Synthesis and reactivity of the bulky germanium(IV) trisamide complexes BrGe[N(SiMe3)2]3 and LiGe[N(SiMe 3)2]3: X-ray crystal structures of BrGe[N(SiMe3)2]3 and [(Me3Si) 2N]3Ge(CH2CH2CH2CH 3)

Article abstract of DOI:10.1016/j.ica.2004.11.035

The bulky germanium(IV) trisamide complex BrGe[N(SiMe₃) ₂]₃ has been prepared and structurally characterized. Reaction of this material with ⁿBuLi resulted in [(Me ₃Si)₂N]₃Ge(CH₂CH₂CH ₂CH₃), which has also been structurally characterized, but no reaction with lithium metal was observed in the attempted preparation of LiGe[N(SiMe₃)₂]₃. However, analytically pure LiGe[N(SiMe₃)₂]₃was prepared by the reaction of 3 equivalents of LiN(SiMe₃)₂ with GeI_2. Evidence is presented that this germyllithium reagent is a major by-product in the synthesis of Ge[N(SiMe₃)₂]₂ when the latter reagent is obtained from GeCl₂ ? (dioxane) prepared by the method of Kolesnikov et al. Reaction of the germyllithium reagent with 2,6-diphenylphenol yields the mixed Ge/Li cage complex [Li(μ₂- OC₆H₃Ph₂-2,6)₃Ge], which had originally been obtained serendipitously via the reaction of 2,6-diphenylphenol with Ge[N(SiMe₃)]₂.

Full text of DOI:10.1016/j.ica.2004.11.035

Inorganica Chimica Acta 358 (2005) 1186–1192
www.elsevier.com/locate/ica
Synthesis and reactivity of the bulky germanium(IV)
trisamide complexes BrGe[N(SiMe3)2]3 and LiGe[N(SiMe3)2]3:
X-ray crystal structures of BrGe[N(SiMe3)2]3
and [(Me3Si)2N]3Ge(CH2CH2CH2CH3)
,1
*
Jennifer L. Walding, Phillip E. Fanwick, Charles S. Weinert
Department of Chemistry, Purdue University, 560 Oval Drive, West Lafayette, IN 47907, USA
Received 21 October 2004; accepted 24 November 2004
Available online 30 December 2004
Dedicated to Professor Duward F. Shriver on the occasion of his 70th birthday
Abstract
The bulky germanium(IV) trisamide complex BrGe[N(SiMe3)2]3 has been prepared and structurally characterized. Reaction of
this material with nBuLi resulted in [(Me3Si)2N]3Ge(CH2CH2CH2CH3), which has also been structurally characterized, but no reac-
tion with lithium metal was observed in the attempted preparation of LiGe[N(SiMe3)2]3. However, analytically pure LiGe[N(Si-
Me3)2]3was prepared by the reaction of 3 equivalents of LiN(SiMe3)2 with GeI2. Evidence is presented that this germyllithium
reagent is a major by-product in the synthesis of Ge[N(SiMe3)2]2 when the latter reagent is obtained from GeCl2 Æ (dioxane) prepared
by the method of Kolesnikov et al. Reaction of the germyllithium reagent with 2,6-diphenylphenol yields the mixed Ge/Li cage com-
plex [Li(l2-OC6H3Ph2-2,6)3Ge], which had originally been obtained serendipitously via the reaction of 2,6-diphenylphenol with
Ge[N(SiMe3)]2.
ꢀ 2004 Elsevier B.V. All rights reserved.
Keywords: Germanium; Crystal structure; NMR spectroscopy
1. Introduction
Si)2N]3MCl of the group 4 metals Ti, Zr, and Hf [5].
We have employed the germanium bisamide compound
Ge[N(SiMe3)2]2 (1) [2,3,6] in reactions with derivatized
phenols and resolved binaphthols for the synthesis of
germanium(II) aryloxide [7] and chiral binaphthoxide
[8] complexes. During the course of this study, we unex-
pectedly obtained a mixed lithium/germanium species
[Li(l2-OC6H3Ph2-2,6)3Ge] from the reaction of 2,6-
diphenylphenol with 1 [9], which had been prepared by
reduction of GeCl4 with Bu3SnH in the presence of diox-
ane [10], followed by reaction with 2 equivalents of
LiN(SiMe3)2 [2,3]. The germanium/lithium complex
[Li(l2-OC6H3Ph2-2,6)3Ge], as well as the entire series
of heavier congeners has previously been prepared by
the reaction of three equivalents of the alkali metal
The use of sterically encumbering ligands often
allows the stabilization of metal centers with low coordi-
nation numbers. In particular, the bulky ligand bis(trim-
ethylsilyl)amido [–N(SiMe3)2] [1] has been employed for
the preparation of stable bivalent compounds of the
group 14 metals Ge and Sn [2,3], the first examples of
three-coordinate group 7 metals M[N(SiMe3)2]3 (M =
Mn, Co) [4], and trisamide compounds [(Me3-
*
Corresponding author. Tel.: +1 405 7446543; fax: +1 405 7446007.
E-mail address: weinert@chem.okstate.edu (C.S. Weinert).
1
Present address: Department of Chemistry, Oklahoma State
University, 316 Physical Science, Stillwater OK 740783071, USA.
0020-1693/$ - see front matter ꢀ 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2004.11.035

J.L. Walding et al. / Inorganica Chimica Acta 358 (2005) 1186–1192
1187
phenoxide with GeI2 and the entire series has been struc-
turally characterized [9]. In our laboratory, the synthesis
of 1 was complicated by the formation of a lithium trisa-
mide LiGe[N(SiMe3)2]3 and this product co-distills with
the desired germanium(II) bisamide and accounts for
the formation of [Li(l2-OC6H3Ph2-2,6)3Ge] upon reac-
tion of this mixture with 2,6-diphenylphenol. Evidence
for this process is the focus of this study. We have char-
acterized the previously reported germanium(IV) trisa-
mide BrGe[N(SiMe3)2]3 (2) [11], which reacts with
nBuLi to yield the alkylgermanium species [(Me3Si)2N]3-
Ge(CH2CH2CH2CH3) (3). Although 2 could not be con-
verted to the germyllithium complex LiGe[N(SiMe3)2]3
4, this material was obtained in pure form by the reac-
tion of GeI2 with 3 equivalents of LiN(SiMe3)2. Reac-
tion of 4 with 2,6-diphenylphenol results in the
formation of [Li(l2-OC6H3Ph2-2,6)3Ge] (5). We have
obtained X-ray crystal structures of complexes 2 and 3
(b.p. = 79 ꢁC at 0.05 torr) and the identity of the product
was confirmed to be that of 4 by elemental analysis.
7
The Li NMR spectrum of pure 4 contains one reso-
nance at 1.05 ppm, while a 7Li NMR spectrum of
Ge[N(SiMe3)2]2 (1) prepared using the original method
of reduction of GeCl4 [10] for the synthesis of
GeCl2 Æ (dioxane) contained a resonance at 1.05 ppm
as well, indicating the presence of 4 as a contaminant
in the product mixture. The formation 4 is likely a result
of side products formed in the reduction of GeCl4 with
Et3SiH during the synthesis of GeCl2 Æ dioxane. This
preparation has been previously proven unreliable and
an improved synthetic method has been reported [12],
which cleanly gives 1 upon reaction with LiN(SiMe3)2.
Additionally, an alternate preparation of 1 also gives
contaminant-free product [6]. Reaction of 2,6-diphenyl-
phenol with compound 1 prepared by either of these
methods gives the expected bisaryloxide Ge(OC6H4-
Ph2-2,6)2, while reduction of GeCl4 by the original
method and subsequent reaction with LiN(SiMe3)2 gives
a product which contains 4 as a major contaminant and
gives the trisaryloxide complex 5 upon reaction with
phenol. The similarity in boiling points of 4 and 1 de-
spite the higher molecular weight of 4 renders separation
by distillation difficult. Complex 5 was obtained by reac-
tion 4 with 3 equivalents of 2,6-diphenylphenol in 82%
yield.
7
and have employed Li NMR spectroscopy to investi-
gate the synthesis of the lithium/germanium complex
via both the rational and serendipitous routes.
2. Results and discussion
2.1. Synthesis
The germanium(IV) trisamide BrGe[N(SiMe3)2]3 (2)
was obtained in moderate yield via the previously re-
ported synthetic method involving insertion of Ge-
[N(SiMe3)2]2 (1) into the N–Br bond of BrN(SiMe3)2
[11]. Compound 2 is stable with respect to oxygen
and moisture due to the presence of three sterically
encumbering bis(trimethylsilyl)amido groups. Stirring
a suspension of 2 in water in air for 7 days did not
result in hydrolysis. Furthermore, protonolysis reac-
tions involving 2 and 2,6-diphenylphenol, which have
been useful for the preparation of germanium(II) aryl-
oxides from 1 [9], were unsuccessful and no reaction
was observed between 2 and various alcohols or other
protic reagents.
2.2. Crystallographic studies
Recrystallization of 2 from hot benzene yielded crys-
tals which were suitable for an X-ray structure determi-
nation. An ORTEP diagram of 2 is shown in Fig. 1 and
selected bond distances and angles are summarized in
Table 1. Compound 2 crystallizes in the trigonal space
group R3c and has a threefold axis of rotation directed
along the Ge–Br bond. The methyl groups of the three –
N(SiMe3)2 ligands interlock in a gear-like fashion. A
comparison of the structure of 2 with that of Ge[N-
(SiMe3)2]2 (1) [13] reveals that the Ge–N bond lengths
˚
of 1.848(3) A in 2 are shorter than those of 1 which have
˚
an average value of 1.876(5) A, while the N–Ge–N an-
Despite the steric bulk of 2, this species could be con-
verted to the butyl complex [(Me3Si)2N]3Ge(CH2-
CH2CH2CH3) (3) by reaction with a slight molar excess
gles of 115.52(6)ꢁ in 2 are substantially more obtuse than
that of 1 which measures 107.1(2)ꢁ. The shorter Ge–N
bonds in 2 are likely due to the presence of the higher
formal charge on germanium (4+) in 2 versus 1 (2+),
while the larger N–Ge–N angles are required to accom-
modate three bis(trimethylsilyl)amido-ligands rather
than two.
n
of BuLi in refluxing benzene for 18 h. However, large
n
molar excesses (greater than 1.4 equivalents) of BuLi
not only cause cleavage of the Ge–Br bond but also
the Ge–N bonds giving a mixture of products. Because
we had unexpectedly obtained [Li(l2-OC6H3Ph2-2,6)3-
Ge] (5) by reaction of 2,6-diphenylphenol with 1, we
were interested in obtaining the lithium/germanium
amide LiGe[N(SiMe3)2]3 (4) from 2 but attempts to con-
vert compound 2 to 4 by reaction with lithium metal
were unsuccessful. However, reaction of GeI2 with three
equivalents of LiN(SiMe3)2 resulted in a pale orange li-
quid which could be purified by vacuum distillation
˚
The Ge–Br bond distance in 2 of 2.3861(6) A is signif-
˚
icantly shorter than the Sn–Br distance of 2.519(2) A in
BrSn[N(SiMe3)2]3 (6) [11,14], reflecting the smaller cova-
˚
lent radius of Ge (1.22 A) relative to Sn (1.40 A) [15].
˚
The Ge–N distance of 1.848(3) A in 2 is shorter than
˚
the Sn–N distance of 2.056(7) A in 6 and also shorter
˚
˚
than the Ti–N bond of 1.940(10) A in the similar group

1188
J.L. Walding et al. / Inorganica Chimica Acta 358 (2005) 1186–1192
Fig. 2. Space filling drawing of BrGe[N(SiMe3)2]3 (2).
Hf) [5]. Elongation of the N–Si bond is indicative of
covalent character in the M–N bond and is most pro-
nounced for 6 while lessened for 2. Tin is slightly less
electropositive than germanium and thus the Sn–N
bond is expected to be more covalent in nature than
the Ge–N bond. The shortening of the N–Si distances
in the group 4 complexes is likely a manifestation of
the ability of transition metals to engage in M–L p-inter-
actions, where the M–N bond exhibits some double
bond character resulting from donation of electron den-
sity from nitrogen to the formally d0 metal center. This
would make the nitrogen atoms electron deficient, and
this deficiency is then alleviated by electron donation
from the Me3Si-groups resulting in a shorter N–Si bond.
A space-filling diagram of 2 is shown in Fig. 2. The
bromine atom is significantly encapsulated by the
methyl groups of the N(SiMe3)2 ligands, while the ger-
manium atom is completely encapsulated by the ligand
system. The bulky ligands of 2 render it kinetically inert,
which explains the lack of reaction with 2,6-diphenyl-
phenol and various other protic reagents. This also ex-
plains the inertness of compound 2 with respect to air
and moisture.
Fig. 1. ORTEP plot of BrGe[N(SiMe3)2]3 (2). Thermal ellipsoids are
drawn at 50% probability.
˚
4 complex ClTi[N(SiMe3)2]3 (d Ti–N = 1.940(10) A) [5],
which might be expected due to the larger covalent
˚
radius of Ti (1.32 A) [15]. However, when correcting
for the difference in covalent radii between germanium
˚
and tin (0.18 A) and germanium and titanium
˚
(0.10 A), all three M–N bond distances are essentially
the same. The Ge–N distance in 2 is also very similar
˚
to the average Ga–N bond of 1.839(4) A in ClGa[N-
(Si-Me3)2]2 [16], where the two elements have very similar
˚
covalent radii (rcov = 1.25 A for Ga). The Ge–N distance
in 2 is significantly longer than the average Ge–N bond
˚
length of 1.780(4) A in the mixed aryloxide/amido com-
plex [Ge(OC6H3Ph2-2,6)2(NMe2)2] [17], which is likely a
result of the increased steric bulk of the three –
N(SiMe3)2 ligands relative to that of the –NMe2 ligands.
The N–Si bonds in 2 have an average value of
˚
1.783(3) A and are longer than that in the tin congener
˚
6 of 1.758(8) A [11,14] but are very similar to the average
˚
N–Si distances in ClM[N(SiMe3)2]3 of 1.77(1) A
The N–M–N bond angles in 2 and 6 and in the group
4 amides are all nearly identical, ranging from 114.1(1)ꢁ
in ClHf[N(SiMe3)2]3 [5] to 115.52(6)ꢁ in 2. All of these
metal complexes can be regarded as isostructural and
crystallize in the R3c space group. They exhibit C3v sym-
metry and contain a crystallographically imposed C3
axis directed along the metal–halide bond. The same is
not true for compound 3, however, which exhibits a lower
symmetry due to the presence of a butyl group. Crystal-
lization of 3 from hot hexane yielded crystals which were
suitable for an X-ray structure determination. An OR-
TEP diagram is shown in Fig. 3, and selected bond dis-
tances and angles are summarized in Table 2.
˚
(M = Ti), 1.766(4) A (M = Zr), and 1.78(1) A (M =
˚
Table 1
˚
Selected bond lengths (A) and angles (ꢁ) for [BrGe{N(SiMe3)2}3] (1)
Ge–Br
Ge–N
2.3861(6)
1.848(3)
N–Si(1)
N–Si(2)
1.781(3)
1.784(3)
N–Ge–N
N–Ge–Br
115.52(6)
N–Si(1)–C(12)
N–Si(1)–C(13)
N–Si(2)–C(21)
N–Si(2)–C(22)
N–Si(2)–C(23)
113.32(1)
111.1(1)
113.5(2)
111.4(1)
112.6(2)
102.39(8)
125.8(2)
118.4(1)
115.3(2)
115.7(1)
Ge–N–Si(1)
Ge–N–Si(2)
Si(1)–N–Si(2)
N–Si(1)–C(11)
The three Ge–N bonds in 3 are longer than those of 2
˚
and measure 1.870(2), 1.891(2), and 1.908(2) A in

J.L. Walding et al. / Inorganica Chimica Acta 358 (2005) 1186–1192
1189
group in 3 is also less sterically encumbering than the
bromine atom in 2, since only the a-methylene group af-
fects the steric environment about the germanium atom.
This diminished steric environment allows the three ami-
do groups in 3 to adopt more acute bond angles than
those of 2.
The germanium–nitrogen bond lengths in 2 and 3 fall
well within the range for germanium complexes contain-
ing one or more –N(SiMe3)2 ligands. A oxadiazagerma-
nine carboxalate complex containing two –N(SiMe3)2
ligands exhibits short Ge–N bond lengths of 1.785(6)
˚
and 1.799(6) A [18], while the germanium(II) complex
[Me3SiNB(tBu)NSiMe3]Ge[N(SiMe3)2] has a long Ge–
˚
N distance of 1.910(2) A [19]. The Ge–N bond in 2 is
very similar in length to those in the germanium(IV)
complex [{(Me3Si)2N}2GeS]2 which have an average dis-
˚
tance of 1.842(4) A [20], and to the Ge–N bond in 1-
bis(trimethylsilyl)aminogermatrane which measures
˚
1.845(3) A [21], which also contains a germanium(IV)
Fig. 3. ORTEP plot of [(Me3Si)2N]3Ge(CH2CH2CH2CH3) (3). Ther-
mal ellipsoids are drawn at 50% probability.
center. However, the Ge–N(3) bond in 3 is exceptionally
long for a germanium(IV) complex containing a
bis(trimethylsilylamido) ligand. The aforementioned
length, with the Ge–N(3) bond being the longest and the
Ge–N(1) bond being the shortest. The Ge–N(1) bond is
slightly shorter than those in Ge[N(SiMe3)2]2 (1), which
[Me3SiNB(tBu)NSiMe3]Ge[N(SiMe3)2] is formally
a
complex of germanium(II) and other structurally char-
acterized species containing a –N(SiMe3)2 ligand with
˚
measure 1.878(5) and 1.873(5) A [13]. Two of the three
N–Ge–N angles measuring 106.23(9)ꢁ and 109.99(9)ꢁ
are substantially more acute in 3 than in 2, but are sim-
ilar to the N–Ge–N angle in 1 which measures 107.1(2)ꢁ
[13]. The third N–Ge–N angle of 116.08(9)ꢁ is more ob-
tuse than those of both 2 and 1. Thus, the three bis(trim-
ethylsilyl)amido-ligands are not symmetry related in 3
and the trimethylsilyl groups do not interlock in a
gear-like fashion as found in compound 2. The longer
Ge–N distances in 3 relative to 2 are a result of the pres-
ence of the less electron-withdrawing butyl group com-
pared with the bromine atom in 2, and the shorter
Ge–N bonds in 2 indicate that the three amido groups
are more tightly held to the germanium metal center
than in 3. In addition to the electronic effects, the butyl
˚
a Ge–N bond longer than 1.90 A are germanium(II) spe-
cies complexed to late transition metals [22,23].
2.3. Variable temperature NMR studies
The steric nature of the three –N(SiMe3)2 groups in 2
and their ability to interlock in a gear-like fashion result
in restricted rotation about the Ge–N bonds in solution
1
and is evident in the H NMR spectrum of 2. The 300
1
MHz H NMR spectrum of 2 in benzene-d6 contains
two resonances of equal intensity at d 0.59 and 0.40
ppm. Three of the six trimethylsilyl ligands are proximal
to the bromine atom, while three are distal which result
in two Si(CH3)3 resonances instead of a single resonance
in an intensity ratio of 1:1. A single resonance was pre-
viously reported for 2 and at d 0.65 ppm at a lower field
strength of 60 MHz [11]. Presumably the two peaks were
not resolvable at lower field.
Table 2
˚
Selected bond lengths (A) and angles (ꢁ) for [{(Me3Si)2N}3-
Ge(CH2CH2CH2CH3)] (2)
Complex 2 was subjected to a variable-temperature
1H NMR experiment in toluene-d8 in order to determine
the barrier to rotation about the Ge–N bonds. At
ꢀ30 ꢁC, the two lines have a linewidth at half height
of Dm1/2 = 4.13 Hz and warming the sample to 25 ꢁC re-
sults in a broadening of the resonances to Dm1/2 = 10.80
Hz. Upon increasing the temperature to 50 ꢁC, the peaks
begin to coalesce and finally become a single broad fea-
ture at d 0.46 ppm at a temperature of 66 ꢁC. This res-
onance becomes sharper with increasing temperature
and has a linewidth of Dm1/2 = 6.23 Hz at 110 ꢁC. The
barrier to rotation for the –SiMe3 groups is calculated
to be DGꢀ = 16.7 0.5 kcal/mol. This value is higher
Ge–N(1)
Ge–N(3)
1.870(2)
1.908(2)
1.527(4)
1.518(4)
1.762(2)
1.780(2)
1.768(2)
Ge–N(2)
Ge–C(41)
1.891(2)
1.968(3)
1.514(4)
1.769(2)
1.754(2)
1.776(2)
C(41)–C(42)
C(43)–C(44)
N(1)–Si(12)
N(2)–Si(22)
N(3)–Si(32)
C(42)–C(43)
N(1)–Si(11)
N(2)–Si(21)
N(3)–Si(31)
N(1)–Ge–N(2)
N(2)–Ge–N(3)
N(2)–Ge–C(41)
Ge–N(1)–Si(11)
Ge–N(2)–Si(21)
Ge–N(3)–Si(31)
Ge–C(41)–C(42)
116.08(9)
106.23(9)
103.0(1)
120.1(1)
131.6(1)
118.4(1)
121.0(2)
N(1)–Ge–N(3)
N(1)–Ge–C(41)
N(3)–Ge–C(41)
Ge–N(1)–Si(12)
Ge–N(2)–Si(22)
Ge–N(3)–Si(32)
109.99(9)
105.4(1)
116.4(1)
121.7(1)
114.3(1)
127.2(1)

1190
J.L. Walding et al. / Inorganica Chimica Acta 358 (2005) 1186–1192
than the values reported for the group 4 metal trisa-
mides ClM[N(SiMe3)2]3 (M = Ti, DGꢀ = 15.7 0.5
kcal/mol; M = Zr, DGꢀ = 14.2 0.5 kcal/mol and
M = Hf, DGꢀ = 14.1 0.5 kcal/mol) [5]. The Ge(IV) cen-
ter has a smaller covalent radius than Ti(IV) and there-
fore rotation about the M–N bonds in 2 is higher than
that of the titanium complex due to steric reasons.
amido complex 3 which was also characterized by VT
NMR spectroscopy and a solid-state structure determi-
nation. Compound 3 contains a less sterically encum-
bered germanium metal center relative to 2 and
therefore exhibits a lower barrier to rotation about
the Ge–N bonds than compound 2.
1
The H NMR spectrum of the alkylgermanium com-
plex 3 at room temperature exhibits a sharp resonance at
0.45 ppm corresponding to the 54 protons of the
bis(trimethylsilyl)amido groups. The single butyl group
of 2 results in two triplets at 0.96 (CH3) and 1.70 (Ge–
CH2), as well as a multiplet resulting in the overlap of
signals for the b- and c-methylene groups centered at
4. Experimental
All manipulations were carried out using standard
Schlenk line, glovebox, and syringe techniques [24].
The compounds Ge[N(SiMe3)2]2 [2,3,6], BrN(SiMe3)2
[25], and HOC6H3Ph2-2,6 [26] were prepared according
to the literature procedures. Solvents were purified using
an Innovative Technologies solvent purification system.
1H NMR spectra were recorded at 300 MHz and refer-
1
1.41 ppm. In contrast to the H NMR spectrum of 2,
the –SiMe3 protons of 3 are all equivalent at room tem-
perature and thus give only one signal. This single reso-
nance splits into two resolved features at 0.53 and 0.48
ppm at ꢀ60 ꢁC as shown by variable temperature
NMR spectroscopy. Thus, the barrier to rotation about
the Ge–N bonds in 3 has a value of DGꢀ = 11.0 0.5
kcal/mol, which is less than that for compound 2 and
for the group 4 compounds ClM[N(SiMe3)2]3 (M = Ti,
Zr, Hf) [5]. The shorter Ge–N bonds in the solid-state
structure of 2 indicate that the three amido groups are
more tightly held to the germanium metal center than
in 3, which accounts for the larger barrier to rotation
of these ligands in 2 versus 3.
7
enced to residual protic solvent. Li NMR spectra were
recorded at 194.3 MHz and referenced to a 1.0 M solu-
tion of LiCl in D2O. Elemental analyses were carried out
in-house at Purdue University.
4.1. Synthesis of BrGe[N(SiMe3)2]3 (2)
Compound 2 was prepared by a slight variation of
the published procedure [11]. To a solution of Ge-
[N(SiMe3)2]2 (4.52 g, 11.5 mmol) in hexane (10 mL)
was added a solution of [BrN(SiMe3)2] (3.04 g, 12.6
mmol) in hexane (10 mL). The reaction mixture became
colorless, then dark brown. The solution was stirred
overnight followed by removal of the volatiles in vacuo,
yielding a brown solid. The solid was washed with hex-
ane (3 · 5 mL), yielding a white solid which was recrys-
tallized from hot benzene to give 2 as colorless needles.
3. Conclusions
In summary, we have structurally characterized the
germanium(IV) trisamide BrGe[N(SiMe3)2]3 2 and con-
verted this species to the alkylgermanium compound
[(Me3Si)2N]3Ge(CH2CH2CH2CH3)] 3 by reaction with
1
Yield: 4.17 g (58%). H NMR (C6D6): d 0.59 (s, 27 H),
0.40 (s, 27 H). Anal. Calc. for C18H54BrGeN3Si6: C,
34.12; H, 8.59; N, 6.63; Br, 12.61. Found: C, 33.65; H,
8.39; N, 6.24; Br, 12.39.
n
one equivalent of BuLi. Although the germyllithium
complex LiGe[N(SiMe3)2]3 4 could not be prepared
from 2 by reaction with lithium metal, 4 was accessible
by the reaction of GeI2 with three equivalents of LiN-
4.2. Synthesis of [(Me3Si)2N]3Ge(CH2CH2CH2CH3)
(3)
7
(SiMe3)2. As shown by Li NMR spectroscopy, com-
pound 4 was found to be present as a product in the
synthesis of Ge[N(SiMe3)2]2 from GeCl2 Æ (dioxane)
and LiN(SiMe3)2 when the germanium(II) chloride com-
plex was prepared by the originally reported reduction
of GeCl4 with Bu3SnH [10]. Compound 4 was subse-
A solution of nBuLi (2.5 M in hexanes, 0.80 mL,
2.0 mmol) was added to a suspension of 2 (1.04 g,
1.64 mmol) in benzene (20 mL). The reaction mixture
was refluxed under a N2 atmosphere for 18 h and was
then allowed to come to room temperature. The vola-
tiles were removed in vacuo to yield a white solid. Hex-
ane (2 mL) was added and the mixture was warmed until
all of the material dissolved. Slow cooling of the solution
quently
employed
for
the
preparation
of
[Li(OC6H3Ph2-2,6)3Ge] 5 and explains the formation
of 5 in our attempts to prepare [Ge(OC6H3Ph2-2,6)2]
using Ge[N(SiMe3)2]2 [9].
The previously reported 2 [11] has been further char-
acterized by X-ray crystallography and variable tem-
perature NMR spectroscopy, which serves to
investigate the barrier to rotation about the Ge–N
bonds of 2. Despite the kinetic inertness of 2, this spe-
cies could be converted to the germanium(IV) alkyl
1
yielded 3 as colorless crystals. Yield: 0.24 g (70%). H
NMR (C6D6): d 1.70 (q, 2H, Ge–CH2CH2–), 1.52–1.30
(m, 4H, Ge–CH2CH 2CH2CH3), 0.96 (t, 3H, CH3–),
0.45 (s, 54H, –Si(CH3)3) ppm. Anal. Calc. for
C22H63GeN3Si6: C, 62.87; H, 15.11; N, 6.88. Found:
C, 63.02; H, 15.04; N, 6.94.

J.L. Walding et al. / Inorganica Chimica Acta 358 (2005) 1186–1192
1191
4.3. Synthesis of LiGe[N(SiMe3)2]3 (4)
data [27]. An empirical absorption correction using
SCALEPACK was applied [28]. Intensities of equivalent
reflections were averaged. The structure was solved
using the structure solution program PATTY in DIR-
DIF92 [29]. The remaining atoms were located in suc-
ceeding difference Fourier syntheses. Hydrogen atoms
were included in the refinement but restrained to ride
on the atom to which they are bonded. The structure
was refined in full-matrix least-squares where the func-
tion minimized was Rw(|Fo|2 ꢀ |Fc|2)2 and the weight w
To a suspension of GeI2 (0.750 g, 2.30 mmol) in ether
(10 mL) was added a solution of LiN(SiMe3)2 (1.23 g,
7.35 mmol) in ether (75 mL) via cannula. The reaction
mixture was stirred for 4 h resulting in a yellow solution.
The volatiles were removed in vacuo and the resulting
material was suspended in hexane. The suspension was
filtered through Celite and the residue was washed with
hexane (3 · 5 mL). The hexane was removed in vacuo
and the resulting oil was vacuum distilled (b.p. = 79 ꢁC
at 0.050 torr), yielding a yellow/orange oil. Yield: 0.67
2
is defined as w ¼ 1=fr2ðF 2oÞ þ ð0:0585PÞ þ 1:4064Pg,
where P ¼ ðF 2o þ 2F 2cÞ=3. Scattering factors were taken
from the ‘‘International Tables for Crystallography’’
[30]. Refinement was performed on a AlphaServer
2100 using SHELX97 [31]. Crystallographic drawings
were done using programs ORTEP3 [32]. See http://
www.rsc.org for crystallographic data in CIF or other
electronic format.
1
7
g (52%). H NMR (C6D6): d 0.32 (s) ppm. Li NMR
(C6D6): d 1.05 (s) ppm. Anal. Calc. for C18H54GeLiN3-
Si6: C, 38.56; H, 9.71; N, 7.49. Found: C, 38.23; H,
9.90; N, 7.74.
4.4. Synthesis of [Li(l2-OC6H3Ph2-2,6)3Ge] (5)
To a solution of 4 (0.34 g, 0.61 mmol) in benzene (10
mL) was added a solution of 2,6-diphenylphenol (0.48 g,
1.9 mmol) in benzene (5 mL). The solution was stirred
for 18 h. The volatiles were removed in vacuo and the
resulting solid was recrystallized from 10 mL of ben-
zene/hexane (1:1, v/v) to yield 5 as colorless crystals.
Yield: 0.40 g (82%). Anal. Calc. for C54H39GeLiO3: C,
79.53; H, 4.82. Found: C, 79.82; H, 5.22.
5. Supplementary material
Crystallographic data for the structural analyses have
been deposited with the Cambridge Crystallographic
Data Center, deposition numbers CCDC 218741 (2)
and CCDC 236920 (3). Copies of this information
may be obtained from The Director, CCDC, 12 Union
Road, Cambridge, CB2 1EZ, UK (fax: +44 1233
336033; email: deposit@ccdc.cam.ac.uk or www:
http://www.ccdc.cam.ac.uk).
4.5. X-ray structure determination
Crystal data and data collection parameters are
contained in Table 3. A suitable crystal was mounted
on a glass fiber in a random orientation under a cold
stream of dry nitrogen. Preliminary examination and
final data collection were performed with Mo Ka radi-
Acknowledgements
C.S.W. gratefully acknowledges Professor Ian P.
Rothwell for helpful discussions. This work was funded
by a grant from the National Science Foundation
(Grant No. CHE-0078405).
˚
ation (k = 0.71073 A) on a Nonius KappaCCD. Lor-
entz and polarization corrections were applied to the
Table 3
Crystallographic data for compounds 1 and 2
1
2
References
Formula
Space group
Unit cell dimensions
C18H54BrGeN3Si6
R3c (# 161)
C22H63GeN3Si6
P21/n (#14)
[1] M.F. Lappert, P.P. Power, A.R. Sanger, R.C. Srivastava, Metal
and Metalloid Amides, Ellis Horwood, Chichester, UK, 1980.
[2] D.H. Harris, M.F. Lappert, J. Chem. Soc., Chem. Commun.
(1974) 895.
˚
a (A)
17.8668(4)
17.8668(4)
16.926(1)
90
11.7654(4)
17.9616(6)
16.8366(8)
90
˚
b (A)
˚
c (A)
[3] M.J.S. Gynane, D.H. Harris, M.F. Lappert, P.P. Power, P.
` `
Riviere, M. Riviere-Baudet, J. Chem. Soc., Dalton Trans. (1977)
a (ꢁ)
b (ꢁ)
c (ꢁ)
90
120
103.33(2)
90
2004.
[4] J.J. Ellison, P.P. Power, S.C. Shoner, J. Am. Chem. Soc. 111
(1989) 8044–8046.
3
˚
V (A )
4679.3(3)
6
1.349
3462.1(2)
4
1.172
Z
[5] C. Airoldi, D.C. Bradley, H. Chudzynska, M.B. Hursthouse,
K.M.A. Malik, P.R. Raithby, J. Chem. Soc., Dalton Trans.
(1980) 2010.
qcalc (g cmꢀ3
Temperature (K)
)
150
150
Radiation
Wavelength (A)
Mo Ka
0.71073
0.030
Mo Ka
0.71073
0.043
[6] Q. Zhu, K.L. Ford, E.J. Roskamp, Heteroatm Chem. 3 (1992)
647.
˚
R
[7] C.S. Weinert, A.E. Fenwick, P.E. Fanwick, I.P. Rothwell, J.
Chem. Soc., Dalton Trans. (2003) 532.
Rw
0.065
0.093

1192
J.L. Walding et al. / Inorganica Chimica Acta 358 (2005) 1186–1192
[8] C.S. Weinert, P.E. Fanwick, I.P. Rothwell, J. Chem. Soc., Dalton
Trans. (2002) 2948.
[21] S.N. Nikolaeva, K. Megges, J. Lorbeth, V.S. Petrosyan, Z.
Naturforsch. B 53 (1998) 973.
[9] C.S. Weinert, P.E. Fanwick, I.P. Rothwell, J. Chem. Soc., Dalton
Trans. (2003) 1795.
[22] K.E. Litz, K. Henderson, R.W. Gourley, M.M. Banaszak Holl,
Organometallics 14 (1995) 5008.
[10] S.P. Kolesnikov, I.S. Rogozhin, O.M. Nefedov, Izv. Akad. Nauk
SSSR, Ser. Khim. (1974) 2379.
[23] K.E. Litz, J.W. Kampf, M.M. Banaszak Holl, J. Am. Chem. Soc.
120 (1998) 7484.
[11] M.F. Lappert, M.C. Misra, M. Onyszchuk, R.S. Rowe, P.P.
Power, M.J. Slade, J. Organomet. Chem. 330 (1987) 31.
[12] T. Fjeldberg, A. Haaland, B.E.R. Schilling, M.F. Lappert, A.J.
Thorne, J. Chem. Soc., Dalton Trans. (1986) 1551.
[13] R.W. Chorley, P.B. Hitchcock, M.F. Lappert, W.-P. Leung, P.P.
Power, M.M. Olmstead, Inorg. Chim. Acta 198–200 (1992) 203.
[14] R.D. Rogers, J.L. Atwood, J. Cryst. Spectrosc. Res 13 (1983) 1.
[15] J. Emsley, The Elements, Oxford University Press, Oxford, 1991.
[16] P.J. Brothers, R.J. Wehmschulte, M.M. Olmstead, K. Ruhlandt-
Senge, S.R. Parkin, P.P. Power, Organometallics 13 (1994) 2792.
[17] G.D. Smith, P.E. Fanwick, I.P. Rothwell, Inorg. Chem. 29 (1990)
3221.
[24] D.F. Shriver, M.A. Drezdzon, The Manipulation of Air Sensitive
Comounds, Wiley, New York, 1986.
[25] N. Wiberg, F. Raschig, J. Organomet. Chem. 10 (1967) 15.
[26] D.E. Dana, A.S. Hay, Synthesis (1982) 164.
[27] P.C. McArdle, J. Appl. Cryst. 29 (1996) 306.
[28] Z. Otwinowski, W. Minor, Methods Enzymol. (1996)
276.
[29] P.T. Beurskens, G. Admirall, G. Beurskens, W.P. Bosman, S.
Garcia-Granda, R.O. Gould, J.M.M. Smits, C. Smykalla,
The DIRDIF92 Program System, Technical Report, Crystal-
lography Laboratory, University of Nijmegen, The Nether-
lands, 1992.
[18] G. Ossig, A. Meller, O. Muller, R. Herbst-Irmer, Organometallics
¨
15 (1996) 5060.
[30] International Tables for Crystallography, vol. C, Kluwer Aca-
demic Publishers, Dordrecht, The Netherlands, 1992, Tables
4.2.6.8 and 6.1.1.4.
[19] S.R. Foley, Y. Zhou, G.P.A. Yap, D.S. Richeson, Inorg. Chem.
39 (2000) 924.
[31] G.M. Sheldrick, SHELXS97. A Program for Crystal Structure
Refinement, University of Gottingen, Germany, 1997.
[32] L.J. Farrugia, J. Appl. Cryst. 30 (1997) 565.
[20] G.L. Wegner, A. Jockisch, A. Schier, H. Schmidbaur, Z.
Naturforsch. B 55 (2000) 347.