of Table 4 reveals that the 15N (imino nitrogen
atom) and 29Si chemical shifts of the neutral pentacoordi-
nate silicon(IV) complexes 1a–1g, 2a–2c, 2e–2g, 3a–3c, and
3e–3g in the solid state depend on the nature of both the
tridentate ligand (S,N,O, O,N,O, N,N,O) and the (pseudo)-
halogeno ligand. The 15N chemical shifts of the imino nitro-
gen atom are in the range δ = –173.2 to –145.4/–140.4 ppm,
and the 29Si chemical shifts range from δ = –116.4 to
–83 ppm. The 29Si chemical shifts of 1a–1g, 2a–2c, 2e–2g,
3a–3c, and 3e–3g determined by NMR spectroscopic stud-
ies in solution (solvent, CD2Cl2; Table 5) are very similar to
those obtained in the solid state (Table 4), indicating that
all these neutral pentacoordinate silicon(IV) complexes also
exist in solution.
Compd. (ligands), Δmax
Si–O(enolato) Si–N(imino) Si–C
1a (S,N,O, F)
2a (O,N,O, F)
3a (N,N,O, F)
Δmax (subgroup)
1.6967(6)[a]
1.6861(12)
1.7225(7)
0.036
2.0096(6)[a]
2.0052(14)
1.9603(8)
0.049
1.8745(7)[a]
1.8661(15)
1.8860(9)
0.020
1b (S,N,O, Cl)
2b (O,N,O, Cl)
3b (N,N,O, Cl)
Δmax (subgroup)
1.6850(8)[b]
1.6802(7)[c]
1.6858(14)[c]
0.006
2.0069(10)[b] 1.8593(11)[b]
1.9843(7)[c] 1.8547(8)[c]
2.0044(14)[c] 1.8762(18)[c]
0.023 0.022
1c (S,N,O, Br)
2c (O,N,O, Br)
3c (N,N,O, Br)
Δmax (subgroup)
1.6802(10)[a]
1.694(2)
1.6853(15)
0.014
1.9851(12)[a] 1.8579(13)[a]
1.974(2)
1.9717(18)
0.013
1.859(3)
1.879(2)
0.021
1d (S,N,O, I)
1.6655(19)[b]
1.936(2)[b]
1.851(3)[b]
1e (S,N,O, CN)
2e (O,N,O, CN)
3e (N,N,O, CN)
Δmax (subgroup)
1.6979(9)[a]
1.6886(11)
1.6958(11)
0.009
1.9451(13)[a] 1.8681(12)[a]
1.9736(13)
1.9918(11)
0.047
1.8659(15)
1.8726(13)
0.007
1f (S,N,O, N3)
2f (O,N,O, N3)
3f (N,N,O, N3)
Δmax (subgroup)
1.6768(10)[a]
1.6999(10)
1.6981(10)
0.023
2.0071(12)[a] 1.8507(14)[a]
2.0094(11)
1.9933(12)
0.016
1.8644(13)
1.8704(14)
0.020
The solid-state 29Si NMR spectra of 2a–2c and 3a–3c
deserve a special discussion. As can be seen from Figure 5,
1g (S,N,O, NCS)
2g (O,N,O, NCS)
3g (N,N,O, NCS)
Δmax (subgroup)
1.6788(13)[a]
1.6888(10)
1.6837(11)
0.010
1.9684(13)[a] 1.8697(16)[a]
a
29Si,19F coupling was observed for the two fluorosili-
1.9670(11)
1.9733(12)
0.006
1.8535(13)
1.8600(15)
0.016
con(IV) complexes 2a and 3a. For the related chloro- and
bromosilicon(IV) complexes 2b, 3b, 2c, and 3c, broad and
structured resonance signals were observed. This phenome-
non can be explained by 29Si,X couplings [X = 35Cl (I =
3/2), 37Cl (I = 3/2); 79Br (I = 3/2), 81Br (I = 3/2)] and the
well known effect that MAS fails to completely eliminate
the effect of dipolar coupling for spin-1/2 nuclei when cou-
pled to quadrupole nuclei with a quadrupole frequency
Δmax (all compounds)
[a] Data taken from ref.[2f] [b] Data taken from ref.[3f] [c] Data taken
from ref.[4]
0.057
0.074
0.035
tortion: 2h·CH3CN, 44.5%; 3h·2CH3CN, 20.1%), with axial comparable to the Zeeman frequency of the nuclei.[5]
N–Si–N angles of 163.12(8)° and 172.58(10)°, respectively. As already reported recently, for compounds 1a, 1b, and
The Si–N(acetonitrile) bond lengths amount to 1.934(2) Å 1f (S,N,O ligand) a dynamic equilibrium between the penta-
and 1.996(2) Å. The Si–O(enolato) [1.6899(16), coordinate species A and the tetracoordinate species B (exi-
1.6770(19) Å], Si–O(phenolato)/Si–N(amido) [1.6761(16)/ sting as two diastereomers) in solution (solvent, CD2Cl2)
1.741(2) Å], and Si–C [1.858(2), 1.873(2) Å] bond lengths of was detected (Scheme 4).[2f] Interestingly, such kind of equi-
2h·CH3CN and 3h·2CH3CN are very similar to the analo- librium could not be observed for the corresponding ana-
gous bond lengths of the related neutral (pseudo)halogeno- logues 2a, 2b, and 2f (O,N,O ligand) and 3a, 3b, and 3f
silicon(IV) complexes (see Tables 2 and 3). In contrast, the (N,N,O) (solvents: CD2Cl2, CD3CN, C6D6). The same
Si–N(imino) bond lengths of 2h·CH3CN [1.8969(19) Å] and holds true for the related compounds in the series 2 and 3
3h·2CH3CN [1.9227(19) Å] are somewhat shortened.
with other (pseudo)halogeno ligands. The special behavior
Table 4. 15N (imino nitrogen atom) and 29Si chemical shifts of compounds 1a–1g, 2a–2c, 2e–2g, 3a–3c, and 3e–3g in the solid state.
Compounds
(ligands)
δ15N
[ppm]
δ29Si
[ppm]
Compounds
(ligands)
δ15N
[ppm]
δ29Si
[ppm]
Compounds
(ligands)
δ15N
[ppm]
δ29Si
[ppm]
1a (S,N,O, F)
1b (S,N,O, Cl)
1c (S,N,O, Br)
1d (S,N,O, I)
1e (S,N,O, CN)
1f (S,N,O, N3)
1g (S,N,O, NCS)
–151.5[a]
–89.1[a]
–83[b]
–89[a]
–91[b]
2a (O,N,O, F)
2b (O,N,O, Cl)
2c (O,N,O, Br)
2d (O,N,O, I)
–160.4
–167.7[c]
–171.7
–108.5
–97[c,d]
–100[d]
3a (N,N,O, F)
3b (N,N,O, Cl)
3c (N,N,O, Br)
3d (N,N,O, I)
3e (N,N,O, CN)
3f (N,N,O, N3)
–173.1
–111.5
–96[c,d]
–97[d,e]
–149.4[b]
–161.5[c]
–165.4[e]
–153.6[a]
–163.8[b]
–158.8[a]
–100.8[a] 2e (O,N,O, CN)
–172.3
–163.0
–113.7
–106.0
–116.4
–163.7
–160.0
–107.1
–101.6
–110.4
[a,f]
–145.4 or –140.4
–87.1[a]
–98.8[a]
2f (O,N,O, N3)
2g (O,N,O, NCS) –173.2
–157.6[a]
3g (N,N,O, NCS) –164.3
[a] Data taken from ref.[2f] [b] Data taken from ref.[3f] [c] Data taken from ref.[4] [d] Center of an asymmetric multiplet; see Figure 5. [e]
Data for 3·0.5C6H6. [f] Resonance signals could not be unambiguously assigned.
3220
www.eurjic.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2012, 3216–3228

Neutral Pentacoordinate Silicon(IV) Complexes
Table 5. 29Si chemical shifts of compounds 1a–1g, 2a–2c, 2e–2g, 3a–3c, and 3e–3g in solution.
Compounds (ligands)
δ29Si [ppm]
Compounds (ligands)
δ29Si [ppm]
Compounds (ligands)
δ29Si [ppm]
1a (S,N,O, F)
1b (S,N,O, Cl)
1c (S,N,O, Br)
1d (S,N,O, I)
1e (S,N,O, CN)
1f (S,N,O, N3)
1g (S,N,O, NCS)
–88.1[b]
–82.1[c]
–86.7[b]
–92.5[c]
–99.9[b]
–87.8[b]
–98.3[b]
2a (O,N,O, F)
2b (O,N,O, Cl)
2c (O,N,O, Br)
2d (O,N,O, I)
2e (O,N,O, CN)
2f (O,N,O, N3)
2g (O,N,O, NCS)
–109.3
–95.5[d]
–98.0
3a (N,N,O, F)
3b (N,N,O, Cl)
3c (N,N,O, Br)
3d (N,N,O, I)
3e (N,N,O, CN)
3f (N,N,O, N3)
3g (N,N,O, NCS)
–108.1
–94.8[d]
–94.1
–81.9
–110.5
–105.5
–114.2
–106.7
–104.7
–111.2
[a] Solvent, CD2Cl2. [b] Data taken from ref.[2f] [c] Data taken from ref.[3f] [d] Data taken from ref.[4]
to the cleavage of the stronger Si–O (2a, 2b, 2f) and Si–N
bond (3a, 3b, 3f) (in this context, see the section Computa-
tional Studies).
Scheme 4. Dynamic equilibrium of 1a, 1b, and 1f between penta-
(A) and tetracoordination (B) in solution.
The 29Si chemical shifts of the cationic pentacoordinate
silicon(IV) complexes 2h and 3h were only determined in
solution (solvent, CD3CN). When using CD2Cl2 as the sol-
vent, both compounds decompose. Attempts to character-
ize 2h·CH3CN and 3h by solid-state NMR spectroscopy
gave unsatisfactory results (NMR spectra of very poor
quality).
Computational Studies
We have conducted computational studies to understand
two of the experimental observations described above using
relativistic density functional theory (DFT). The first objec-
tive was to understand why the pentacoordinate silicon(IV)
complex 1b (X = S) is experimentally observed to be in
equilibrium with the tetracoordinate silicon(IV) species 1bЈ/
1bЈЈ, whereas such an equilibrium is not observed for the
analogous pentacoordinate silicon(IV) complexes 2b (X =
O) and 3b (X = NMe) (Scheme 5).
Figure 5. 29Si VACP/MAS NMR spectra of A) 2a (X = F; O,N,O),
B) 3a (X = F; N,N,O), C) 2b (X = Cl; O,N,O), D) 3b (X = Cl;
N,N,O), E) 2c (X = Br; O,N,O), and F) 3c (X = Br; N,N,O).
of 1a, 1b, and 1f with their S,N,O ligand (compared with
their corresponding analogues containing the O,N,O or
N,N,O ligand) can be understood in terms of a more fa-
vored cleavage of the Si–S bond of 1a, 1b, and 1f compared Scheme 5. Equilibrium between penta- and tetracoordination.
Eur. J. Inorg. Chem. 2012, 3216–3228
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3221

F. M. Bickelhaupt, R. Tacke et al.
FULL PAPER
The second objective was to understand why the neutral line with the experimental observation, the formation of the
pentacoordinate iodosilicon(IV) complexes 2d (X = O) and tetracoordinate species 1bЈ (X = S) and 2bЈ (X = O) is rela-
3d (X = NMe) undergo an iodo/acetonitrile exchange, lead- tively favorable (ΔG = 8.4 kcal/mol, in CH2Cl2) and prohib-
ing to the cationic pentacoordinate silicon(IV) complexes itively unfavorable (ΔG = 21.6 kcal/mol, in CH2Cl2), respec-
2h and 3h, whereas the analogue 1d (X = S) does not show tively. However, the formation of the tetracoordinate species
such a ligand exchange in acetonitrile and keeps the iodo 3bЈ (X = NMe) is associated with the same Gibbs free en-
ligand (Scheme 6).
ergy of reaction (ΔG = 8.4 kcal/mol, in CH2Cl2) as that of
1bЈ, whereas the latter species is experimentally observed
and the former is not.
Table 6. Computed relative Gibbs free energies ΔG [kcal/mol] for
the reactions associated with Scheme 5.[a]
Medium
Gas phase
CH2Cl2
System
ΔG
b
I
bЈ
bЈЈ
1 (X = S)
0
0
0
0
0
0
16.7
26.1
28.4
15.9
25.9
30.7
3.0
15.9
5.9
8.4
21.6
8.4
4.0
17.2
6.2
8.9
22.8
8.4
2 (X = O)
3 (X = NMe)
1 (X = S)
Scheme 6. Iodo/acetonitrile ligand exchange.
All calculations were performed using the Amsterdam
Density Functional (ADF) program,[6] using density func-
tional theory at BP86/TZ2P for geometry optimization and
2 (X = O)
3 (X = NMe)
[a] Computed at ZORA-BP86/TZ2P using COSMO to estimate
energies.[7] Relativistic effects were taken into account using solvent effects by CH2Cl2.
the zeroth-order regular approximation (ZORA).[8] In all
cases, solvation in dichloromethane (CH2Cl2) or aceto-
nitrile (CH3CN) was simulated using the conductor-like
Table 7. Analysis of the heterolytic Si–X bonding mechanism asso-
ciated with Scheme 7 [kcal/mol unless stated otherwise].[a]
screening model (COSMO).[9] All stationary points were
verified to be minima through vibrational analysis. Gibbs
free energies were computed using electronic energies aug-
mented with the COSMO free energies of solvation plus the
gas-phase internal energy and entropy corrections of the
solute. The bonding mechanism was analyzed within the
framework of quantitative Kohn–Sham molecular orbital
(MO) theory in combination with a quantitative energy de-
composition analysis (EDA). In this EDA approach, the
bond energy ΔE is first decomposed into the preparation
energy ΔEprep needed to deform the constituting molecular
fragments from their own equilibrium structure to the ge-
ometry they adopt in the overall complex plus the interac-
tion energy ΔEint between these deformed fragments, i.e.,
ΔE = ΔEprep + ΔEint. The interaction energy ΔEint is
furthermore decomposed into classical electrostatic attrac-
tion ΔVelstat, Pauli repulsion ΔEPauli between occupied or-
bitals, and stabilizing orbital interactions ΔEoi, such as
HOMO–LUMO interaction (for details, see ref.[10]). Atomic
charges were computed using the Voronoi deformation den-
sity (VDD) method.[11]
Medium
Energy
1M
2M
3M
Gas phase εHOMO of PhX–
ΔQ(X) in PhX–
0.364 eV
0.612 eV
1.177 eV
–0.582 a.u. –0.550 a.u. –0.411 a.u.
ΔQ(Si) in [Si]+ fragment 0.430 a.u. 0.437 a.u. 0.456 a.u.
d(Si–X)
ΔEPauli
2.30 Å
154.1
–179.4
–114.3
–139.6
35.9
–103.7
–90.6
–7.3
1.74 Å
195.0
–224.9
–140.4
–170.3
47.0
–123.3
–108.1
–20.6
1.78 Å
248.8
–263.9
–179.1
–194.2
73.9
–120.3
–102.9
–21.1
ΔV
elstat
ΔEoi
ΔEint
ΔEprep
ΔE
ΔG
ΔG
CH2Cl2
[a] Computed at ZORA-BP86/TZ2P using COSMO to estimate
solvent effects by CH2Cl2.
The experimental trend, however, can be understood, if
we take into account the intermediate species I potentially
involved in the equilibrium shown in Scheme 5. In I, the
Si–X bond of b has already been heterolytically broken, and
the new X–C bond in bЈ (or bЈЈ) is not yet formed. The
intermediate 1I (X = S) is comparatively low in energy, at
ΔG = 15.9 kcal/mol (in CH2Cl2), whereas it is prohibitively
high for both 2I (X = O) and 3I (X = NMe), at 25.9 and
30.7 kcal/mol (in CH2Cl2), respectively (see Table 6).
The results of the COSMO-ZORA-BP86/TZ2P calcula-
tions concerning the equilibrium between penta- and tetra-
coordination (Scheme 5) are summarized in Tables 6 and 7
and in Figure 6. Note that the experiments concerning this
issue were carried out in CH2Cl2, and our computations
therefore also simulate solvation in CH2Cl2 using the
COSMO model. The different behavior between 1b, on one
hand, and 2b and 3b, on the other hand, can not be ex-
plained by thermodynamic factors (see Table 6). The forma-
tion of the respective tetracoordinate species bЈ/bЈЈ is in all
cases endothermic, with the diastereomers bЈ being only
slightly (by up to ca. 1 kcal/mol) more stable than bЈЈ. In
3222
www.eurjic.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2012, 3216–3228

Neutral Pentacoordinate Silicon(IV) Complexes
Figure 6. X– lone-pair orbitals (HOMO–1 and HOMO) and LUMO of the cationic silicon-containing fragment for the true intermediate
1I (X = S) (upper) and for the separate fragments of the model system 1M (X = S) shown in Scheme 7 (below).
The fact that the formation of intermediate 1I is rela- the corresponding orbitals, as they emerge from our Kohn–
tively favorable compared to the intermediates 2I and 3I Sham MO analyses, are depicted for both, the two separate
can be directly traced to the strength of the Si–X bond and fragments of the model system 1M (X = S) used in the bond
electron-donating capacity or basicity of the X– group. analyses (Table 7) as well as the true intermediate 1I (X =
Thus, for the related model system M shown in Scheme 7, S). The corresponding orbitals for 2M and 2I (X = O) as
we computed heterolytic bond free energies ΔG in CH2Cl2 well as for 3M and 3I (X = NMe) (not depicted) are similar
of –7.3, –20.6, and –21.1 kcal/mol for the Si–S, Si–O, and in character. Importantly, however, the aforementioned
Si–N bonds, respectively (see Table 7). In the gas phase, the rank order of the lone-pair orbital energies is SϽOϽNMe
corresponding bond free energies are much stronger (due to (see εHOMO in Table 7). Consequently, the HOMO–LUMO
the absence of differential solvation of the charged frag- gap diminishes, which leads to an increasingly stabilizing
ments) but essentially the same trend emerges, that is, the orbital interaction ΔEoi and thus interaction energy ΔEint
Si–S bond is substantially weaker than the Si–O and Si–N along the series (see Table 7). Interestingly, this trend in in-
bonds.
teraction is also directly related to the basicity (proton af-
finity), alkyl cation affinity, and nucleophilicity of RX– rea-
gents.[12] The electrostatic attraction ΔVelstat also becomes
more stabilizing along X = S, O, and NMe (see Table 7).
This is not the result of an increase in the partial negative
charge on X, which is relatively constant and in fact even
slightly decreases. The trend in ΔVelstat follows from the sub-
stantial shortening of the Si–X bond from X = S to X = O
and the increasing steric contact between the fragments
from X = O to X = NMe (vide infra). The overall bond
energy ΔE follows the trend of the increasingly stabilizing
interaction energy ΔEint from X = S to X = O but not from
X = O to X = NMe. Here, we see the effect of introducing a
Scheme 7. Heterolytic Si–X dissociation (X = S, O, NMe) of the
model system M leading to separated charged fragments.
The gas-phase bond analyses show that the trend in ΔE methyl substituent at X. The concomitant increase in steric
essentially derives from two factors: (i) the increasing en- repulsion shows up not only in a higher Pauli repulsion
ergy of the PhX– lone-pair orbitals along X = S, O, and ΔEPauli (and a more stabilizing electrostatic attraction) but,
NMe and (ii) the introduction of an extra substituent on X in particular, in a substantially more destabilizing ΔEprep
in the case of X = NMe. The orbital interactions are domi- (see Table 7). The latter effect is simply the result of a par-
nated by the donor–acceptor interaction between the tial alleviation of the initial Pauli repulsion in the case of X
HOMO–1 and the HOMO of PhX– (X = S, O, NMe), = NMe by deforming the molecular fragments as to avoid
which has mainly X– lone-pair character, and the LUMO steric contact, inter alia, by bending away the methyl sub-
of the cationic silicon-containing fragment, which has a stituent. Note that the Si–N bond (1.74 Å) is also somewhat
large 3p-derived amplitude on the silicon atom. In Figure 6, longer than the Si–O bond (1.78 Å; see Table 7).
Eur. J. Inorg. Chem. 2012, 3216–3228
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3223

F. M. Bickelhaupt, R. Tacke et al.
FULL PAPER
The results of the COSMO-ZORA-BP86/TZ2P calcula- analogues of the already known silicon(IV) complexes 1a–
tions concerning the iodo/acetonitrile ligand exchange reac- 1c and 1e–1g that contain a tridentate S,N,O ligand, with a
tion associated with Scheme 6 are summarized in Table 8. soft sulfur ligand atom instead of a hard oxygen (series 2)
This issue turns out to be a more subtle phenomenon. As or nitrogen ligand atom (series 3). The silicon(IV) com-
can be seen from Table 8, the intrinsic behavior of this reac- plexes of series 2 and 3 are stable both in the solid state and
tion system is to have no exchange of the iodo ligand at all in solution (solvent, CD2Cl2).
because this reaction goes with charge separation, which is
Comparison of the respective S/O/NMe analogues of the
highly unfavorable in the gas phase, with Gibbs free ener- series 1–3 revealed insight into the impact of the soft sulfur
gies of reaction of 81.7–83.9 kcal/mol. Also, there is no pro- ligand atom (compared to the hard oxygen and nitrogen
nounced trend for the respective S/O/NMe analogues, only ligand atoms) on (i) the molecular structure in the crystal,
a slight preference for the iodo/acetonitrile exchange in the (ii) the NMR spectroscopic parameters in the solid state
case of 2d (X = O). This does not match the qualitative and in solution, and (iii) the chemical reactivity. All com-
trend observed in the experiment in which the ligand ex- pounds studied have a strongly distorted trigonal-bipyrami-
change occurs for 2d (X = O) and 3d (X = NMe) but not dal structure, with the N(imino) ligand atom and the
for 1d (X = S).
(pseudo)halogeno ligand in the two axial positions. How-
ever, the degree of distortion of the trigonal-bipyramidal
silicon coordination polyhedra differs significantly and de-
pends on the tridentate ligand, with the highest degree of
distortion observed for series 2 (O,N,O ligand), followed by
series 1 (S,N,O) and series 3 (N,N,O). There are only a few
exceptions from this general trend. Striking differences be-
tween the respective S/O/NMe analogues were also ob-
served in terms of chemical reactivity. While some com-
pounds of series 1 (1a, 1b, and 1f; S,N,O ligand) isomerize
upon dissolution in organic solvents (equilibrium between
the pentacoordinate silicon complex and two isomeric tetra-
Table 8. Computed relative Gibbs free energies ΔG [kcal/mol] for
the reactions associated with Scheme 6.[a]
Medium
Gas phase
CH3CN
System
ΔG
d
h
1 (X = S)
0
0
0
0
0
0
83.3
81.7
83.9
6.5
4.1
3.7
2 (X = O)
3 (X = NMe)
1 (X = S)
2 (X = O)
3 (X = NMe)
[a] Computed at ZORA-BP86/TZ2P using COSMO to estimate coordinate species), no such kind of isomerization could be
solvent effects by CH3CN.
observed for the respective analogues of series 2 (O,N,O)
and 3 (N,N,O). Quantum-chemical calculations show that
As can be seen from Table 8, solvation in acetonitrile
strongly stabilizes the situation of charge separation, that is,
after the iodo ligand has been exchanged for an acetonitrile
ligand. As a result, ligand exchange in acetonitrile as the
solvent is only slightly endothermic, with reaction Gibbs
free energies of 3.7–6.5 kcal/mol (Table 8). Absolute energy
differences are thus not much larger than in the gas phase
but on a relative scale they differ now significantly. Impor-
tantly, the qualitative trend after the incorporation of the
large solvent effects is now in the right order, namely: ex-
change should occur more readily for 2 (X = O) and 3 (X
= NMe) than for 1 (X = S), in accordance with the experi-
ment.
this can be ascribed to the smaller energy demand associ-
ated with heterolytic dissociation of the relatively weak Si–
S bond. The higher stability of the Si–O and Si–N bonds
with respect to heterolytic dissociation could be traced to
the higher energy of the O and N lone-pair orbitals, which
is also responsible for the higher basicity and alkyl cation
affinity of oxygen and nitrogen bases.[12] Another difference
between the S/O/NMe analogues concerns the iodosil-
icon(IV) complexes 1d, 2d, and 3d. The pentacoordinate
iodosilicon(IV) complex 1d (S,N,O ligand) could be synthe-
sized from the corresponding chlorosilicon(IV) complex 3b
as a stable crystalline product, whereas the analogous iodo-
silicon(IV) complex 2d (O,N,O) could not be prepared at
all, and the analogue 3d (N,N,O) could only be generated at
low temperature in solution and decomposed upon heating.
Obviously, the soft iodo ligand is tolerated in the presence
of the soft sulfur ligand atom (1d), whereas the combination
Conclusions
With the synthesis of compounds 2a, 2c, 2e–2g, 3a, 3c, of the hard oxygen (2d) or nitrogen ligand atom (3d) with
and 3e–3g, a series of novel neutral pentacoordinate sili- the soft iodo ligand is significantly less favored. Instead of
con(IV) complexes has been made available. These com- the neutral pentacoordinate iodosilicon(IV) complexes 2d
pounds contain an SiFO2NC, SiBrO2NC, SiO2NC2, and 3d, the monocationic pentacoordinate silicon(IV) com-
SiO2N2C, SiFON2C, SiBrON2C, SiON2C2, or SiON3C plexes 2h and 3h (with an iodide counterion each) and the
skeleton. The (pseudo)halogenosilicon(IV) complexes 2a, dicationic tetracoordinate dinuclear silicon(IV) complex 2i
2c, 2e–2g, 3a, 3c, and 3e–3g were synthesized from the cor- (with two iodide counterions) were obtained. That is, re-
responding chlorosilicon(IV) complexes 2b (SiClO2NC placement of the soft sulfur ligand atom of 1d by a hard
skeleton) and 3b (SiClON2C), respectively. All these com- oxygen (2d) or nitrogen ligand atom (3d) leads to a dramatic
pounds contain a tridentate dianionic O,N,O (series 2) or decrease in stability. This is not what one would expect if
N,N,O ligand (series 3), a phenyl ligand, and a (pseudo)- it is assumed that hard ligand atoms (such as oxygen and
halogeno ligand (F, Cl, Br, CN, N3, NCS). They represent nitrogen) should favor pentacoordination of the hard sili-
3224
www.eurjic.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2012, 3216–3228

Neutral Pentacoordinate Silicon(IV) Complexes
Synthesis of 2c: Bromotrimethylsilane (306 mg, 2.00 mmol) was
added in a single portion at 20 °C to a stirred suspension of 2b
(547 mg, 1.66 mmol) in acetonitrile (10 mL). The reaction mixture
was heated to 95 °C and then concentrated to a volume of ca. 6 mL
by distillation at ambient pressure. The remaining solution was
slowly cooled to 20 °C (formation of crystals) and then kept undis-
turbed at this temperature for 1 h and at –20 °C for a further 3 h.
The resulting precipitate was isolated by filtration, washed with
diethyl ether (10 mL), and dried in vacuo (0.01 mbar, 20 °C, 5 h)
to give 2c in 61% yield (383 mg, 1.02 mmol) as a yellow crystalline
solid; m.p. Ͼ120 °C (decomp.). For analytical data, see the Sup-
porting Information.
con(IV) coordination center (expected rank order of sta-
bility: OϾNϾS). Obviously, this concept is too simple to
explain the experimental results obtained in this study.
Experimental Section
General Procedures: All syntheses were carried out under dry nitro-
gen. The organic solvents used were dried and purified according
to standard procedures and stored under nitrogen. Melting points
were determined with a Büchi Melting Point B-540 apparatus using
samples in sealed capillaries. The solution-state 1H, 13C, 19F, and
29Si NMR spectra were recorded at 23 °C by using a Bruker DRX-
300 (1H, 300.1 MHz; 13C, 75.5 MHz; 19F, 282.4 MHz; 29Si,
59.6 MHz) or a Bruker Avance 500 NMR spectrometer (1H,
500.1 MHz; 13C, 125.8 MHz; 29Si, 99.4 MHz). CD2Cl2 or CD3CN
served as the solvent. Chemical shifts (ppm) were determined rela-
tive to internal CHDCl2 (1H, δ = 5.32 ppm; CD2Cl2), CHD2CN
(1H, δ = 1.93 ppm; CD3CN), CD2Cl2 (13C, δ = 53.8 ppm; CD2Cl2),
CD3CN (13C, δ = 1.3 ppm; CD3CN), or external TMS (29Si, δ =
0 ppm; CD2Cl2, CD3CN). Assignment of the 13C NMR spectro-
scopic data was supported by DEPT 135 experiments and 13C,1H
correlation experiments. Solid-state 13C, 15N, and 29Si VACP/MAS
NMR spectra were recorded at 22 °C by using a Bruker DSX-400
NMR spectrometer with bottom layer rotors of ZrO2 (diameter,
7 mm) containing ca. 300 mg of sample (13C, 100.6 MHz; 15N,
40.6 MHz; 29Si, 79.5 MHz; external standard, TMS (13C, 29Si; δ =
0 ppm) or glycine (15N, δ = –342.0 ppm); spinning rate, 5–7 kHz;
Synthesis of 2e: Cyanotrimethylsilane (339 mg, 3.42 mmol) was
added in a single portion at 20 °C to a stirred suspension of 2b
(950 mg, 2.88 mmol) in acetonitrile (20 mL), and the reaction mix-
ture was then stirred at 20 °C for 1.5 h. The volatile components
of the reaction mixture were removed in vacuo, acetonitrile (8 mL)
was added to the residue, and the resulting suspension was heated
until a clear solution was obtained. This solution was allowed to
cool to 20 °C and then kept undisturbed at this temperature for 1 h
and at –20 °C for a further 16 h. The resulting precipitate was iso-
lated by filtration, washed with diethyl ether (10 mL), and dried
in vacuo (0.01 mbar, 20 °C, 3 h) to give 2e in 64% yield (593 mg,
1.85 mmol) as a red to orange-colored crystalline solid; m.p.
Ͼ155 °C (decomp.). For analytical data, see the Supporting Infor-
mation.
Synthesis of 2f: Azidotrimethylsilane (324 mg, 2.81 mmol) was
added in a single portion at 20 °C to a stirred suspension of 2b
(773 mg, 2.34 mmol) in acetonitrile (10 mL). The reaction mixture
was heated to 90 °C and then concentrated to a volume of ca. 7 mL
by distillation at ambient pressure. The remaining solution was
slowly cooled to 20 °C and then kept undisturbed at this tempera-
ture for 24 h and at –20 °C for a further 3 h. The resulting precipi-
tate was isolated by filtration, washed with diethyl ether (15 mL),
and dried in vacuo (0.01 mbar, 20 °C, 3 h) to give 2f in 79% yield
(623 mg, 1.85 mmol) as a yellow crystalline solid; m.p. Ͼ120 °C
(decomp.). For analytical data, see the Supporting Information.
1
contact time, 1 ms (13C), 3 ms (15N), or 5 ms (29Si); 90° H trans-
mitter pulse length, 3.6 μs; repetition time, 4 s).
Synthesis of 2a: Freshly sublimed ammonium fluoride (119 mg,
3.18 mmol) was added in a single portion at 20 °C to a stirred solu-
tion of 2b (1.00 g, 3.03 mmol) in tetrahydrofuran (20 mL), and the
reaction mixture was then stirred at 20 °C for 48 h. The resulting
solid was filtered off, washed with tetrahydrofuran (5 mL), and dis-
carded. The solvent of the filtrate (including the wash solution) was
removed in vacuo, toluene (8 mL) was added to the residue, and
the resulting suspension was heated until a clear solution was ob-
tained. This solution was allowed to cool to 20 °C and was then
kept undisturbed at this temperature for 4 h and at –20 °C for a
further 16 h. The resulting precipitate was isolated by filtration,
washed with n-pentane (8 mL), and dried in vacuo (0.01 mbar,
20 °C, 4 h) to give 2a in 57% yield (538 mg, 1.72 mmol) as a yellow
Synthesis of 2g: (Thiocyanato-N)trimethylsilane (560 mg,
4.27 mmol) was added in a single portion at 20 °C to a stirred sus-
pension of 2b (1.41 g, 4.27 mmol) in acetonitrile (20 mL), and the
reaction mixture was then stirred at 20 °C for 2 h. The volatile com-
ponents of the reaction mixture were removed in vacuo, acetonitrile
(18 mL) was added to the residue, and the resulting suspension was
heated until a clear solution was obtained. This solution was al-
lowed to cool to 20 °C and then kept undisturbed at this tempera-
ture for 2 h and at –20 °C for a further 16 h. The resulting precipi-
tate was isolated by filtration, washed with n-pentane (20 mL), and
dried in vacuo (0.01 mbar, 20 °C, 2 h) to give 2g in 88% yield
(1.32 g, 3.74 mmol) as a yellow crystalline solid; m.p. Ͼ216 °C (de-
comp.). For analytical data, see the Supporting Information.
1
solid; m.p. Ͼ160 °C (decomp.). H NMR (CD2Cl2, 500.1 MHz): δ
4
= 2.18 [d, J(1H,1H) = 0.5 Hz, 3 H, CCH3], 2.52 (s, 3 H, CCH3),
5.66–5.69 (m, 1 H, CCHC), 6.92–6.97, 7.10–7.13, 7.19–7.24, 7.27–
7.32, 7.35–7.40, and 7.44–7.48 ppm (m, 9 H, C6H5, C6H4). 13C
NMR (CD2Cl2, 125.8 MHz): δ = 24.3 (CCH3), 24.9 (CCH3), 106.2
[d, 4J(13C,19F) = 0.5 Hz, CCHC], 115.8 [d, J(13C,19F) = 0.8 Hz],
4
120.5, 120.7, 127.9 (2 C) [d, J(13C,19F) = 0.7 Hz], 128.8, 130.1 [d,
J(13C,19F) = 0.6 Hz], 132.7, 134.5 (2 C) [d, 3J(13C,19F) = 3.5 Hz],
2
135.5 [d, J(13C,19F) = 32.5 Hz], and 151.8 [d, J(13C,19F) = 7.7 Hz]
Synthesis of 2h·CH3CN: Iodotrimethylsilane (350 mg, 1.75 mmol)
was added in a single portion at 20 °C to a stirred suspension of
2b (503 mg, 1.52 mmol) in acetonitrile (5 mL), and the resulting
mixture was stirred at 20 °C for 1 h. The volatile components of
the reaction mixture were removed in vacuo, acetonitrile (4 mL)
was added to the residue, and the resulting suspension was heated
until a clear solution was obtained. This solution was allowed to
cool to 20 °C and then kept undisturbed at this temperature for 1 h
and at 4 °C for a further 24 h. The resulting precipitate was isolated
by filtration, washed with n-pentane (6 mL), and dried in vacuo
(0.01 mbar, 20 °C, 3 h) to give 2h·CH3CN in 44% yield (339 mg,
673 μmol) as a yellow crystalline solid; m.p. Ͼ85 °C (decomp.). 1H
(C6H5, C6H4), 168.9 (NCCH3 or OCCH3), 172.3 ppm [d, J(13C,19F)
= 7.9 Hz, NCCH3 or OCCH3]. 19F NMR (CD2Cl2, 282.4 MHz): δ
= –118.1 ppm. 29Si NMR (CD2Cl2, 99.4 MHz): δ = –109.3 ppm [d,
1J(29Si,19F) = 204.8 Hz]. 13C VACP/MAS NMR: δ = 23.8 (CCH3),
24.6 (CCH3), 106.9 (CCHC), 116.6, 119.0 (2 C), 121.0, 128.7 (2 C),
129.8, 131.5 (3 C), 136.6, and 151.4 (C6H5, C6H4), 166.8 (NCCH3
or OCCH3), 167.8 ppm (NCCH3 or OCCH3). 15N VACP/MAS
NMR: δ = –160.4 ppm. 29Si VACP/MAS NMR: δ = –108.5 ppm
1
[d, J(29Si,19F) = 206 Hz]. C17H16FNO2Si (313.40): calcd. C 65.15,
H 5.15, N 4.47; found C 64.8, H 5.4, N 4.5.
Synthesis of 2b: Compound 2b was synthesized according to ref.[4]
Eur. J. Inorg. Chem. 2012, 3216–3228
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3225

F. M. Bickelhaupt, R. Tacke et al.
FULL PAPER
NMR (CD3CN, 500.1 MHz):[13] δ = 1.95 (s, 6 H, NϵCCH3), 2.32
(30 mL) and triethylamine (465 mg, 4.60 mmol), and the reaction
mixture was then stirred at 20 °C for 1 h. The resulting solid was
filtered off, washed with tetrahydrofuran (5 mL), and discarded.
The solvent of the filtrate (including the wash solution) was re-
4
[d, J(1H,1H) = 0.5 Hz, 3 H, CCH3], 2.59 (s, 3 H, CCH3), 6.14 [q,
4J(1H,1H) = 0.5 Hz, 1 H, CCHC], 7.09–7.16, 7.19–7.24, 7.30–7.48,
and 7.62–7.67 ppm (m, 9 H, C6H5, C6H4). 13C NMR (CD3CN,
125.8 MHz):[13] δ = 1.75 (2 C) (NϵCCH3), 23.9 (CCH3), 25.1 moved in vacuo, acetonitrile (11 mL) was added to the residue, and
(CCH3), 109.0 (CCHC), 132.0 (2 C) (br., NϵCCH3), 116.3, 122.3, the resulting suspension was heated until a clear solution was ob-
123.4, 129.2 (2 C), 130.8, 132.0, 132.3, 132.5, 133.5 (2 C), and 149.7
tained. This solution was allowed to cool to 20 °C and was then
(C6H5, C6H4), 172.4 (NCCH3 or OCCH3), 174.3 ppm (NCCH3 or
kept undisturbed at this temperature for 2 h and at –20 °C for a
OCCH3). 29Si NMR (CD3CN, 99.4 MHz):[13] δ = –103.2 ppm. further 19 h. The resulting precipitate was isolated by filtration,
C21H22IN3O2Si (503.41): calcd. C 50.10, H 4.40, N 8.35; found C
49.8, H 4.5, N 8.5.
washed with diethyl ether (10 mL), and dried in vacuo (0.01 mbar,
20 °C, 5 h) to give 3a in 61% yield (918 mg, 2.81 mmol) as a yellow
crystalline solid. The analytical data of the product match with
those obtained for the product synthesized according to method A.
Synthesis of 2i: Iodotrimethylsilane (374 mg, 1.87 mmol) was added
in a single portion at 20 °C to a stirred suspension of 2b (515 mg,
1.56 mmol) in dichloromethane (7 mL), and the resulting mixture
was stirred for 10 min and then kept undisturbed at 20 °C for 20 h
excluding light. The resulting precipitate was isolated by filtration,
washed with diethyl ether (10 mL), and dried in vacuo (0.01 mbar,
20 °C, 3 h) to give 2i in 58% yield (383 mg, 455 μmol) as a yellow
crystalline solid; m.p. Ͼ190 °C (decomp.). 13C VACP/MAS NMR:
δ = 23.9 (CCH3), 27.4 (CCH3), 112.1 (CCHC), 119.3, 121.2, 125.0,
129.5 (3 C), 132.0, 134.0 (3 C), 135.9, and 146.4 (C6H5, C6H4),
178.3 (NCCH3 or OCCH3), 189.3 ppm (NCCH3 or OCCH3). 15N
VACP/MAS NMR: δ = –208.3 ppm. 29Si VACP/MAS NMR: δ =
–56.9 ppm. C34H32I2N2O4Si2 (842.62): calcd. C 48.46, H 3.83, N
3.32; found C 48.1, H 3.8, N 3.4.
Synthesis of 3b: Compound 3b was synthesized according to ref.[4]
Synthesis of 3c·0.5C6H6: Bromotrimethylsilane (1.69 g, 11.0 mmol)
was added in a single portion at 20 °C to a stirred suspension of
3b (3.16 g, 9.22 mmol) in benzene (30 mL). The reaction mixture
was slowly heated from 20 °C to 85 °C within 5 h and then concen-
trated to a volume of ca. 30 mL by distillation at ambient pressure.
The remaining solution was cooled to 20 °C, and bromotrimethyl-
silane (1.69 g, 11.0 mmol) was added again in a single portion. The
reaction mixture was heated slowly from 20 °C to 85 °C within 5 h,
and the volatile components were removed by distillation at ambi-
ent pressure. Subsequently, benzene (20 mL) was added to the resi-
due, and the resulting suspension was heated until a clear solution
was obtained. This solution was allowed to cool to 20 °C and then
kept undisturbed at this temperature for 24 h. The resulting precipi-
tate was isolated by filtration, washed with n-pentane (10 mL), and
dried in vacuo (0.01 mbar, 20 °C, 4 h) to give 3c·0.5C6H6 in 84%
yield (3.00 g, 7.74 mmol) as an orange-colored crystalline solid;
m.p. Ͼ170 °C (decomp.). For analytical data, see the Supporting
Information.
Synthesis of 3a. Method A: Freshly sublimed ammonium fluoride
(95.6 mg, 2.58 mmol) was added in a single portion at 20 °C to a
stirred solution of 3b (883 mg, 2.58 mmol) in tetrahydrofuran
(20 mL), and the reaction mixture was then stirred at 20 °C for
20 h. The resulting solid was filtered off, washed with tetra-
hydrofuran (5 mL), and discarded. The solvent of the filtrate (in-
cluding the wash solution) was removed in vacuo, acetonitrile
(9 mL) was added to the residue, and the resulting suspension was
heated until a clear solution was obtained. This solution was al-
lowed to cool to 20 °C and then kept undisturbed at this tempera-
ture for 3 h and at –20 °C for a further 24 h. The resulting precipi-
tate was isolated by filtration, washed with diethyl ether (10 mL),
and dried in vacuo (0.01 mbar, 20 °C, 4 h) to give 3a in 49% yield
(410 mg, 1.26 mmol) as a yellow solid; m.p. Ͼ143 °C (decomp.). 1H
NMR (500.1 MHz, CD2Cl2): δ = 2.16 [d, 4J(1H,1H) = 0.6 Hz, 3 H,
CCH3], 2.47 (s, 3 H, CCH3), 3.14 (s, 3 H, NCH3), 5.60 (m, 1 H,
CCHC), 6.72–6.77, 6.82–6.86, 7.13–7.29, and 7.36–7.39 ppm (m, 9
H, C6H5, C6H4). 13C NMR (125.8 MHz, CD2Cl2): δ = 24.1
(CCH3), 25.1 (CCH3), 33.0 [d, 3J(13C,19F) = 12.4 Hz, NCH3], 106.1
[d, 4J(13C,19F) = 0.8 Hz, CCHC], 110.8 [d, J(13C,19F) = 2.8 Hz],
Synthesis of 3d: In an NMR tube, iodotrimethylsilane (58 mg,
290 μmol) was added in a single portion at –80 °C to a solution
of 3b (33 mg, 96 μmol) in [D2]dichloromethane (750 μL), and the
resulting mixture was studied immediately by NMR spectroscopy.
1H NMR (500.1 MHz, CD2Cl2, –80 °C): δ = 2.47 (s, 3 H, CCH3),
2.65 (s, 3 H, CCH3), 3.24 (s, 3 H, NCH3), 6.43 (s, 1 H, CCHC),
6.92–6.98, 7.02–7.07, 7.25–7.29, 7.30–7.36, 7.37–7.42, and 7.44–
7.50 (m, 9 H, C6H5, C6H4). 13C NMR (125.8 MHz, CD2Cl2): δ =
23.8 (CCH3), 25.0 (CCH3), 32.7 (NCH3), 108.3 (CCHC), 111.2,
118.6, 120.3, 128.1 (2 C), 128.5, 129.4, 130.5, 131.6, 132.4 (2 C),
and 142.3 (C6H5, C6H4), 169.9 (NCCH3 or OCCH3), 171.4 ppm
(NCCH3 or OCCH3). 29Si NMR (99.6 MHz, CD2Cl2): δ =
–81.9 ppm.
4
116.4, 120.1, 127.6 (2 C) [d, J(13C,19F) = 0.8 Hz], 128.4, 129.3 [d,
Synthesis of 3e: Cyanotrimethylsilane (210 mg, 2.12 mmol) was
added in a single portion at 20 °C to a stirred suspension of 3b
(662 mg, 1.93 mmol) in acetonitrile (15 mL), and the reaction mix-
ture was then stirred at 20 °C for 24 h. The volatile components of
the reaction mixture were removed in vacuo, acetonitrile (3 mL)
was added to the residue, and the resulting suspension was heated
until a clear solution was obtained. This solution was allowed to
cool to 20 °C and then kept undisturbed at this temperature for 3 h
and at –20 °C for a further 24 h. The resulting precipitate was iso-
lated by filtration, washed with n-pentane (5 mL), and dried in
vacuo (0.01 mbar, 20 °C, 4 h) to give 3e in 81% yield (521 mg,
1.56 mmol) as an orange-colored crystalline solid; m.p. Ͼ137 °C
(decomp.). For analytical data, see the Supporting Information.
J(13C,19F) = 1.0 Hz], 131.0, 134.0 (2 C) [d, 3J(13C,19F) = 3.9 Hz],
2
138.8 [d, J(13C,19F) = 36.1 Hz], and 146.1 [d, J(13C,19F) = 3.6 Hz]
(C6H5, C6H4), 166.8 (NCCH3 or OCCH3), 170.2 ppm [d, J(13C,19F)
= 7.8 Hz, NCCH3 or OCCH3]. 19F NMR (282.4 MHz, CD2Cl2): δ
= –112.4 ppm. 29Si NMR (99.4 MHz, CD2Cl2): δ = –108.1 ppm [d,
1J(29Si,19F) = 222 Hz]. 13C VACP/MAS NMR: δ = 24.9 (CCH3),
26.3 (CCH3), 33.1 (NCH3), 107.9 (CCHC), 109.6, 116.1, 118.9,
127.2 (3 C), 129.7, 131.7, 133.5, 136.0, 141.4, and 145.4 (C6H5,
C6H4), 169.2 (NCCH3 or OCCH3), 170.2 ppm (NCCH3 or
OCCH3). 15N VACP/MAS NMR: δ = –306.2 (NCH3), –173.1 ppm
1
(NCCH3). 29Si VACP/MAS NMR: δ = –111.5 ppm [d, J(29Si,19F)
= 227 Hz]. C18H19FN2OSi (326.45): calcd. C 66.23, H 5.87, N 8.58;
found C 65.73, H 5.85, N 8.76.
Synthesis of 3f: Azidotrimethylsilane (117 mg, 1.02 mmol) was
Methode B: Silver tetrafluoroborate (895 mg, 4.60 mmol) was added in a single portion at 20 °C to a stirred suspension of 3b
added in a single portion at 20 °C with exclusion of light to a
stirred solution of 3b (1.58 g, 4.61 mmol) in tetrahydrofuran
(348 mg, 1.01 mmol) in acetonitrile (10 mL), and the reaction mix-
ture was then stirred at 20 °C for 2 h. The volatile components of
3226
www.eurjic.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2012, 3216–3228

Neutral Pentacoordinate Silicon(IV) Complexes
the reaction mixture were removed in vacuo, acetonitrile (10 mL)
was added to the residue, and the resulting suspension was heated
until a clear solution was obtained. This solution was allowed to
cool to 20 °C and then kept undisturbed at this temperature for 2 h
and at –20 °C for a further 24 h. The resulting precipitate was iso-
lated by filtration, washed with n-pentane (8 mL), and dried in
vacuo (0.01 mbar, 20 °C, 5 h) to give 3f in 88% yield (310 mg,
887 μmol) as an orange-colored crystalline solid; m.p. Ͼ172 °C (de-
comp.). For analytical data, see the Supporting Information.
CCDC-873533 (for 2a), -873534 (for 2c), -874092 (for 2e), -873535
(for 2f), -873536 (for 2g), -873537 (for 2h·CH3CN), -873538 (for
2i), -873539 (for 3a), -873540 (for 3c·0.5C6H6), -873541 (for 3e),
-873542 (for 3f), -873543 (for 3g), and -873544 (for 3h·2CH3CN)
contain the supplementary crystallographic data for this paper.
These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
Supporting Information (see footnote on the first page of this arti-
cle): Crystallographic data for compounds 2a, 2c, 2e–2g,
2h·CH3CN, 2i, 3a, 3c·0.5C6H6, 3e–3g, and 3h·2CH3CN. Molecular
structures of 2a, 2e–2g, 2i (cation), and 3c·0.5C6H6, 3e–3g. NMR
spectroscopic data and elemental analyses for compounds 2c, 2e–
2g, 3c·0.5C6H6, and 3e–3g. Cartesian coordinates and total energies
for 1b, 2b, 3b, 1bЈ, 1bЈЈ, 2bЈ, 2bЈЈ, 3bЈ, 3bЈЈ, 1d, 1h, 2d, 2h, 3d, and
3h.
Synthesis of 3g: (Thiocyanato-N)trimethylsilane (567 mg,
4.32 mmol) was added in a single portion at 20 °C to a stirred sus-
pension of 3b (1.48 g, 4.32 mmol) in acetonitrile (30 mL), and the
reaction mixture was then stirred at 20 °C for 16 h. The volatile
components of the reaction mixture were removed in vacuo, aceto-
nitrile (12 mL) was added to the residue, and the resulting suspen-
sion was heated until a clear solution was obtained. This solution
was allowed to cool to 20 °C and then kept undisturbed at this
temperature for 24 h and at –20 °C for a further 16 h. The resulting
precipitate was isolated by filtration, washed with diethyl ether
(9 mL), and dried in vacuo (0.01 mbar, 20 °C, 4 h) to give 3g in
70% yield (1.10 g, 3.01 mmol) as a yellow crystalline solid; m.p.
Ͼ165 °C (decomp.). For analytical data, see the Supporting Infor-
mation.
Acknowledgments
R. T. thanks the State of Bavaria for financial support. C. F. G.
and F. M. B. thank the National Research School Combination–
Catalysis for a postdoctoral NRSC-C fellowship for C. F. G. and
the Netherlands Organization for Scientific Research (NWO-CW
and NWO-NCF) for further financial support.
Synthesis of 3h: Iodotrimethylsilane (1.47 g, 7.35 mmol) was added
in a single portion at 20 °C to a stirred suspension of 3b (2.51 g,
7.32 mmol) in acetonitrile (50 mL), and the resulting mixture was
stirred at 20 °C for 1 h and then kept undisturbed at –20 °C for
24 h. The resulting precipitate was isolated by filtration, washed
with n-pentane (20 mL), and dried in vacuo (0.01 mbar, 20 °C, 6 h)
to give 3h in 64% yield (2.24 g, 4.71 mmol) as an orange-colored
solid; m.p. Ͼ107 °C (decomp.). 1H NMR (500.1 MHz, CD3CN):[13]
δ = 1.95 (s, 3 H, NϵCCH3), 2.34 [d, 4J(1H,1H) = 0.6 Hz, 3 H,
CCH3], 2.50 (s, 3 H, CCH3), 2.28 (s, 3 H, NCH3), 6.06 [d, 4J(1H,1H)
= 0.6 Hz, 1 H, CCHC], 6.88–6.92, 7.07–7.11, 7.25–7.29, 7.34–7.38,
and 7.46–7.49 ppm (m, 9 H, C6H5, C6H4). 13C NMR (125.8 MHz,
CD3CN):[13] δ = 1.79 (NϵCCH3), 23.8 (CCH3), 24.9 (CCH3), 33.8
(NCH3), 108.5 (CCHC), 118.2 (NϵCCH3), 113.1, 119.7, 121.7,
129.0 (2 C), 130.0, 130.4, 131.5, 132.6 (2 C), 135.2, and 144.0
(C6H5, C6H4), 171.0 (NCCH3 or OCCH3), 171.9 ppm (NCCH3 or
OCCH3). 29Si NMR (99.4 MHz, CD3CN):[13] δ = –96.4 ppm.
C20H22IN3OSi (475.40): calcd. C 50.53, H 4.66, N 8.84; found C
50.9, H 4.8, N 9.0.
[1] For selected reviews dealing with higher-coordinate silicon(IV)
compounds, see: a) R. R. Holmes, Chem. Rev. 1996, 96, 927–
950; b) V. Pestunovich, S. Kirpichenko, M. Voronkov, in: The
Chemistry of Organic Silicon Compounds, vol. 2, part 2 (Eds.:
Z. Rappoport, Y. Apeloig), Wiley, Chichester, 1998, pp. 1447–
1537; c) C. Chuit, R. J. P. Corriu, C. Reye, in: Chemistry of
Hypervalent Compounds (Ed.: K.-y. Akiba), Wiley-VCH, New
York, 1999, pp. 81–146; d) R. Tacke, M. Pülm, B. Wagner, Adv.
Organomet. Chem. 1999, 44, 221–273; e) M. A. Brook, Silicon
in Organic, Organometallic, and Polymer Chemistry, Wiley, New
York, 2000, pp. 97–114; f) R. Tacke, O. Seiler, in: Silicon Chem-
istry: From the Atom to Extended Systems (Eds.: P. Jutzi, U.
Schubert), Wiley-VCH, Weinheim, Germany, 2003, pp. 324–
337; g) D. Kost, I. Kalikhman, Adv. Organomet. Chem. 2004,
5, 1–106; h) D. Kost, I. Kalikhman, Acc. Chem. Res. 2009, 42,
303–314; i) E. P. A. Couzijn, J. C. Slootweg, A. W. Ehlers, K.
Lammertsma, Z. Anorg. Allg. Chem. 2009, 635, 1273–1278.
[2] For selected recent publications dealing with higher-coordinate
silicon(IV) compounds, see: a) S. Yakubovich, I. Kalikhman,
D. Kost, Dalton Trans. 2010, 39, 9241–9244; b) K. Junold, C.
Burschka, R. Bertermann, R. Tacke, Dalton Trans. 2010, 39,
9401–9413; c) P. Bombicz, I. Kovács, L. Nyulászi, D. Szieberth,
P. Terleczky, Organometallics 2010, 29, 1100–1106; d) E.
Kertsnus-Banchik, B. Gostevskii, M. Botoshansky, I. Kalikh-
man, D. Kost, Organometallics 2010, 29, 5435–5445; e) S. Cota,
M. Beyer, R. Bertermann, C. Burschka, K. Götz, M. Kaupp,
R. Tacke, Chem. Eur. J. 2010, 16, 6582–6589; f) S. Metz, B.
Theis, C. Burschka, R. Tacke, Chem. Eur. J. 2010, 16, 6844–
6856; g) R. S. Ghadwal, S. S. Sen, H. W. Roesky, M. Granitzka,
D. Kratzert, S. Merkel, D. Stalke, Angew. Chem. 2010, 122,
4044–4047; Angew. Chem. Int. Ed. 2010, 49, 3952–3955; h) D.
Schöne, D. Gerlach, C. Wiltzsch, E. Brendler, T. Heine, E.
Kroke, J. Wagler, Eur. J. Inorg. Chem. 2010, 461–467; i) A. R.
Bassindale, M. Sohail, P. G. Taylor, A. A. Korlyukov, D. E. Ar-
khipov, Chem. Commun. 2010, 46, 3274–3276; j) R. S. Ghad-
wal, K. Pröpper, B. Dittrich, P. G. Jones, H. W. Roesky, Inorg.
Chem. 2011, 50, 358–364; k) S. Yakubovich, B. Gostevskii, I.
Kalikhman, M. Botoshansky, L. E. Gusel’nikov, V. A. Pestuno-
vich, D. Kost, Organometallics 2011, 30, 405–413; l) K. Junold,
C. Burschka, R. Bertermann, R. Tacke, Dalton Trans. 2011,
40, 9844–9857; m) C. Kobelt, C. Burschka, R. Bertermann, C.
Crystal Structure Analyses: Suitable single crystals of compounds
2a, 2c, 2e–2g, 2h·CH3CN, 2i, 3a, 3c·0.5C6H6, 3e–3g, and
3h·2CH3CN were isolated directly from the respective reaction mix-
tures. The crystals were mounted in inert oil (perfluoropolyalkyl
ether, ABCR) on a glass fiber and then transferred to the cold
nitrogen gas stream of the diffractometer (3a and 3g: Bruker Non-
ius KAPPA APEX II diffractometer, Montel mirror, Mo-Kα radia-
tion, λ = 0.71073 Å; 2a, 2c, 2e–2g, 2h·CH3CN, 2i, 3c·0.5C6H6, 3e–
3f, and 3h·2CH3CN: Stoe IPDS diffractometer, graphite-mono-
chromated Mo-Kα radiation, λ = 0.71073 Å). All structures were
solved by direct methods.[14] For compound 2h·CH3CN the PLA-
TON routine “SQUEEZE” was used to account for regions of dis-
ordered solvent in the structure.[15] In all the structures, the non-
hydrogen atoms were refined anisotropically, and a riding model
was employed in the refinement of the CH hydrogen atoms.[14] Ac-
cording to the X-ray data, the cyano ligand in 2e is disordered (in
this context, see also ref.[16]). Best refinement results are obtained
when 70% of the coordinating atom is nitrogen and 30% is carbon
(R1 = 0.039). Although the 15N NMR spectroscopic data do not
indicate any N coordination (Si–NC), the refinement with 100% C
coordination (Si–CN) results in a higher residual of R1 = 0.045.
Eur. J. Inorg. Chem. 2012, 3216–3228
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3227

F. M. Bickelhaupt, R. Tacke et al.
FULL PAPER
Fonseca Guerra, F. M. Bickelhaupt, R. Tacke, Dalton Trans.
2012, 41, 2148–2162; n) K. Junold, C. Burschka, R. Tacke, Eur.
J. Inorg. Chem. 2012, 189–193; o) B. Theis, J. Weiß, W. P. Lip-
pert, R. Bertermann, C. Burschka, R. Tacke, Chem. Eur. J.
2012, 18, 2202–2206.
[9]
a) A. Klamt, G. Schüürmann, J. Chem. Soc. Perkin Trans. 2
1993, 799–805; b) A. Klamt, J. Phys. Chem. 1995, 99, 2224–
2235; c) C. C. Pye, T. Ziegler, Theor. Chem. Acc. 1999, 101,
396–408; d) M. Swart, E. Rösler, F. M. Bickelhaupt, Eur. J.
Inorg. Chem. 2007, 3646–3654.
[3] a) R. Willeke, R. Tacke, Z. Anorg. Allg. Chem. 2001, 627, 1537–
1541; b) R. Tacke, M. Mallak, R. Willeke, Angew. Chem. 2001,
113, 2401–2403; Angew. Chem. Int. Ed. 2001, 40, 2339–2341;
c) R. Bertermann, A. Biller, M. Kaupp, M. Penka, O. Seiler,
R. Tacke, Organometallics 2003, 22, 4104–4110; d) O. Seiler,
M. Büttner, M. Penka, R. Tacke, Organometallics 2005, 24,
6059–6062; e) D. Troegel, C. Burschka, S. Riedel, M. Kaupp,
R. Tacke, Angew. Chem. 2007, 119, 7131–7135; Angew. Chem.
Int. Ed. 2007, 46, 7001–7005; f) S. Metz, C. Burschka, D. Platte,
R. Tacke, Angew. Chem. 2007, 119, 7136–7139; Angew. Chem.
Int. Ed. 2007, 46, 7006–7009; g) S. Metz, C. Burschka, R.
Tacke, Eur. J. Inorg. Chem. 2008, 4433–4439; h) S. Metz, C.
Burschka, R. Tacke, Organometallics 2009, 28, 2311–2317.
[4] B. Theis, S. Metz, F. Back, C. Burschka, R. Tacke, Z. Anorg.
Allg. Chem. 2009, 635, 1306–1312.
[10]
a) F. M. Bickelhaupt, E. J. Baerends, in: Reviews in Computa-
tional Chemistry, vol. 15 (Eds.: K. B. Lipkowitz, D. B. Boyd),
Wiley-VCH, New York, 2000, pp. 1–86; b) F. M. Bickelhaupt,
A. Diefenbach, S. P. de Visser, L. J. de Koning, N. M. M. Nibb-
ering, J. Phys. Chem. A 1998, 102, 9549–9553.
a) F. M. Bickelhaupt, N. J. R. van Eikema Hommes, C. Fon-
seca Guerra, E. J. Baerends, Organometallics 1996, 15, 2923–
2931; b) C. Fonseca Guerra, J.-W. Handgraaf, E. J. Baerends,
F. M. Bickelhaupt, J. Comput. Chem. 2004, 25, 189–210.
a) J. M. Ruiz, R. J. Mulder, C. Fonseca Guerra, F. M. Bickel-
haupt, J. Comput. Chem. 2011, 32, 681–688; b) R. J. Mulder,
C. Fonseca Guerra, F. M. Bickelhaupt, J. Phys. Chem. A 2010,
114, 7604–7608; c) A. P. Bento, F. M. Bickelhaupt, J. Org.
Chem. 2008, 73, 7290–7299; d) M. Swart, F. M. Bickelhaupt,
J. Chem. Theory Comput. 2006, 2, 281–287.
[11]
[12]
[5] J. Böhm, D. Fenzke, H. Pfeifer, J. Magn. Reson. 1983, 55, 197–
204.
[13] The data given refer to the respective isotopomer with an
SiNϵCCD3 moiety resulting from a ligand exchange (CH3CN/
CD3CN) in solution (solvent, CD3CN).
[14] G. M. Sheldrick, Acta Crystallogr., Sect. A 2008, 64, 112–122.
[15] A. L. Spek, PLATON, Utrecht University, The Netherlands,
2008.
[16] J. K. Rasmussen, S. M. Heilmann, L. R. Krepski, in: Advances
in Silicon Chemistry, vol. 1 (Ed.: G. L. Larson), Jai Press Inc.,
Greenwich, UK, 1991, pp. 65–187.
[6] a) G. te Velde, F. M. Bickelhaupt, E. J. Baerends, C. Fon-
seca Guerra, S. J. A. van Gisbergen, J. G. Snijders, T. Ziegler, J.
Comput. Chem. 2001, 22, 931–967; b) http://www.scm.com.
[7] a) S. H. Vosko, L. Wilk, M. Nusair, Can. J. Phys. 1980, 58,
1200–1211; b) A. D. Becke, Phys. Rev. A 1988, 38, 3098–3100;
c) J. P. Perdew, Phys. Rev. B 1986, 33, 8822–8824 (Erratum:
Phys. Rev. B 1986, 34, 7406.
[8] a) E. van Lenthe, E. J. Baerends, J. G. Snijders, J. Chem. Phys.
1994, 101, 9783–9792; b) E. van Lenthe, R. van Leeuwen, E. J.
Baerends, J. G. Snijders, Int. J. Quantum Chem. 1996, 57, 281–
293.
Received: March 26, 2012
Published Online: May 23, 2012
3228
www.eurjic.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2012, 3216–3228