3Ge+ a free germyl cation with aryl ligands

Article abstract of DOI:10.1039/b818931e

The reaction of Ar₃GeBr (Ar = 2,6-(OtBu)₂C ₆H₃), which is the side product of the synthesis of the metalloid germanium cluster compound Ge₈Ar₆, with the silver salt of the weakly coordinating anion (WCA) [Al(OR_f) ₄]^- (R_f = C(CF₃)₃) gives the free germyl cation Ar₃Ge⁺. Quantum chemical calculations open an insight into the bonding situation of this first free cation exhibiting aryl ligands and a first reaction leading to Ar ₃GeOH is presented.

Full text of DOI:10.1039/b818931e

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/dalton | Dalton Transactions
[(OtBu)2C6H3]3Ge+ a free germyl cation with aryl ligands†‡
Christian Schenk,a Christian Drostb and Andreas Schnepf*a
Received 27th October 2008, Accepted 27th November 2008
First published as an Advance Article on the web 15th December 2008
DOI: 10.1039/b818931e
The reaction of Ar3GeBr (Ar = 2,6-(OtBu)2C6H3), which is
the side product of the synthesis of the metalloid germanium
cluster compound Ge8Ar6, with the silver salt of the weakly
coordinating anion (WCA) [Al(ORf)4]- (Rf = C(CF3)3) gives
the free germyl cation Ar3Ge+. Quantum chemical calcula-
tions open an insight into the bonding situation of this first
free cation exhibiting aryl ligands and a first reaction leading
to Ar3GeOH is presented.
We now report on another free cation of germanium that we
were able to synthesise using the bulky aryl ligand Ar = 2,6-
(OtBu)2C6H3. For the synthesis of a free germyl cation, a sufficient
precursor is needed that can be transformed into the correspond-
ing cationic species, and additionally, a weak coordinating anion
is needed so that no anion–cation interaction disturbs the free
cation character.6 For the synthesis of tri-coordinated cationic
heavier group 14 centres R3M (M = Si, Ge, Sn; R = organic
substituent), hydride abstraction of the corresponding hydride
R3MH with the triphenylmethyl cation (trityl cation) is a common
route.7 However, this synthetic route fails when hydrides R3MH
with bulky ligands are used since the hydride atom is sterically
protected from an attack of the trityl cation. Nevertheless, bulky
ligands are necessary to get access to a “free” cation, where no
solvent molecules are coordinated to the cationic centre. Thus
for the synthesis of cations exhibiting bulky ligands, possessing
the possibility to be “free” in solution and the solid state, other
synthetic routes have to be used. One possibility to do so is a
one electron oxidation of a radical, the way used from Sekiguchi
et al. for the synthesis of the free cation [Ge(SitBu2Me)3]+. Another
possibility can be seen in the halide abstraction reaction from an
R3GeX precursor (X = halide atom) (Scheme 2).
Free germyl cations as well as germanium radicals are a subject
of current interest as they can be seen as fundamental highly
reactive species in germanium chemistry.1 Especially in the case
when both species are known as isolable and characterisable
compounds the influence of the occupation of the 4pz orbital
can be directly observed. The first free cationic species that
exhibits no cation–anion interaction and no coordination to
a solvent molecule in the solid state was the cyclopropylium
cation [Ge3(SitBu3)3]+, which was synthesised by Sekiguchi et al.
from the corresponding cyclotrigermene Ge3(SitBu)4.2 In this
cation, the positive charge is delocalised over three germanium
centres, thus the cation is stabilised by conjugation to the Ge–
Ge p-bond, comparable to the situation in the cyclopropenylium
cation C3H3 .3 Thus, [Ge3(SitBu3)3]+ can be seen as a cationic
+
aromatic 2p electron system. The first free germyl cation that lacks
conjugation to p bonds, (tBu2MeSi)3Ge+ was isolated some years
later4 from the same group by a one electron oxidation of the
corresponding free radical (tBu2MeSi)3Ge∑.5 To the best of our
knowledge, this is the only system were both tri-coordinated
species; the radical and the free cation are known as isolable
species. In the case of (tBu2MeSi)3Ge+ as well as (tBu2MeSi)3Ge∑,
the highly reactive germanium atom is shielded by the bulky ligand
SitBu2Me. Both compounds can be converted into each other by
a one electron oxidation/reduction using Ph3C+ and tBu- as the
oxidizing/reducing agent, respectively (Scheme 1).
Scheme 2 Halide abstraction reaction for the synthesis of a germyl cation.
An ideal starting material for such a reaction seems to be the
compound Ar3GeBr 1 (Ar = 2,6-(OtBu)2C6H3), which can be
isolated as the oxidation product in the synthesis of the metalloid
cluster compound Ge8Ar6 from GeBr.8 In 1 the Ge–Br distance is
at 238 pm slightly elongated with respect to other known Ge–Br
distances (e.g. 232 pm in Ph3GeBr9) and therefore the elimination
of Br- should perform much easier as in other R3GeBr compounds.
Additionally, the three bulky ligands might be able to shield the
cationic centre against the exterior, so that no solvent molecules
can coordinate to the germanium atom. For the Br- abstraction
we decided to use the silver salt of the weakly coordinating anion
(WCA) [Al(ORf)4]- (Rf = C(CF3)3), which was used recently for
Scheme 1 Synthesis of the radical/germyl cation pair (R = SitBu2Me).
aInstitut fu¨r Anorganische Chemie der Universita¨t Karlsruhe (TH), En-
gesserstr., Geb. 30.45, D-76131 Karlsruhe, Germany. E-mail: schnepf@
chemie.uni-karlsruhe.de; Fax: +49 (721) 608-4854; Tel: +49 (721) 608-
2951
+
the synthesis of the carbocation CI3 from CI4 using a similar
reaction course.10
Both compounds (Ag[Al(ORf)4] (Rf = C(CF3)3) and 2,6-
(OtBu)2C6H3)3GeBr 1) were dissolved in CH2Cl2 at -30 ◦C leading
to a pale greenish solution in which a grey-white solid (AgBr)
is suspended. The solution was filtered at -30 ◦C and upon
concentrating the reaction mixture, pale green crystals can be
isolated. X-Ray crystal structure analysis of these crystals reveals
that the desired cation Ar3Ge+ 2 (Ar = 2,6-(OtBu)2C6H3) had
bInstitut fu¨r Anorganische Chemie der Universita¨t Leipzig, Johannisallee 29,
D-04103 Leipzig, Germany
† CCDC reference numbers 698439 (2) and 698440 (4). For crystallo-
graphic data in CIF or other electronic format see DOI: 10.1039/b818931e
‡ We thank the DFG for financial support and Prof. H. Schno¨ckel
for helpful discussions. We also gratefully thank Prof. I Krossing, who
provided the silver salt of the weak coordinating anion.
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 773–776 | 773
©

View Article Online
formed, whose molecular structure is shown in Fig. 1. As the
shortest C–F contact between the anion and the cation is 340 pm,
an anion–cation interaction is not present, so 2 is a free germyl
cation and it is the first one where aryl ligands are bound to the
central germanium atom.
expected due to the covalent radii.12,8 This behaviour is different to
that found for the other free germyl cation (tBu2MeSi)3Ge+, where
the Ge–Si bond length is elongated with respect to a normal Ge–
Si single bond. In the case of the free cation 2, the shortening of
the Ge–C bond hints to a certain bonding interaction between
the empty 4pz orbital at the germanium centre and the p electron
density at the aromatic ring system as emphasised by the Lewis
forms in Scheme 3.
Scheme 3 Lewis forms of the cation R3Ge+ 2.
To verify this assumption quantum chemical calculations on
model compounds R3Ge+ (R = Ph (2a); = SiMe3 (2b); Table 1)
have been performed.13 For 2a, a similar arrangement of the phenyl
rings around the germanium centre is calculated and also a short
Ge–C bond of 191 pm was calculated. The multiple bond character
of the Ge–C bond could be verified by an Ahlrichs-Heinzmann
population analysis, for which a shared electron number (SEN14)
of 1.37 is calculated for the Ge–C bond. In the model compound
2b, a SEN of only 1.08 is calculated for the Ge–Si bond. These
results show that a Ge–C multiple bond character is present, likely
due to back bonding between the p system of the aromatic ring
and the empty 4pz orbital of the germanium atom. Hereby, the
ideal value of the dihedral angle of 0◦ can not be realised due
to steric reasons of the three aryl ligands. Such a hindrance is
also present in the model compound 2a, where a dihedral angle
of 26.5◦ is calculated. Thus, the electron deficiency at the cationic
centre is reduced by an interaction with the p-electrons of the
organic rings. Consequently, the calculated positive charge of the
germanium atom in the model compound 2a is +0.6 and not +1.
The p bonding interaction additionally leads to a hindered
rotation of the aryl ligands around the Ge–C bond as the rotation
would lead to a complete loss of the back bonding ability in
the transition state where the dihedral angle is 90◦. Thus, the
interconversion of both enantiomeric forms of 2 is hindered by
electronic reasons. To get a first guess about the activation barrier
for this process we calculated the intermedial structure of 2a,
where all aromatic ring planes are perpendicular to the GeC3
plane (D3h symmetry). In this case, the calculated energy of the
molecule is 105 kJ mol-1 higher than that of the ground state. In
this orientation, no back bonding between the empty 4pz orbital
and the p orbitals of the aromatic system is possible as they are
oriented perpendicular to each other (dihedral angle 90◦) and
therefore the calculated positive charge on the germanium atoms
increases from +0.6 to +0.9. The magnitude of the barrier between
+
Fig. 1 Molecular structure of Ge[(OtBu)2C6H3]3 2, without hydrogen
atoms. Vibrational ellipsoids with 50% probability. Selected distances
(pm) and angles (◦): Ge1–C1 189.8(3); Ge1–C10 189.3(2); Ge1–C20
189.9(2); Ge1–O5 286.5(3); Ge1–O26 287.7(2); C1–Ge1–C10 120.99(10);
C1–Ge1–C20 121.20(10); C10–Ge1–C20 117.80(10).
The three aryl ligands in 2 are oriented in a paddlewheel fashion
with a dihedral angle between the GeC3 plane and the aromatic
ring plane of 56◦ (see Fig. 1). The sum of the C–Ge–C angles is
359.99◦, thus a planar arrangement around the central cationic
germanium centre is present. This arrangement is different to that
found in the recently synthesised cationic compound of silicon
[Me2SiAr*]+ (Ar* = 2,6-(2,3,5,6-Me4-C6H)2C6H3),11 where the
geometry around the central silicon atom is pyramidal; the sum
of the covalent bond angles is 346◦ and the silicon atom is around
40 pm out of the plane of the three covalently bound carbon atoms.
This difference is due to an interaction of the silicon atom with
the carbon atoms of the flanking aryl substituent. In the case of 2,
any possible interaction between the central germanium and the
tert-butoxy groups located in the 2,6-positions of each aryl ligand
can be excluded, as the Ge–O distances are long at 286–288 pm,
and significant interaction would lead to a distortion of the planar
arrangement.
The Ge–C distances in 2 are at 189.7 pm, thus shorter than the
normal value of a Ge–C single bond, where a distance of 199 pm is
Table 1 Comparison of bonding features of the calculated model compounds (Ar = 2,6-OtBu2C6H3)
Compound
Ge–E Distance/pm
Ge–E 2 center SEN
Torsion angle of ligand/◦
Charge Ge
Sum of Ge–E angles/◦
+
GePh3 (D3h)
193.8
191.2
245.7
192.4
1.16
1.37
1.08
1.32
90
26.3
—
0.98
0.6
0.48
0.69
360
360
359.3
360.8
+
GePh3
+
Ge(SiMe3)3
+
GeAr3
51.6
774 | Dalton Trans., 2009, 773–776
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
the two enantiomers of 2 is at 105 kJ mol-1 in line with comparable
systems as B(2,6-Me2-C6H3)2(2,4,6-Me3-3-iPr-C6H), for which an
activation barrier of 74.47 kJ mol-1 was measured via dynamic
NMR studies.15 As it is to be expected from the reaction pathway,
2 is present as a racemic mixture in solution and it crystallises as a
racemic mixture in the space group P21/n, with both enantiomers
present in the crystal.
germanol 4. To prove that such a reaction course had taken place,
an experiment where the reaction of KOtBu and 2 was performed
in a sealed glass tube, in which we were able to detect the NMR
signals of the elimination product iso-butene (Scheme 4).
The development of the subsequent reaction might be due to
steric reasons as the steric amount of the OH group is less than
that of the OtBu group.17 In 4, the Ge–C bond lengths are now at
196 pm and in the range of a normal Ge–C single bond and the
Ge–O distance is 179 pm, also in the range of known Ge–O bond
lengths in R3GeOH compounds.18
We have prepared the first crystalline and highly reactive tri-
arylgermyl cation/weak coordinating anion compound 2 without
neither anion/cation, nor inner cationic donor atom contacts
in the solid state. This was shown by X-ray analysis; additional
quantum chemical calculations were performed in order to
verify its bonding situation. Its highly electrophilic character was
established in a first reaction with KOtBu and will be further
investigated in future experiments.
1
Variable temperature H-NMR experiments revealed that 2 is
only stable in solution at low temperatures as signals of 2 vanish
when heated above -20 ◦C, leading to an undefined spectra of
numerous products.16 Due to the lack of stabilizing anion–cation
interactions, 2 appears to be a highly electrophilic compound being
non stable in solution at elevated temperatures. However, when
2 is treated at low temperatures with nucleophiles e.g. KOtBu,
as emphasised in Scheme 4, we were able to isolate a product
in the form of colourless crystals after work-up procedures. X-
Ray crystal structure analysis of these crystals revealed that not
the expected ether Ar3GeOtBu 3 has formed but the germanol
Ar3GeOH 4 (Fig. 2).
General considerations
All manipulations were carried out under nitrogen or in vacuo
in Schlenk-type glassware on a dual manifold Schlenk line.
Solvents were pre-dried over molecular sieves. GeBrAr3 (Ar =
2,6-OtBu2C6H3) was synthesised by the reaction of GeBr and LiR
as described in the literature.8
Preparation of [2,6-OtBu2C6H3]3Ge+ 2
82 mg of [2,6-OtBu2C6H3]3GeBr (0.1 mmol) and 107 mg of
AgAl[OC(CF3)3]4 were added to a flask in a glove box. The solid
mixture was then cooled to -78 ◦C and -78 ◦C cold CH2Cl2 was
added via a steel cannula leading to a pale yellow greenish solution.
+
Scheme 4 Subsequent reaction of GeAr3 2 with KOtBu.
◦
The solution was then warmed to -30 C and stirred for 2 d at
this temperature leading to a pale yellow solution in which a grey-
white solid of AgBr was suspended. Afterwards, the solution was
filtered and concentrated in vacuo. At -30 ◦C, pale greenish crystals
of {[2,6-OtBu2C6H3]3Ge}{Al[OC(CF3)3]4} 2 are formed (78 mg,
45%). 1H-NMR (400 MHz, CD2Cl2) d/ppm: 1.24 (d, CH3, 18H),
6.82 (dd, Ar, 2H), 7.33 (t, Ar, 1H). 13C-NMR (100 MHz) d/ppm:
28.8 (d, CH3), 81.8 (d, CCH3), 110.8 (d, Ar) 116.7 (s, Ge–Ar), 121.6
(q, CF3), 133.0 (s, Ar), 159.8 (d, O–Ar). 19F-NMR (376 MHz)
d/ppm: -75.8 (s, CF3). 27Al-NMR (104 MHz) d/ppm: 36.0 (s).
Preparation of [2,6-OtBu2C6H3]3GeOH 4
138 mg of [2,6-OtBu2C6H3]3GeBr (0.17 mmol) and 182 mg of
AgAl[OC(CF3)3]4 were added to a flask in a glove box. The solid
mixture was then cooled to -78 ◦C and -78 ◦C cold CH2Cl2 was
added via a steel cannula leading to ◦a pale yellow greenish solution.
The solution was warmed to -30 C and stirred for 2 d leading
to a pale yellow solution in which a grey-white solid of AgBr was
suspended. To◦this suspension, 20 mg KOtBu dissolved in THF
cooled to -40 C was added slowly while the suspension turned
brown immediately. The solvent was removed and the brown
residue was extracted with toluene. Colourless crystals (52 mg,
0.07 mmol, 40.6%) of [2,6-OtBu2C6H3]3GeOH 4 were obtained
Fig. 2 Molecular structure of 4 without hydrogen atoms. Vibra-
tional ellipsoids with 50% probability. Selected bond lengths (pm) and
angles (◦): Ge1–C1 195.1(3); Ge1–C10 196.6(3); Ge1–C20 197.1(3);
Ge1–O100 179.0(2); C1–Ge1–C10 113.81(13); C1–Ge1–C20 115.77(14);
C10–Ge1–C20 114.37(13).
1
Thus, a subsequent elimination reaction of the primary product
upon concentrating the toluene solution. H-NMR (250 MHz,
3 as shown in Scheme 4 has taken place, leading to the isolated
C6D6) d/ppm: 1.25 (s, tBu, 9H), 1.33 (s, tBu, 9H), 6.67 (m, Ar,
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 773–776 | 775
©

View Article Online
2H), 7.08 (t, Ar, 1H). 13C-NMR (100 MHz) d/ppm: 29.7 (tBu),
68.6 (tBu), 110.8 (Ar), 119.9 (Ar), 127.3 (Ar), 161.9 (Ar).
the molecule are present within the crystal with an occupancy
of 95 : 5%, thus the disorder could only be modelled for the
central Ge–O group as the 5% disorder in the O-tBu group was
not possible to model.
X-Ray crystallography†
Table 2 contains the crystal data and details of the X-
+
Notes and references
ray structural determination for {[2,6-OtBu2C6H3]3Ge}
2
-
{Al[OC(CF3)3]4} ·1.875CH2Cl2 and [2,6-OtBu2C6H3]3GeOH 4.
The data were collected at 100 K (2) and 150 K (4) using
a Bruker IPDS II diffractometer employing monochromated
1 V. Ya. Lee and A. Sekiguchi, Acc. Chem. Res., 2007, 40, 410.
2 A. Sekiguchi, M. Tsukamoto and M. Ichinohe, Science, 1997, 275, 60.
3 C. H. Suresh and N. Koga, Internet Electro., J. Mol. Design, 2002, 1,
603.
4 A. Sekiguchi, T. Fukawa, V. Ya. Lee, M. Nakamoto and M. Ichinohe,
Angew. Chem., 2003, 115, 1175, (Angew. Chem., Int. Ed., 2003, 42,
1143).
5 A. Sekiguchi, T. Fukawa, M. Nakamoto, V. Ya. Lee and M. Ichinohe,
J. Am. Chem. Soc., 2002, 124, 9865.
6 I. Krossing, Chem.–Eur. J., 2001, 7, 490.
7 J. Y. Corey, J. Am. Chem. Soc., 1975, 97, 3237.
8 A. Schnepf and C. Drost, Dalton Trans., 2005, 20, 3277.
9 H. Preut and F. Huber, Acta Crystallogr., Sect. B: Struct. Crystallogr.
Cryst. Chem., 1979, 35, 83.
10 I. Krossing, A. Bihlmeier, I. Raabe and N. Trapp, Angew. Chem., 2003,
115, 1569, (Angew. Chem., Int. Ed., 2003, 42, 1531).
11 S. Duttwyler, Q.-Q. Do, A. Linden, K. K. Baldridge and J. S. Siegel,
Angew. Chem., 2008, 120, 1743, (Angew. Chem., Int. Ed., 2008, 47,
1719).
˚
MoKa (0.71073 A) radiation from a sealed tube and equipped
with an Oxford Cryosystems cryostat. A numeric absorption
correction was applied using the optically determined shape of
the crystals. The structure was solved by direct methods and
refined by full-matrix least-square techniques (Programs used:
SHELXS and SHELXL19). The non-hydrogen atoms were refined
anisotropically and the hydrogen atoms were calculated using
a riding model, except the hydrogen atom of the OH group
in 4, which was found in the difference Fourier map. In both
compounds, a disorder appears, which was assigned using a split
model. In 2, one C(CF3)3 group is disordered at two positions
with an occupancy of 55:45%. In 4, two different orientations of
12 H. Wiberg, Lehrbuch der Anorganischen Chemie, Walter de Gruyter,
Berlin, Germany, 102nd edn, 2007, p. 138.
13 Quantum-chemical calculations were carried out with the RI-DFT
version of the Turbomole program package, by employing the Becke-
Perdew 86-functional. The basis sets were of SVP quality. The electronic
structure was analysed with the Ahlrichs-Heinzmann population
analysis based on occupation numbers. Turbomole: O. Treutler and
R. Ahlrichs, J. Chem. Phys., 1995, 102, 346–354. BP-86-functional:
J. P. Perdew, Phys. Rev. B: Condens. Matter Mater. Phys., 1986, 33,
8822–8824; A. D. Becke, Phys. Rev. A, 1988, 38, 3098–3100. RI-DFT:
Table 2 Crystal data and detail of structural determinations
+
{[2,6-OtBu2C6H3]3Ge} 2
[2,6-OtBu2C6H3]3-
-
Compound
{Al[OC(CF3)3]4} ·1.875CH2Cl2 GeOH 4
Formula
GeCl3.75AlF36O10C59.88H66.75
1862.89
100
Monoclinic
P21/n
18.1010(9)
24.3749(8)
18.2722(8)
90
106.088(4)
90
7746.2(6)
4
0.674
GeO7C42H64
753.52
150
Triclinic
¯
P1
11.869(2)
12.774(3)
13.757(3)
90.70(3)
95.43(3)
90.95(3)
2075.9(7)
2
FW/g mol-1
T/K
¨
K. Eichkorn, O. Treutler, H. Ohm, M. Ha¨ser and R. Ahlrichs, Chem.
Crystal system
Space group
Phys. Lett., 1995, 240, 283–290. SVP: A. Scha¨fer, H. Horn and R.
Ahlrichs, J. Chem. Phys., 1992, 97, 2571–2577. Ahlrichs-Heinzmann
population analysis: E. R. Davidson, J. Chem. Phys., 1967, 46, 3320–
3324; K. R. Roby, Mol. Phys., 1974, 27, 81–104; R. Heinzmann and
R. Ahlrichs, Theor. Chim. Acta, 1976, 42, 33–45; C. Erhardt and R.
Ahlrichs, Theor. Chim. Acta, 1985, 68, 231–245.
˚
a/A
˚
b/A
˚
c/A
a/◦
b/◦
g /◦
14 The shared electron numbers (SENs) for bonds are a reliable measure
of the covalent bonding strength. For example, the SEN for the Ge–Ge
single bond in the model compound R3Ge–GeR3 (R = NH2) is 1.04.
15 J. P. Hummel, D. Gust and K. Mislow, J. Am. Chem. Soc., 1974, 96,
3679.
16 NMR experiments show a higher yield of 2 than the isolated 45%.
17 DFT calculations also show that the elimination of iso-butene is
exothermic by 40 kJ mol-1.
18 K. Goto, I. Shimo and T. Kawashima, Bull. Chem. Soc. Jpn., 2003, 76,
2389; A. Kawachi, Y. Tanaka and K. Tamao, Organometallics, 1997,
16, 5102.
19 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2008,
64, 112.
3
˚
V/A
Z
m/mm-1
0.783
1.205
D/g cm-3
1.597
Reflections measured 54 448
Reflections observed 12 332
14 158
7372
0.0667
0.948
Rint
0.0641
1.046
GOF
R1 (I < 2s)
wR2 (all data)
0.0438
0.1138
0.0475
0.1179
776 | Dalton Trans., 2009, 773–776
This journal is
The Royal Society of Chemistry 2009
©