Lena Albers1, Mohammad Aghazadeh Meshgi2, Judith Baumgartner3, Christoph Marschner2, Thomas Müller1. 1. Institut für Chemie, Carl von Ossietzky Universität Oldenburg , Carl von Ossietzky-Strasse 9-11, D-26129 Oldenburg, Federal Republic of Germany. 2. Institut für Anorganische Chemie, Technische Universität Graz , Stremayrgasse 9, 8010 Graz, Austria. 3. Institut für Chemie, Karl Franzens Universität Graz , Stremayrgasse 9, 8010 Graz, Austria.
Abstract
The rearrangement of tris(trimethylsilyl)silyltrimethylgermane 1 to give tetrakis(trimethylsilyl)germane 2 was investigated as a typical example for Lewis acid catalyzed Wagner-Meerwein-type rearrangements of polysilanes and polygermasilanes. Direct 29Si NMR spectroscopic evidence is provided for several cationic intermediates during the reaction. The identity of these species was verified by independent synthesis and NMR characterization, and their transformation was followed by NMR spectroscopy.
The rearrangement of tris(trimethylsilyl)silyltrimethylgermane 1 to give tetrakis(trimethylsilyl)germane 2 was investigated as a typical example for Lewis acid catalyzed Wagner-Meerwein-type rearrangements of polysilanes and polygermasilanes. Direct 29Si NMR spectroscopic evidence is provided for several cationic intermediates during the reaction. The identity of these species was verified by independent synthesis and NMR characterization, and their transformation was followed by NMR spectroscopy.
Despite the most interesting
properties and the widespread potential
use of oligo- and polysilanes, synthetic access to this class of compounds
is mainly limited to the Wurtz-type coupling reaction, in which the
Si – Si chain is built up in a reductive process using silicon
halides and alkali metals.[1] Up to date
alternative methods for the synthesis of oligosilanes such as dehydrocoupling
of hydrosilanes[2] or electrochemically mediated
Si–Si bond forming processes[3] do
not reach the prominence of the Wurtz-type coupling. While this process
works fairly well for the preparation of dialkylated linear polysilane
polymers and small cyclosilanes, it does not provide satisfactory
access to compounds of even slightly enhanced structural complexity.
In particular, the generation of well-defined polycyclic molecules
in acceptable yields has proven to be very challenging by this method.[1] Regarding the potential for such materials being
useful for technical applications, there is intrinsic interest in
novel approaches that open the possibility to develop new synthetic
tools for the preparation of open chain, cyclic, and polyhedral compounds
with silicon containing backbones. Pioneering work by the group of
Kumada and Ishikawa,[4] further developed
by the groups of West,[5] Märkl,[6a] and Pannell,[6b] revealed
that the Lewis acid catalyzed Wagner–Meerwein-type rearrangement
of oligosilanes is a powerful method to create structures of higher
complexity. The high yield synthesis of an all-silicon analogue of
adamantane from a structural isomer, reported previously by two of
us, is certainly one of the more spectacular examples for the potential
of this synthetic methodology.[7]The
description of this process as being a silicon variation of
the Wagner–Meerwein rearrangement suggests the presence of
cationic intermediates.[4,5] However, despite the fact that
the chemistry of silyl cations has experienced immense popularity
in recent years,[8] experimental evidence
for polysilanyl-substituted silyl cations and their possible involvement
in rearrangement or fragmentation reactions is indeed rather scarce.
Lambert and co-workers reported on the formation of tris(trimethylsilyl)silylium,
(Me3Si)3Si+, most probably in the
form of its arene complexes; the obtained 29Si NMR spectra
suggest, however, substantial decomposition of the material.[9] Structural aspects and the question of aromaticity
and homoaromaticity had been the focus in several publications of
the Sekiguchi group concerning some cyclic polysilasilyl cations.[10,11] Sekiguchi and co-workers also demonstrated the occurrence of a degenerated
intramolecular 1,3-methyl shift occurring in a dialkyloligosilanylsilyl
cation.[12] Interestingly, in none of these
reports were parallels drawn between the subject of study and the
likely involvement of similar species in the sila-Wagner–Meerwein
rearrangement. During our own experimental studies of the Wagner–Meerwein
rearrangement in oligosilanes,[13] we found
that for germylsilanes a particular preference exists for the formation
of products in which the germanium atoms occupy central positions
in the oligosilane frameworks.[14] A deeper
insight was provided by a computational study of the fundamental reaction
of germylsilane 1 to give germane 2 (reaction
1, Scheme ). The results
of density functional theory (DFT) calculations suggested an exothermic
multistep rearrangement via several isomeric silyl- and germyl cations 3–8 (Scheme ). According to this investigation, the isomerization
occurs via consecutive 1,2-silyl and 1,2-methyl shifts with an overall
barrier of less than 50 kJ mol–1.[14]
Scheme 1
Rearrangement of Germylsilane 1 to Silylgermane 2 Using Lewis Acids
Scheme 2
Suggested Isomerization Cascade of Germylium Ion 3 to
Silylium Ion 8(14)
An experimental investigation
on the course of the principal reaction
1 would greatly benefit from the simplicity of the reactants, proposed
intermediates, and products and from the straightforward interpretation
of their NMR spectra. The close chemical relationship between silicon
and germanium suggests that conclusions drawn from a mechanistic study
of reaction 1 will be of significance also in the case of polysilanes.
Here, we report on a 29Si NMR investigation of the exemplary
reaction 1, which provides direct evidence for cationic intermediates
in the sila-Wagner–Meerwein rearrangement and for the clean
transformation of the incipient germyl cation 3 into
the product-forming silyl cation 8 at low temperatures
along a cascade of isomeric cations 4–7 (Scheme ).
Results
and Discussion
As a starting point for our experimental investigation,
we tested
additional cationic Lewis acids as stoichiometric reagents or catalysts
for the rearrangement of model compound 1. The idea behind
using a cationic Lewis acid is to introduce weakly coordinating anions
such as the tetrakis(pentafluorophenyl)borate, [B(C6F5)4]−,[15] which will be beneficial for the stabilization of the postulated
cationic intermediates. Indeed, we found that already 4 mol % of tritylborate [Ph3C][B(C6F5)4], 9, induces the complete transformation of germylsilane 1 into the symmetricgermane 2 in dichloromethane
at room temperature (Scheme ).[16] For the generation of stoichiometric
amounts of the postulated cationic intermediates in the rearrangement
reaction shown in Scheme , we switched to the even more electrophilic trialkylsilylarenium borates, [R3Si(arene)] [B(C6F5)4] (R = Me, Et, Pr; arene
= C6H5CH3, C6H6, C6H5Cl), 10,[8] and used arenes as solvents. In the past, it was shown[8] that the combination of arene solvents and weakly
coordinating anions is suitable for the detection of highly reactive
silylium ions. In order to allow the NMR spectroscopic detection of
the postulated cationic intermediates, we tested at first the stoichiometric
reaction between germylsilane 1 and triethylsilyl benzenium
borate, [Et3Si(C6D6))] [B(C6F5)4], 10a. The reaction in benzene
at 6 °C was completed after 2 h and resulted in a biphasic reaction
mixture that is typical for solutions of tetrakis(pentafluorophenyl)borates
in aromatic hydrocarbons.[8]29Si NMR spectra obtained from the ionic phase showed many signals,
which indicated the formation of a complicated reaction mixture. The 29Si NMR analysis of the second, nonpolar layer showed the
presence of at least nine different neutral compounds. These nine
compounds were identified by GC/MS spectrometry and NMR spectroscopy
as the expected germane 2, three different methylethylsilanes,
and a series of silylated germanes 11 and 12 (Scheme and Supporting Information for details). Noteworthy,
germanes 11 and 12, which all have cation 8 as a common intermediate, are substituted with a different
number of ethyl groups. When the same reaction was performed with
tri-iso-propylsilyl toluenium borate, [Pr3Si(C7D8)][B(C6F5)4], 10b, only two side
products were detected, namely, (Me3Si)3GeSiMe2Pr and Pr3SiMe. Finally, when trimethylsilyl toluenium
borate [Me3Si(C7D8)] [B(C6F5)4], 10c, was applied as reaction
partner, only the expected rearrangement product 2 and
tetramethylsilane were identified in the nonpolar phase (see Supporting Information). Although with all three
silylarenium borates 10 the clean production of cationic
species at room temperature was not achieved, the results of these
stoichiometric reactions indicate that the initial step of the rearrangement
reaction is cation formation by cleavage of either a Ge–C-
or Si–C bond. Judged from the calculated bond strengths, the
cleavage of a Ge–C bond is preferred over that of a Si–C
bond (De(Ge–C) = 305 kJ mol–1;
(De(Si–C) = 347 kJ mol–1).[14] This clearly supports our initial assumption
that the first cationic intermediate formed should be germyl cation 3 which undergoes a rearrangement reaction to give silyl cation 8. Furthermore, the results suggest that the terminating step
consists in back transfer of an alkyl group to the rearranged cation
which yields besides germane 2 the alkylated products 13 (Scheme ). The multiple substitution[5] in the case
of the ethyl substituted silyl arenium ion demonstrates the ease of
alkyl group exchange under the applied conditions, which is only limited
by the steric requirements of the exchanging groups.
Scheme 3
Stoichiometric
Reaction between Triethylsilyl Benzenium Borate 10a and
Silylgermane 1 in Benzene at T = 6°C
Scheme 4
Reaction of Silylgermane 1 with Different Triakylsilyl
Arenium Borates 10 at r.t. (Borate Anion Not Shown)
In contrast to the reaction
at room temperature, the reaction of
silylgermane 1 with tri-iso-propylsilyl
toluenium borate, 10b, in toluene-d8 at −20 °C yielded cleanly a single ionic compound
as indicated by only two resonances that were detected in the 29Si NMR spectra of the ionic phase. The NMR data (δ29Si = −87.9 (Siq) and δ29Si = −7.9 (SiMe3), integral ratio in the 29Si{1H} inverse gated experiment, 1:3) is consistent with
the structure of germyl toluenium ion 3(C7D8). This assignment was supported by the fact that the
same cationic species was formed by the conventional Bartlett Condon
Schneider hydride transfer[17] reaction of
hydridogermane 14 with trityl borate 9 (Scheme ). In addition, the
results of quantum mechanical calculations[18,19] of 29Si NMR chemical shifts for an optimized molecular
structure of 3(C7H8) predict values
that are very close to the experiment (see Table ).
Scheme 5
Formation of Silyl Toluenium Ion 8(C7D8) from Different Precursor Compounds
Table 1
Experimental and
Calculated 29Si NMR Chemical Shifts of Solvent Complexes
of Germylium Ions 3 and
7 and Silylium Ion 8a
compound
δ29Si (Siq)
δ29Si (SiMe3)
δ29Si (Si+Me2)
δ29Si (SiMe2)
[3(C7D8)]
–87.9 (−93)
–7.9 (−7)
7(C7D8)]
(8; −13)
(−13)
8(C7D8)]
–2.4 (0)
98.1 (108)
[8(ClC6D5)]b
–1.5 (0)
154.3 (157)
Calculated values for the nondeuterated
compounds at GIAO/M06-L/6-311G(2d,p)//M06-2X/6-311+G(d,p) are in parentheses.
Calculated for the chloronium
ion
structure [8–Cl-C6H5]. Chloronium
ions are the dominating species when silyl cations are generated in
chlorinated arenes; see ref (22).
Calculated values for the nondeuterated
compounds at GIAO/M06-L/6-311G(2d,p)//M06-2X/6-311+G(d,p) are in parentheses.Calculated for the chloronium
ion
structure [8–Cl-C6H5]. Chloronium
ions are the dominating species when silyl cations are generated in
chlorinated arenes; see ref (22).Interestingly,
already after 2 h at −20 °C two new 29Si NMR
resonances were detected in both cases (δ29Si = −2.4
and +98.1, integral ratio in the 29Si{1H} inverse
gated experiment, 3:1; see Figure and Supporting
Information).[20] Over time, the intensity
of the new signals increased significantly while the original 29Si NMR signals of germyl cation 3(C7D8) vanished (see Figure ), thereby confirming the clean conversion of silylgermyltoluenium ion 3(C7D8) into a new
cationic species in toluene at −20 °C. The new high field
signal appears in the typical region for trimethylsilyl groups attached
to germanium,[21] but the measured chemical
shift is not specific. More informative is the low field resonance
at δ29Si = 98.1, which is characteristic for silylarenium ions.[8] The 29Si NMR
chemical shifts of these cationic species are reported to be very
sensitive to changes of the solvent due to the replacement of the
arene molecule attached to the silicon atom. Indeed, the 29Si NMR chemical shift of the low field resonance varies significantly
with the arene solvent from δ29Si = 98.1 in toluene-d8 to δ29Si = 154.3 in chlorobenzene-d5 (see Table ).[22] On the basis of these
results, the two new 29Si NMR signals were tentatively
assigned to silyl toluenium ion 8(C7D8). This assignment was supported by its independent synthesis
by hydride transfer reaction from the appropriate silane 15 (see Scheme and Figure ) and by quantum
mechanical calculations of the 29Si NMR chemical shifts
of optimized molecular structures of arene complexes of 8 (see Table ). According
to the NMR results, solutions of the silyltoluenium borate[8(C7D8)][B(C6F5)4] in toluene at T = −20 °C
are stable for at least 3 days. When the temperature of the NMR probe
was gradually raised in steps of 10 °C, already at T = −10 °C new signals in the 29Si NMR spectrum
appeared, and exposure of the NMR sample to temperatures of about
+20 °C resulted in complete decomposition of cation 8(C7D8) (see Supporting
Information, Figure S21).[23] The
thermal instability of arenium borate [8(C7D8)][B(C6F5)4] in solution
prevents its isolation in substance and additional characterization.
These NMR experiments demonstrated the clean conversion of silylgermyltoluenium ion 3(C7D8) to germylsilyl
toluenium 8(C7D8) in toluene at
−20 °C. According to our previous computational study,
this sila-variant of the Wagner–Meerwein rearrangement is expected
to proceed via several cationic intermediates (see Scheme ).[14] Additional DFT computations at the B3LYP/6-311+G(d,p) level of theory[18,19] indicated that complexation of the formed silyl and germyl cations
with the arene solvent has no significant influence on the calculated
reaction coordinate. The calculated relative energies of the toluene
complexes of cations 3–8 reveal only
slight modifications of the previously predicted stability sequence
for the free silylium and germylium ions (see Figure S22, Supporting Information). In particular, the silyltoluenium
ion 8(C7H8) is predicted to be
the most stable along the complete reaction sequence, and the overall
reaction 3(C7H8) → 8(C7H8) is found to be exothermic by
44 kJ mol–1. Experimentally, however, no indications
for the cationic intermediates 4–7 or for the related arenium ions were detected. In order to test
our mechanistic proposal, we decided to generate one specific cation
along the reaction cascade, shown in Scheme , and to monitor its rearrangement. Therefore,
we subjected silylgermane 16 to the hydride transfer
reaction in toluene in an attempt to generate the corresponding germyltoluenium 7(C7D8) (Scheme ). This last step of the reaction sequence, 7(C7H8) → 8(C7H8), is predicted by the calculations to be exothermic
by 15 kJ mol–1 (Figure S22, Supporting Information). 29Si{1H} NMR
spectra recorded after 15 min at −20 °C showed no signals
expected for cation 7(C7H8) (compare
Figure S19c (Supporting Information) and
calculated values in Table ), but the characteristic resonances for silyltoluenium 8(C7D8) already dominated. In agreement
with the computational results, this experiment shows that germyltoluenium
ion 7(C7D8) undergoes a fast 1,2-shift
of the trimethylsilyl group to form the more stable cation 8(C7D8).
Figure 1
99 MHz 29Si{1H}
NMR spectra (toluene-d8, 253 K) of the
rearrangement of germyl toluenium
ion 3(C7D8) (•) to silyl
toluenium 8(C7D8) (*) (↓ Pr3SiMe). (a) Spectrum 2 h after
the addition of [Pr3Si(C7D8)][B(C6F5)4];
(b) after 9 h; (c) after 25 h; (d) after 120 d at 210 K.
Figure 2
99 MHz 29Si{1H} NMR spectrum (toluene-d8, 253 K) of silyl toluenium ion 8(C7D8) at 253 K in toluene synthesized by hydride
transfer from silane 15.
99 MHz 29Si{1H}
NMR spectra (toluene-d8, 253 K) of the
rearrangement of germyl toluenium
ion 3(C7D8) (•) to silyltoluenium 8(C7D8) (*) (↓ Pr3SiMe). (a) Spectrum 2 h after
the addition of [Pr3Si(C7D8)][B(C6F5)4];
(b) after 9 h; (c) after 25 h; (d) after 120 d at 210 K.99 MHz 29Si{1H} NMR spectrum (toluene-d8, 253 K) of silyl toluenium ion 8(C7D8) at 253 K in toluene synthesized by hydride
transfer from silane 15.Obviously, the barrier for this rearrangement is rather low
and
in a similar range as the 18 kJ mol–1 that is predicted
for the transformation of the free cations 7 → 8 (Figure S22, Supporting Information),[14] and therefore, germyltoluenium ion 7(C7D8) cannot be detected even at temperatures
as low as −20 °C.
Conclusions
We were able to synthesize
polysilanyl-substituted silyl and germyl
cations, such as silyl cation 8 and germyl cation 3 in the form of their arene complexes at low temperatures
using the weakly coordinating [B(C6F5)4]− anion. The cation synthesis was done using either
the stoichiometric reaction of germa-oligosilanes with a cationic
Lewis acid such as tri-iso-propylsilyltoluenium,
or, in a more regio-controlled way, by the standard hydride transfer
reaction between hydrido-silanes and -germanes with the trityl cation.
The cationic species were identified by low-temperature NMR spectroscopy
supported by the results of quantum mechanical NMR chemical shift
calculations. Applying these techniques, we were able to verify several
silyl- and germyl cations and their solvent complexes as intermediates
in the sila-Wagner–Meerwein rearrangement of silagermane 1 (eq 1, Scheme ) and to provide strong evidence for their clean interconversion
at low temperatures. We are convinced that this reaction is archetypical
for many synthetic useful Lewis acid catalyzed skeletal rearrangements
in polysilanes and polygermasilanes. Additional stable ion studies
for sila-Wagner–Meerwein rearrangement of oligosilanes of higher
complexity are currently under investigation in our laboratory to
identify general rules and reaction patterns.
Experimental
Section
All manipulations of air- and moisture-sensitive
compounds were
carried out under argon or nitrogen atmosphere using Schlenk techniques
or a standard glovebox (Braun Unilab). Glassware was dried in an oven
at 120 °C and evacuated prior to use. The solvents tetrahydrofuran
(THF), dimethoxyethane (DME), n-pentane, benzene,
and toluene were dried over sodium and distilled under nitrogen atmosphere.
Chlorobenzene was dried over CaCl2 and stored over molecular
sieves. Deuterated benzene and toluene were stored over molecular
sieves after drying over sodium. Dichlorodimethylgermane, chlorodimethylsilane,
triethylsilane, tri-iso-propylsilane and trimethylchlorosilane
were obtained from commercial suppliers, and the silanes were dried
over molecular sieves. Sodium methanolate was prepared by addition
of sodium to an excess of abs. methanol. After all
sodium was consumed, the solvent was removed in vacuo. Triphenylmethyl tetrakis(pentafluorophenyl) borate 9 ([Ph3C][B(C6F5)4]) was
prepared according to a modified literature procedure.[24] Tetrakis(trimethylsilyl)silane,[25] tetrakis(trimethylsilyl)germane[26]2, tris(trimethylsilyl)silylpotassium,[27] tris(trimethylsilyl)germylpotassium,[28] tris(trimethylsilyl)silyltrimethylgermane[29]1, chloropentamethyldisilane,[30] and trimethylsilane[31] were synthesized according to reported procedures. GC-MS spectra
were performed on a Thermo Focus DSQ. NMR spectra were recorded on
Bruker Avance 500, Avance III 500 and Varian Inova 300 spectrometers. 1H NMR spectra were calibrated against the residual proton
signal of the solvent as internal reference (benzene-d6/δ1H(C6D5H) =
7.20; toluene-d8/δ1H(CD2H) = 2.08; chloroform-d1/δ1H(CHCl3) = 7.24; chlorobenzene-d5/δ1H(C6D4HCl)
= 7.14) and 13C NMR spectra by using the central line of
the solvent signal (benzene-d6/δ13C(C6D6) = 128.0, toluene-d8/δ13C(C6D5CD3) = 20.4, chloroform-d1/δ13C(CDCl3) = 77.0, chlorobenzene-d5/δ13C(C6D5Cl) = 134.2). 29Si{1H} NMR spectra were calibrated
against an external standard (29Si(Me2SiHCl)
= 11.1 versus tetramethylsilane (TMS)). The 29Si{1H} NMR inverse gated spectra were recorded with a relaxation delay
D1 = 10 s. On the basis of our experiences, at −20 °C
this delay is long enough to allow a reliable integration of the peaks.
The 29Si{1H} INEPT spectra were recorded with
delays D3 = 8.4 ms and D4 = 31.3 ms. IR spectra were recorded on a
Bruker Tensor 27 instrument. Analysis values for carbon show often
too low values, which we attribute to the formation and incomplete
combustion of silicon carbide, although vanadium pentoxide as combustion
aid was used.
Dimethoxydimethylgermane (17)
This compound
was prepared according to slightly modified literature procedures.[32] NaOMe (2.84 g) (3.5 eq., 52.57 mmol) was suspended
in 40 mL of pentane, and 1.74 mL (15.00 mmol) of dichlorodimethylgermane
was slowly added with a syringe. The mixture was stirred overnight
at room temperature. The excess of NaOMe and formed NaCl were separated
from the solution by using a centrifuge (20 min, 2000 rpm), and then
the product-containing pentane solution was decanted using a Teflon
tube. The salts were washed with 10 mL of pentane and again centrifuged
and decanted. The pentane solutions were combined, and the product
was separated from the solvent by fractionated distillation (bp.:
118 °C at normal pressure (1.45 g, 58%). 1H NMR (500.13
MHz, 297.9 K, C6D6, δ): 0.30 (s, 6H, (CH3)2Ge), 3.54 (s, 6H, Ge(OCH3)2). 13C{1H} NMR (125.77 MHz, 298.1K, C6D6, δ): −2.9 ((CH3)2Ge), 51.6 (Ge(OCH3)2). Mass required for C4H12GeO2: 166.0. Mass found GC/MS: 164.9
(0.5) [M+-H], 150.8 (100) [M+-Me], 135.9 (72)
[M+-OMe], 120.9 (88) [M+-OMe-Me], 104.9 (84)
[M+-OMe-2Me].
A solution of 4.50 mmol tris(trimethylsilyl)silylpotassium[27] in 40 mL of pentane and a solution of 0.75 g
(4.50 mmol) of dimethoxydimethylgermane 17 in 10 mL of
pentane were cooled to 0 °C. The silyl potassium compound was
added dropwise to the germane solution. The ice bath was allowed to
warm to room temperature overnight. The reaction mixture was then
hydrolyzed with 1 M hydrochloric acid. The organic layer was separated
and dried over sodium sulfate. The solvent was removed under reduced
pressure, and the product was purified by Kugelrohr distillation (0.62
g, 36%). Because of the use of hydrochloric acid, about 14% of the
corresponding germyl chloride was formed as a byproduct, which was
detected in the GC chromatograms and NMR spectra. 1H NMR
(499.87 MHz, 305.0 K, C6D6, δ ppm): 0.34
(s, 27H, (CH3)3Si), 0.60 (s, 6H, (CH3)2Ge), 3.54 (s, 3H, CH3OGe). 13C{1H} NMR (125.69 MHz, 305.0 K, C6D6, δ
ppm): 2.6 ((CH3)3Si), 4.1 ((CH3)2Ge), 52.6 (CH3OGe). 29Si{1H} INEPT NMR (99.31 MHz, 305.0 K, C6D6, δ
ppm): −124.6 (((CH3)3Si)3Si), −9.9 (((CH3)3Si)3Si). Mass required for C12H36GeOSi4: 382.1. Mass found GC/MS: 367.2 (1) [M+-Me], 351.2 (0.5) [M+-OMe], 278.0 (8) [M+-SiMe3-OMe], 205.1 (13) [M+-2SiMe3-OMe], 73.0 (100) [Me3Si+]. No satisfactory
combustion analysis was available due to contamination with the chloride
side-product.
Tris(trimethylsilyl)silyldimethylgermane
(14)
A solution of 0.62 g (1.62 mmol) tris(trimethylsilyl)silyldimethylmethoxygermane 18 in 30 mL of THF and a suspension of 0.062 g (1.62 mmol)
of LiAlH4 in 50 mL of THF were cooled to 0 °C with
an ice bath. The solution of silagermane 18 was added
to the LiAlH4 suspension and the reaction mixture was stirred
for 20 min at 0 °C before it was allowed to warm to room temperature
and stirred for another 20 min. The mixture was slowly added to ice
cold 2 M sulfuric acid. The phases were separated, and the aqueous
phase was extracted two times with 50 mL of diethyl ether. The combined
organic phases were dried over sodium sulfate and filtered, and the
solvent was removed under reduced pressure. The product was crystallized
from ethanol as a waxy, colorless solid (0.38 g, 1.09 mmol, 67%). 1H NMR (499.87 MHz, 305.0 K, C6D6, δ
ppm): 0.30 (s, 27H, (CH3)3Si), 0.50 (d, 3JH,H = 4.2 Hz, 6H, (CH3)2Ge), 4.04 (sept, 3JH,H = 4.2 Hz, 1H, GeH). 13C{1H} NMR (125.69
MHz, 305.0 K, C6D6, δ ppm): −2.3
((CH3)2Ge), 2.5 ((CH3)3Si). 29Si{1H} NMR (99.31 MHz, 305.0 K, C6D6, δ ppm): −128.3 (((CH3)3Si)3Si), −9.4
(((CH3)3Si)3Si). Mass required for C11H34GeSi4: 352.1. Mass found GC/MS: 351.1 (0.1) [M+-H], 337.1 (0.6)
[M+-Me-H], 278.0 (30) [M+-SiMe3-H],
189.1 (13) [M+-2SiMe3-Me-H], 174.0 (4) [M+-SiMe3-GeMe2H], 73.1 (100) [Me3Si+]. IR (ATR, neat): νGe–H 1982
cm–1. Anal. found/calcd. for C11H34GeSi4: C 37.63/37.60, H 10.67/9.75.
Tris(trimethylsilyl)germyldimethylsilane
(15)[34]
Solutions
of 2.99 mmol tris(trimethylsilyl)germylpotassium[28] in 30 mL of DME and of 0.6 mL (excess, 5.52
mmol) of chlorodimethylsilane in 30 mL of DME were cooled to 0 °C
with an ice bath. The germylpotassium compound was slowly added to
the chlorosilane solution during 1 h. The ice bath was allowed to
warm to room temperature overnight. The reaction mixture was then
hydrolyzed with 1 M sulfuric acid. The organic layer was separated,
and the aqueous phase was extracted with 10 mL of diethyl ether. The
combined organic phases were dried over sodium sulfate, and the filtrate
was concentrated to 5 mL under reduced pressure. The product was crystallized
by adding 2 mL of acetonitrile as a colorless, waxy solid (0.847 g,
80.6%). 1H NMR (499.87 MHz, 305.0 K, CDCl3,
δ ppm):[34] 0.22 (s, 27H, (CH3)3Si), 0.27 (d, 3JH,H = 4.2 Hz, 6H, (CH3)2Si), 4.12 (sept, 3JH,H = 4.2 Hz, 1H, SiH). 13C{1H} NMR (125.71 MHz, 305.0 K, CDCl3, δ ppm): −1.4 ((CH3)2Si), 3.1
((CH3)3Si). 29Si INEPT NMR (99.31
MHz, 305.0 K, CDCl3, δ ppm): −29.8 (dsept, 1JSi,H = 180.5 Hz, 2JSi,H = 7.0 Hz, SiH), −4.7 ((CH3)3Si). Mass required for C11H34GeSi4: 352.1. Mass found GC/MS: m/z (%)
= 351.1 (0.3) [M+-H], 337.1 (2.5) [M+-Me-H],
278.1 (64) [M+-SiMe3-H], 189.9 (22) [M+-2SiMe3-Me-H], 174.0 (2) [M+-SiMe3-GeMe2H], 73.0 (100) [Me3Si+]. IR
(ATR, neat): νSi–H 2085 cm–1. Anal. found/calcd. for C11H34GeSi4: C 36.31/37.60, H 9.98/9.75.
A solution of 1.37 mmol tris(trimethylsilyl)germylpotassium·18-crown-6[28] in 3 mL of benzene was added dropwise to a solution
of 0.25 g (1.51 mmol) of chloropentamethyldisilane[30] in 3 mL of benzene. After 5 h, the solution mixture was
quenched with 1 M sulfuric acid, and the phases were separated. The
aqueous phase was extracted with pentane, and the combined organic
phases were dried over sodium sulfate and filtered, and the solvent
was removed under reduced pressure. The product was obtained as colorless
crystals by crystallization from methanol/diethyl ether 1:2 (0.42
g, 73%). 1H NMR (299.94 MHz, 298.0 K, C6D6, δ ppm): 0.22 (s, 9H, Si(CH3)2Si(CH3)3), 0.36 (s,
27H, ((CH3)3Si)3Ge), 0.40 (s, 6H, Si(CH3)2Si(CH3)3). 13C{1H} NMR
(75.43 MHz, 298.0 K, C6D6, δ ppm): −0.8
(Si(CH3)2Si(CH3)3), −0.4 (Si(CH3)2Si(CH3)3), 4.0 (((CH3)3Si)3Ge). 29Si{1H} INEPT NMR (59.59 MHz, 295.0 K, C6D6, δ ppm): −34.0 (Si(CH3)2Si(CH3)3), −15.5 (Si(CH3)2Si(CH3)3), −5.2 (((CH3)3Si)2Ge). Mass required for C14H42GeSi5: 424.1. Mass found GC/MS: m/z (%)
= 424 (1) [M+]; 408 (1) [M+-Me-H]; 351 (3) [M+-SiMe3]; 278 (10) [M+-2SiMe3]; 259 (1) [GeSi3C7H17+]; 243 (1) [GeSi3C6H13+]; 219 (3) [GeSi2C6H17+]; 203 (11) [M+-3SiMe3-2H]; 187(8)[M+-3SiMe3-Me-4H]; 147 (7) [GeSiMe3+]; 131 (35) [SiMe3SiMe2+]; 73 (100)
[SiMe3+]. Anal. found/calcd. for C14H42GeSi5 C 39.33/39.70, H 9.50/9.99.
A mixture of 0.21 g (0.49 mmol) of germapolysilane 19, 0.062 g (0.51 mmol) of KOBu, and 0.134 g (0.51 mmol) of 18-crown-6 ether was dissolved in
2 mL of benzene. After the complete formation of germylpotassium compound 20 was confirmed by NMR spectroscopy, the solution was added
to a stirred mixture of 10 mL of degassed diethyl ether and 20 mL
of degassed 2 M sulfuric acid cooled with an ice bath. The phases
were separated, the aqueous phase was extracted with degassed diethyl
ether, and the combined organic phases dried over sodium sulfate.
The solvents were removed under reduced pressure, and the product
was obtained as a colorless oil (0.15 g, 91%). The compound is sensitive
to oxygen and should be stored under argon at −20 °C.
General Preparation of
Trialkylsilyl Arenium Borates (10a–d)[33]
Triphenylmethyl tetrakis(pentafluorophenyl)borate
was dissolved
in 3 mL of the indicated solvent, and the silane was added. The formation
of two phases could be observed, and the biphasic reaction mixture
was vigorously stirred for 30 min. The upper, nonpolar phase was removed,
and the lower, polar phase was washed with 2 mL of the used solvent,
and again the nonpolar phase was removed. The polar phase was dried
under reduced pressure for 30 min and then dissolved in the respective
deuterated solvent.
Triethylsilyl Benzenium Borate (10a)
Triphenylmethyl
tetrakis(pentafluorophenyl)borate (0.50 g (0.54 mmol)) was dissolved
in 3 mL of benzene, and 0.14 mL (1.6 eq., 0.87 mmol) of triethylsilane
was added.
Tri-iso-propylsilyl Toluenium
Borate (10b)
Triphenylmethyl tetrakis(pentafluorophenyl)borate
(0.46 g (0.50 mmol)) was dissolved in 3 mL of toluene, and 0.11 mL
(1.1 eq., 0.55 mmol) of triiso-propylsilane was added.
Trimethylsilyl Toluenium Borate (10c)
Triphenylmethyl
tetrakis(pentafluorophenyl)borate (0.46 g (0.50 mmol))
was dissolved in 3 mL of toluene, and 0.06 mL (1.1 eq., 0.55 mmol)
of trimethylsilane was added.
General Procedure for the
Rearrangement of Tris(trimethylsilyl)silyltrimethylgermane
(1) with Trialkylsilyl Arenium Borates (10a–c)
A solution of 0.18 g (0.50 mmol)
of silagermane 1 in 1 mL of the named deuterated solvent
was added to a precooled solution of the named trialkylsilylareniumborate 10a–c. The reaction mixture
was stirred for 2 h at the specified temperature and then allowed
to warm to room temperature. The polar phase and the nonpolar phase
were each transferred to separate NMR tubes to be analyzed independently.
In the following reactions, the NMR spectra of the polar phase showed
too may signals to be analyzable, but the compounds in the nonpolar
phase were identified by NMR and GC/MS spectroscopy (see Figures S8–11
in the Supporting Information for details).
Rearrangement of Tris(trimethylsilyl)silyldimethylgermyl
Toluenium Borate (3(C7D8)[B(C6F5)4]) to Tris(trimethylsilyl)germyldimethylsilyl
Toluenium Borate (8(C7D8)[B(C6F5)4]) Starting from Silagermane (1)
To a solution of 0.18 g (0.50 mmol) of tris(trimethylsilyl)silyltrimethylgermane 1 in 1 mL of toluene-d8 cooled
to −20 °C, 1 equiv of tri-iso-propylsilyl
toluenium borate 10b was slowly added via a Teflon tube.
The mixture was stirred for 1 h at −20 °C. The brown polar
phase and the light yellow nonpolar phase were each transferred to
separate NMR tubes at −20 °C and stored at −60
°C overnight until the NMR spectra were recorded the next morning.
The polar phase contained borates [3(C7D8)][B(C6F5)4] and [8(C7D8)][B(C6F5)4]). The nonpolar phase contained methyl-triiso-propylsilane
and the rearrangement product 2. Polar phase1H NMR (499.87 MHz, 253.0 K, C7D8, δ ppm): −0.15 (((CH3)3Si)3GeSi(CH3)2+), 0.07 (((CH3)3Si)3GeSi(CH3)2+), 0.12
(((CH3)3Si)3SiGe(CH3)2+), 0.28 (((CH3)3Si)3SiGe(CH3)2+). 13C{1H} NMR (125.71
MHz, 253.0 K, C7D8, δ ppm): 1.6 (((CH3)3Si)3SiGe(CH3)2+), 2.5 (((CH3)3Si)3GeSi(CH3)2+), 5.5 (((CH3)3Si)3GeSi(CH3)2+), 12.1 (((CH3)3Si)3SiGe(CH3)2+). 29Si{1H} NMR (99.31 MHz, 253.0 K, C7D8, δ ppm): −87.9 (((CH3)3Si)3SiGe(CH3)2+), −7.9 (((CH3)3Si)3SiGe(CH3)2+), −2.4
(((CH3)3Si)3GeSi(CH3)2+), 98.1 (((CH3)3Si)3GeSi(CH3)2+) (see Figures S12a–c in the Supporting Information). Nonpolar phase1H NMR (499.87 MHz, 305.0 K, C7D8, δ ppm): −0.17 (s, ((CH3)2CH)3SiCH3), 0.26 (s, ((CH3)3Si)4Ge), 0.86 (sept, 3JH,H = 7.3 Hz, ((CH3)2CH)3SiCH3), 0.97 (d, 3JH,H = 7.3 Hz, ((CH3)2CH)3SiCH3). 13C{1H} NMR (127.71 MHz, 305.0 K, C7D8, δ ppm): −10.1 (((CH3)2CH)3SiCH3), 3.5 (((CH3)3Si)4Ge), 11.8 (((CH3)2CH)3SiCH3), 18.8 (((CH3)2CH)3SiCH3). 29Si{1H}
NMR (99.31 MHz, 305.0 K, C7D8, δ ppm):
−5.1 (((CH3)3Si)4Ge), 9.0
(((CH3)2CH)3SiCH3) (see
Figures S13a–d in the Supporting Information).
Rearrangement of Tris(trimethylsilyl)silyldimethylgermyl
toluenium Borate (3(C7D8)[B(C6F5)4]) to Tris(trimethylsilyl)germyldimethylsilyl
Toluenium Borate (8(C7D8)[B(C6F5)4]) Starting from Hydrogen Substituted
Silagermane (14)
tris(trimethylsilyl)silyldimethylgermane 14 (0.14 g (0.40 mmol)) and 0.37 g (0.40 mmol) of triphenylmethyl
tetrakis(pentafluorophenyl)borate were each dissolved in 1 mL of toluene-d8 and cooled to −20 °C. Silagermane 14 was slowly added to the borate salt via a Teflon tube,
and the mixture was stirred at −20 °C for 1.5 h. The brown
polar phase and the light yellow nonpolar phase were each transferred
to NMR tubes at −20 °C. The NMR spectra were recorded
at −20 °C. The polar phase contained borates [3(C7D8)][B(C6F5)4] and [8(C7D8)][B(C6F5)4]). The nonpolar phase contained triphenylmethane
and the rearrangement product 2. Polar phase1H NMR (499.87 MHz, 253.0 K, C7D8, δ ppm): −0.15 (((CH3)3Si)3GeSi(CH3)2+), 0.07 (((CH3)3Si)3SiGe(CH3)2+), 0.12
(((CH3)3Si)3GeSi(CH3)2+), 0.28 (((CH3)3Si)3SiGe(CH3)2+). 13C{1H} NMR (125.71
MHz, 253.0 K, C7D8, δ ppm): 1.6 (((CH3)3Si)3SiGe(CH3)2+), 2.4 (((CH3)3Si)3GeSi(CH3)2+), 5.5 (((CH3)3Si)3GeSi(CH3)2+), 12.1 (((CH3)3Si)3SiGe(CH3)2+). 29Si{1H} NMR (99.31 MHz, 253.0 K, C7D8, δ ppm): −87.9 (((CH3)3Si)3SiGe(CH3)2+), −7.9 (((CH3)3Si)3SiGe(CH3)2+), −2.4
((CH3)3Si)3GeSi(CH3)2+)), 98.1 (((CH3)3Si)3GeSi(CH3)2+) (see Figures S14a–c and S15a–b
in the Supporting Information). Nonpolar
phase1H NMR (499.87 MHz, 305.0 K, C7D8, δ ppm): 0.30 (s, ((CH3)3Si)4Ge), 5.38 (s, Ph3CH), 6.98–7.09 (m, Ph3CH). 13C{1H} NMR (127.71 MHz, 305.0 K, C7D8, δ ppm): 3.5 ((CH3)3Si)4Ge), 57.1 (Ph3CH), 125.4 (Ph3CH), 128.3 (Ph3CH), 129.2 (Ph3CH), 144.3
(Ph3CH). 29Si{1H} NMR (99.31 MHz, 305.0 K, C7D8, δ ppm):
−5.1 ((CH3)3Si)4Ge) (see Figures
S16a–c in the Supporting Information).
Tris(trimethylsilyl)germyldimethylsilyl Toluenium Borate
(8(C7H8)[B(C6F5)4]) from Silane 15
Tris(trimethylsilyl)germyldimethylsilane 15 (0.18 g (0.50 mmol)) and 0.46 g (0.50 mmol) of triphenylmethyl
tetrakis(pentafluorophenyl)borate were both dissolved in 1 mL of toluene-d8 and cooled to −20 °C. Germylsilane 15 was slowly added to the borate salt via a Teflon tube,
and the mixture was stirred at −20 °C for 1.5 h. The brown
polar phase and the light yellow nonpolar phase were each transferred
to separate NMR tubes at −20 °C and stored at −60
°C overnight until the NMR spectra were recorded the next morning.
The polar phase contained borate[8(C7D8)][B(C6F5)4]. The nonpolar
phase contained triphenylmethane and the rearrangement product 2. Polar phase1H NMR (499.87 MHz,
253.0 K, C7D8, δ ppm): −0.15 (((CH3)3Si)3GeSi(CH3)2+), 0.12 (((CH3)3Si)3GeSi(CH3)2+). 13C{1H} NMR (125.71 MHz,
253.0 K, C7D8, δ ppm): 2.5 (((CH3)3Si)3GeSi(CH3)2+), 5.5 (((CH3)3Si)3GeSi(CH3)2+). 29Si{1H} NMR (99.31 MHz, 253.0
K, C7D8, δ ppm): −2.4 (((CH3)3Si)3GeSi(CH3)2+), 98.1 (((CH3)3Si)3GeSi(CH3)2+) (see Figures S17a–c in the Supporting Information). Nonpolar phase1H NMR (499.87 MHz, 305.0 K, C7D8, δ
ppm): 0.29 (s, ((CH3)3Si)4Ge), 5.38
(s, Ph3CH), 6.98–7.10 (m, Ph3CH). 13C{1H} NMR
(125.69 MHz, 305.0 K, C7D8, δ ppm): 3.5
(((CH3)3Si)4Ge), 57.2 (Ph3CH), 125.4 (Ph3CH), 128.3 (Ph3CH), 129.2
(Ph3CH), 144.3 (Ph3CH). 29Si{1H} NMR (99.31 MHz, 305.0
K, C7D8, δ ppm): −5.1 (((CH3)3Si)4Ge) (see Figure S18a–c
in the Supporting Information).
Tris(trimethylsilyl)germyldimethylsilyl-phenylchloronium
Borate (8(C6D5Cl)[B(C6F5)4]) from Silane 15
Tris(trimethylsilyl)germyldimethylsilane 15 (0.09 g
(0.25 mmol)) and 0.23 g (0.25 mmol) of triphenylmethyl tetrakis(pentafluorophenyl)borate
were both dissolved in 0.5 mL of chlorobenzene-d5 and cooled to −20 °C. Germylsilane 15 was slowly added to the borate salt via a Teflon tube, and the mixture
was stirred at −20 °C for 1.5 h. The brown solution was
transferred into a NMR tube at −20 °C and stored at −60
°C overnight until the NMR spectra were recorded the next morning.
The mixture contained borate [8(C6D5Cl)][B(C6F5)4], the rearrangement
product 2, and triphenylmethane. 29Si{1H} NMR (99.31 MHz, 253.0 K, C6D6Cl,
δ ppm): −5.1 (((CH3)3Si)4Ge), −1.4 (((CH3)3Si)3GeSi(CH3)2+), 154.3
(((CH3)3Si)3GeSi(CH3)2+).
Tris(trimethylsilyl)germyldimethylsilyl
Toluenium Borate
(8(C7D8)[B(C6F5)4]) from Germane 16
Bis(trimethylsilyl)pentamethyldisilanylgermane 16 (0.11 g (0.32 mmol)) and 0.29 g (0.32 mmol) of triphenylmethyl
tetrakis(pentafluorophenyl)borate were both dissolved in 1 mL of toluene-d8 and cooled to −20 °C. Silylgermane 18 was slowly added to the borate salt via a Teflon tube,
and the mixture was stirred at −20 °C for 5 min. The brown
polar phase and the light yellow nonpolar phase were each transferred
to separate NMR tubes at −20 °C and stored at −60
°C for 5 h until the NMR spectra were recorded. At −60
°C, the polar phase solidifies, and no further reaction is expected.
The NMR spectra of the polar phase recorded at −20 °C
contained nearly exclusively borate[8(C7D8)][B(C6F5)4]. The nonpolar
phase contained triphenylmethane and the rearrangement product 2. Polar phase1H NMR (499.87 MHz,
253.0 K, C7D8, δ ppm): −0.16 (((CH3)3Si)3GeSi(CH3)2+), 0.12 (((CH3)3Si)3GeSi(CH3)2+). 13C{1H} NMR (125.71 MHz,
253.0 K, C7D8, δ ppm): 2.5 (((CH3)3Si)3GeSi(CH3)2+), 5.5 (((CH3)3Si)3GeSi(CH3)2+). 29Si{1H} NMR (99.31 MHz, 253.0
K, C7D8, δ ppm): −2.4 (((CH3)3Si)3GeSi(CH3)2+)), 98.2 (((CH3)3Si)3GeSi(CH3)2+)) (see Figures S19a–c in the Supporting Information). Nonpolar phase1H NMR (499.87 MHz, 305.0 K, C7D8, δ
ppm): 0.30 (s, ((CH3)3Si)4Ge), 5.38
(s, Ph3CH), 6.98–7.10 (m, Ph3CH). 13C{1H} NMR
(125.69 MHz, 305.0 K, C7D8, δ ppm): 3.5
((CH3)3Si)4Ge), 57.1 (Ph3CH), 125.4 (Ph3CH), 128.2 (Ph3CH), 129.1
(Ph3CH), 144.3 (Ph3CH). 29Si{1H} NMR (99.31 MHz, 305.0
K, C7D8, δ ppm): −5.1 (((CH3)3Si)4Ge) (see Figure S20a–c
in the Supporting Information).