Synthesis and molecular structures in the gas phase of N,N-dimethylaminoxy-trimethylsilane and -trimethylgermane

Article abstract of DOI:10.1021/ic990689c

The compounds Me₃SiONMe₂ and Me₃GeONMe₂ have been prepared by reacting Me₃SiCl and Me₃GeBr with LiONMe₂. Their identity was proven by gas-phase IR and solution NMR spectroscopy of the nuclei ¹H, ¹³C, ¹⁵N, ¹⁷O, ²⁹Si, and ⁷³Ge and by mass spectrometry and elemental analyses. Their molecular structures in the gas phase have been determined by analysis of electron diffraction data augmented by restraints derived from ab initio calculations. The molecules adopt C_s symmetry. Important geometry parameter values for Me₃SiONMe₂ are: Si-O 1.688(2), Si-C_in-plane 1.870(4), Si-C_out-of-plane 1.872(2), O-N 1.481(6) ?, Si-O-N 107.9(6), O-Si-C_in-plane 102.1(6), O-Si-C_out-of-plane 110.9(7), for Me₃GeONMe₂: Ge-O 1.812(2), Ge-C_in-plane 1.949(3), Ge-C_out-of-plane 1.950(2), O-N 1.475(6) ?, Ge-O-N 108.9(7), O-Ge-C_in-plane 101.7(6), O-Ge-C_out-of-plane 110.3(14). The structure data are interpreted in terms of weak attractive interactions between the nitrogen donor and the silicon/germanium acceptor atoms. The results are discussed in comparison with other structure data from the literature: the strength of a donor-acceptor interaction in Me₃SiONMe₂ is weaker than in H₃SiONMe₂ or ClH₂SiONMe₂, but stronger than in Me₃SiON(CF₃)₂.

Full text of DOI:10.1021/ic990689c

Inorg. Chem. 1999, 38, 5323-5328
5323
Synthesis and Molecular Structures in the Gas Phase of
N,N-Dimethylaminoxy-trimethylsilane and -trimethylgermane
Norbert W. Mitzel,*,† Udo Losehand,† and Alan D. Richardson‡
Anorganisch-chemisches Institut, Technische Universita¨t Mu¨nchen, Lichtenbergstrasse 4,
85747 Garching, Germany, and Department of Chemistry, Oregon State University,
Corvallis, Oregon 97331-4003
ReceiVed June 15, 1999
The compounds Me3SiONMe2 and Me3GeONMe2 have been prepared by reacting Me3SiCl and Me3GeBr with
LiONMe2. Their identity was proven by gas-phase IR and solution NMR spectroscopy of the nuclei 1H, 13C, 15N,
17O, 29Si, and 73Ge and by mass spectrometry and elemental analyses. Their molecular structures in the gas phase
have been determined by analysis of electron diffraction data augmented by restraints derived from ab initio
calculations. The molecules adopt Cs symmetry. Important geometry parameter values for Me3SiONMe2 are:
Si-O 1.688(2), Si-Cin-plane 1.870(4), Si-Cout-of-plane 1.872(2), O-N 1.481(6) Å, Si-O-N 107.9(6),
O-Si-Cin-plane 102.1(6), O-Si-Cout-of-plane 110.9(7)°, for Me3GeONMe2: Ge-O 1.812(2), Ge-Cin-plane 1.949(3),
Ge-Cout-of-plane 1.950(2), O-N 1.475(6) Å, Ge-O-N 108.9(7), O-Ge-Cin-plane 101.7(6), O-Ge-Cout-of-plane
110.3(14)°. The structure data are interpreted in terms of weak attractive interactions between the nitrogen donor
and the silicon/germanium acceptor atoms. The results are discussed in comparison with other structure data
from the literature: the strength of a donor-acceptor interaction in Me3SiONMe2 is weaker than in H3SiONMe2
or ClH2SiONMe2, but stronger than in Me3SiON(CF3)2.
Introduction
SiONMe2,8 which adopts a Si-O-N angle of only 79.7(1)° in
the crystal, but does not undergo even weak intermolecular
The reasons for attractive interactions between geminal
acceptor and donor atoms have been the subject of intensive
studies for various combinations of elements. They have been
postulated for E-C-N linkages (E ) Si, Ge, Sn; e.g., in Me3-
SnCH2NMe2), to rationalize the unusually low basicities of such
amines (also named the R-effect),1 and explained in terms of a
“closed type” three-center bond caused by transitions of
σ-electrons of the N atom participating in the N-C bond into
the empty d-orbitals of the group 14 element.2 Typical examples
for three-membered ring systems with one Lewis-donor-
acceptor bond include the compounds (F3C)2BCPh2NMe2, RNd
C{Al[CH(SiMe3)2]}2 and (Me3Si)2CdN-N{Al[CH(SiMe3)2]}2
with B-C-N,3 Al-CdN,4 and Al-N-N5 linkages forming
three-membered ring systems in the solid state.
interactions. A theoretical treatment of this example showed
negative hyperconjugation to be one suitable model to describe
the attraction as lp(N)fσ*(Si-Cl) interaction.
9
A recent study of the gas-phase structure of Me3SiON(CF3)2
sheds more light on the nature of the Si‚‚‚N attraction. In this
molecule the acceptor ability of the Si atom is reduced by
methylation and the donor ability of the nitrogen center is
reduced because of the two CF3 substituents. Nevertheless, the
Si-O-N angle is still much less than expected on the basis of
Bartell’s one-angle radii10 for Si and N. Only if the N basicity
is even more reduced by silylation, as verified MesF2SiON-
(SiMe3)2,11 the Si‚‚‚N distance (2.692 Å) is consistent with the
sum of Bartell’s one-angle radii (2.69 Å), indicating the absence
of a significant attraction between Si and N atoms.
The most intensely studied systems in this context are SiON6
and SiNN7 linkages. The most intriguing example so far is ClH2-
Determination of the molecular structure of Me3SiONMe2
seemed desirable to prove the consistency of our present
knowledge on those systems. Furthermore, we wanted to extend
our studies to the heavier elements of group 14 acceptors and
report here about the synthesis and gas-phase molecular structure
of Me3GeONMe2. A structural investigation of Me3SiONMe2
should also improve our knowledge on trialkylsilicon systems
with hydoxylamine and oxime groups, which are used as cold-
curing catalysts in silicone polymer production.12
* Author to whom correspondence should be addressed. E-mail:
N.Mitzel@lrz.tum.de.
† Technische Universita¨t Mu¨nchen.
‡ Oregon State University.
(1) (a) Voronkov, M. G.; Feshin, V. P.; Mironov, V. F.; Mikhailyants, S.
A.; Gar, T. K.; Zh. Obshch. Khim. 1971, 41, 2211. (b) Voronkov, M.
G.; Kashik, T. V.; Lukevits, E. Y.; Deriglazova, E. S.; Pestunovich,
A. E.; Moskovich, R. Y. Zh. Obshch. Khim. 1974, 44, 2211.
(2) Rivie`re, P.; Rivie`re-Baudet, M.; Satge´, J. In ComprehenshiVe Orga-
nometallic Chemistry; Wilkinson, G. W., Stone, F. G. A., Abel, E.
W., Eds.; Pergamon Press: Oxford, 1982.
(3) Brauer, D. J.; Bu¨rger, H.; Buchheim-Spiegel, S.; Pawelke, G. Eur. J.
Inorg. Chem. 1999, 255.
(4) Uhl, W.; Schu¨tz, U.; Hiller, W.; Heckel, M. Chem. Ber. 1994, 127,
1587.
(5) Uhl, W.; Hannemann, F. Eur. J. Inorg. Chem. 1999, 201.
(6) (a) Mitzel N. W.; Losehand, U. Angew. Chem., Int. Ed. Engl. 1997,
36, 2807. (b) Losehand, U.; Mitzel, N. W. Inorg. Chem. 1998, 37,
3175.
(8) Mitzel, N. W.; Losehand, U. J. Am. Chem. Soc. 1914, 36, 2807.
(9) Hertel, T.; Jakob, J.; Minkwitz, R.; Oberhammer, H. Inorg. Chem.
1998, 37, 5092.
(10) (a) Bartell, L. S. J. Chem. Phys. 1960, 32, 827. (b) Glidewell, C. Inorg.
Chim. Acta 1975, 12, 219.
(11) Wolfgramm, R.; Klingebiel, U. Z. Anorg. Allg. Chem. 1998, 624, 859.
(12) Voronkov, M. G.; Maletina, E. A.; Roman, V. K. Heterosiloxanes,
Vol. 2, DeriVatiVes of Nitrogen and Phosphorus; Harwood Academic
Publishers GmbH; Chur, Switzerland, 1991.
(7) Mitzel, N. W. Chem. Eur. J. 1998, 4, 692.
10.1021/ic990689c CCC: $18.00 © 1999 American Chemical Society
Published on Web 10/22/1999

5324 Inorganic Chemistry, Vol. 38, No. 23, 1999
Mitzel et al.
Table 1. Geometry Parameters, Their Values (in ra and rR), and Parameter Restraints for the GED Refinements of Me3SiONMe2 and
a
Me3GeONMe2
Me3SiONMe2
Me3GeONMe2
value
value
restraint
definition
no.
param
ra
rR
restraint
MP2/6-31G*
ra
rR
restraint
MP2/6-31G*
p1 E-O
1.688(2)
1.872(2)
1.870(4)
1.481(6)
1.452(4)
1.108(4)
1.117(2)
1.684(1)
1.863(2)
1.860(4)
1.461(6)
1.438(4)
1.057(3)
1.067(2)
1.707
1.877
1.878
1.471
1.460
1.090
1.095
111.0
103.1
106.0
104.6
111.0
110.0
111.2
111.0
108.8
-177.0
-176.9
1.812(2)
1.950(2)
1.949(3)
1.475(6)
1.460(4)
1.107(3)
1.111(3)
110.3(14)
101.7(6)
108.9(7)
105.4(4)
111.2(10)
107.1(42)
111.5(6)
109.6(7)
107.7(8)
179.5(44)
1.808(2)
1.940(2)
1.938(3)
1.459(6)
1.442(4)
1.050(3)
1.053(3)
110.0(12)
101.9(7)
109.8(7)
106.1(5)
110.0(11)
107.5(33)
112.2(6)
110.3(7)
108.4(8)
177.4(46)
1.833
1.938
1.937
1.473
1.459
1.091
1.094
109.9
102.7
102.9
104.3
111.0
109.1
111.8
109.9
108.6
178.3
-176.6
p2 E-C7
p3 E-C6
p4 O-N
p5 N-C
p3 - p2
p5 - p4
p7 - p6
0.001(5)
-0.011(10)
0.005(5)
-0.001(5)
-0.014(10)
0.003(5)
p6 C-H(N)
p7 C-H(E)
p8
p9
OEC(7)
OEC(6)
E-O-N
O-N-C
C-N-C
C-E-C
110.9(7)
110.7(7)
102.1(6)
107.9(6)
104.8(4)
112.7(12)
108.1(23)
111.3(10)
110.6(14)
106.1(6)
103.4(6)
108.9(5)
105.0(4)
110.3(10)
107.8(23)
112.4(9)
112.3(13)
106.5(5)
p10
p11
p12
p13
p14
p15
p16
ECH(7) p14
111.2(10)
-0.2(10)
108.8(10)
111.8(10)
-1.9(10)
108.6(10)
178.3(50)
ECH(6) p15 - p14
NCH
p16
p17
p18
p17 τCECH
p18 τCNCH
-169.1(34) -173.9(39) -177.0(50)
-160.3(31) -168.8(30) -176.9(50)
-181.0(29) -181.1(29) -176.6(50)
a The geometrical parameter values of the ab initio calculations (MP2/6-31G*) are listed for comparison. Distances are given in angstroms,
angles and torsion angles in degrees.
Results and Discussion
bonding partners of oxygen is different (Si and Ge), we cannot
provide a more detailed explanation for this fact.
Synthesis and Characterization of Me3SiONMe2 and
Me3GeONMe2. Me3SiONMe2 and Me3GeONMe2 were pre-
pared from lithiated N,N-dimethylhydroxylamine and trimeth-
ylchlorosilane and trimethylbromogermane, respectively. They
were isolated as colorless liquids by fractionation through a
series of cold traps in vacuo. Their identity was proven by their
gas-phase IR and solution (C6D6) NMR spectra (1H, 13C, 15N,
17O, 29Si, 73Ge) and by mass spectrometry and elemental
analyses. Me3SiONMe2 was previously prepared by West and
Boudjouk by a different method.13
Molecular Structures of Me3SiONMe2 and Me3GeONMe2.
All attempts to grow single crystalline material for X-ray
diffraction experiments from samples of Me3SiONMe2 and Me3-
GeONMe2 were unsuccessful, because the compounds solidify
as glass upon cooling. However, the high volatility allowed us
to determine their molecular structures in the gas phase by means
of electron diffraction. The limitations of structural analyses
on electron diffraction data can be overcome to some extent by
augmenting the data by restraints taken from ab initio calcula-
tions of the molecular geometries in an extended method called
SARACEN,16 a procedure combining and extending Bartell’s
approach of predicate values17 and Scha¨fer’s MOCED method.18
The definition of the models for both compounds was identical
and is described in the Experimental Section. The ab initio
geometries were calculated at the MP2/6-31G* level of theory,
under imposed Cs symmetry. The nature of the stationary points
was tested by frequency calculations at the same level. This
yielded a force field used for the calculation of vibrational
amplitudes (after scaling the force constants by a common factor
of 0.93), which were included as further restraints in the GED
(Gas-Phase Electron Diffraction) refinements (see Experimental
Section for details). Perpendicular amplitude corrections were
calculated from the same force field to allow for a refinement
of the structures in rR space.
The 29Si NMR chemical shift of Me3SiONMe2 (-17.18 ppm)
is shifted to lower frequencies relative to the isoelectronic Me3-
SiOCHMe2 (12.1 ppm).14 This is compatible with a shielding
by increased electron density at the silicon center, as could be
expected if the Me2NO substituent donated electrons via its
NMe2 group.
In a 73Ge NMR experiment Me3GeONMe2 gave a signal at
522 ppm vs GeMe4, which is noteworthy, because we have
never observed 73Ge NMR signals for other aminoxygermanes
even after prolonged periods and varied conditions of NMR
experiments, not even for the symmetrically substituted Ge-
(ONMe2)4 or Ge(ONEt2)4. So far, because of the lack of data,
a comparison with other 73Ge NMR chemical shifts has been
impossible.
More telling are the 15N NMR resonances of Me3SiONMe2
and Me3GeONMe2 at -247.8 and -250.6 ppm, which are
shifted more than 16 ppm to lower frequencies, i.e. shielded
with respect to the simplest aminoxysilane H3SiONMe2 (-234.0
ppm),6 but slightly to higher frequencies, i.e. deshielded,
compared with HONMe2 (-259.4 ppm).15 These data are
consistent with a less pronounced interaction between the N
and the Si or Ge atoms in Me3SiONMe2 and Me3GeONMe2
than in H3SiONMe2, which should result in wider Si-O-N
and Ge-O-N angles. However, care is suggested with such
interpretations, because chemical shifts of many similar and
related compounds fall over a range of 10 ppm. The similarity
between the chemical shifts of Me3SiONMe2 and Me3GeONMe2
also indicates the similar electronic nature of the nitrogen atoms.
At -13 and 146 ppm the 17O chemical shifts of Me3SiONMe2
and Me3GeONMe2 are markedly different. Although one of the
The parameter values and the applied restraints for both
refinements, each in ra and rR space, and for the ab initio values
are listed in Table 1. Several low calculated frequencies (for a
list see the Supporting Information) led to an overestimation of
the corrections for some bond lengths in the rR refinements (e.g.,
(13) West, R.; Boudjouk, P. J. Am. Chem. Soc. 1973, 95, 3987.
(14) Schraml, J.; Pola, J.; Jacke, H.; Engelhardt, G.; Cerny, M. Collect.
Czech. Chem. Commun. 1976, 41, 360.
(15) Mitzel, N. W.; Schmidbaur, H. Z. Anorg. Allg. Chem. 1994, 620, 1087.
(16) Blake, A. J.; Brain, P. T.; McNab, H.; Miller, J.; Morrison, C. A.;
Parsons, S.; Rankin, D. W. H.; Robertson, H. E.; Smart, B. A. J. Phys.
Chem. 1996, 100, 12280.
(17) Bartell, L. S.; Romenensko D. J.; Wong, T. C. In Molecular Structure
by Diffraction Methods; Specialist Periodical Reports, The Chemical
Society, 1975; Vol. 3, p 72.
(18) Klimkowski, V. J.; Ewbank, J. D.; v. Alsenoy, C.; Scha¨fer, L. J. Am.
Chem. Soc. 1982, 104, 1476.

Molecular Structures of Me3SiONMe2 and Me3GeONMe2
Inorganic Chemistry, Vol. 38, No. 23, 1999 5325
Figure 1. Molecular scattering intensity and final difference curves
(vs model) as obtained by electron diffraction of gaseous Me3SiONMe2.
Figure 3. Radial distribution and difference curve for the electron
diffraction refinement of Me3SiONMe2. Vertical lines indicate atom
pairs with their height being proportional to their scattering contribution.
Figure 2. Molecular scattering intensity and final difference curves
(vs model) as obtained by electron diffraction of gaseous Me3GeONMe2.
the N-C distances) if established standard values for such
lengths were considered, which led us to prefer the uncorrected,
but in this way more realistic ra values. However, the structural
information from both types of refinement is very similar, and
the most important geometry parameters, the E-O-N angles,
are the same within the standard deviations. This is why the ra
values are used for comparison and structure discussion
throughout the rest of this paper.
Figure 4. Radial distribution and difference curve for the electron
diffraction refinement of Me3GeONMe2. Vertical lines indicate atom
pairs with their height being proportional to their scattering contribution.
the Ge-O distance [1.812(2) Å], which is longer than that of
Me3GeOGeMe3 [1.770(10) Å]21 and the Ge-C bond lengths
[1.949(3) and 1.950(2) Å], which are almost identical with those
of GeMe4 [1.945(3) Å].22
Although eight restraints had been used, most of the important
parameters defining the geometry of the skeleton were freely
refined. The satisfactory agreement between theory and experi-
ment in these parameters justified the application of supporting
restraints derived from calculations for differences between
similar parameters, parameters describing hydrogen positions
and vibrational amplitudes. The experimental structures pre-
sented correspond to ra geometries. The success of the refine-
ments of the molecular geometries of Me3SiONMe2 and
Me3GeONMe2 may be assessed from the small residuals in the
electron-scattering intensity curves and radial distribution curves,
which are presented in Figures 1-4. The final R factors were
0.0314 and 0.0581 for the refinements on Me3SiONMe2 and
Me3GeONMe2.
At 1.688(2) Å the Si-O bond length in Me3SiONMe2 is
longer than in Me3SiOCH3 [1.639(4) Å],19 which is the best
comparable siloxane studied in the gas phase. It is also longer
than in ClH2SiONMe2 [1.641(3) Å in the gauche conformer],
but is still shorter than the ab initio estimate of 1.707 Å. The
Si-C distances in Me3SiONMe2 are similar to those in SiMe4
[1.875(2) Å].20 Similar comparisons for Me3GeONMe2 include
The N-O distances in both compounds, Me3SiONMe2
[1.481(6) Å] and Me3GeONMe2 [1.475(6) Å], are longer than
in the nonsilylated HONMe2 [1.448(11) Å].23 The angles about
the nitrogen centers are very close to that in HONMe2, as is
indicated by their sum of 322.3° (Me3SiONMe2) and 322.0°
(Me3GeONMe2) vs 323.8° in HONMe2.
The most interesting part in these structures is the geometry
of the SiON and GeON skeletons. The occurrence of an
attractive force between the nitrogen donor center and the
acceptor atoms Si and Ge should lead to a shortening of the
Si‚‚‚N and Ge‚‚‚N distances, which is better described by a
contraction of the angles Si-O-N and Ge-O-N, which are
107.9(6)° and 108.9(7)°. On the one hand, this is substantially
smaller than typical Si-O-C and Ge-O-C angles [compare
Me3SiOCH3 Si-O-C 122.5(6)°]19 or the even wider angles
Si-O-Si and Ge-O-Ge in Me3SiOSiMe3 [148(3)°]24 and Me3-
GeOGeMe3 [141(1)°].25 On the other hand, the Si-O-N and
(21) Vilkov, L. V. Kem. Kozl. 1971, 35, 375.
(22) Hender, J. L.; Mustoe, F. J. Can. J. Chem. 1975, 53, 3542.
(23) Mitzel, N. W.; Smart, B. A.; Parsons, S.; Robertson, H. E.; Rankin,
D. W. H. J. Chem. Soc., Perkin Trans. 2 1996, 2727.
(19) Csa´kva´ri, B.; Wagner, Z.; Go¨mo¨ry, P.; Hargittai, I.; Rozsondai, B.;
Mijlhoff, F. C. Acta Chim. Acad. Sci. Hung. 1976, 90, 149.
(20) Mastryukov, V. In Stereochemical Applications of Gas-Phase Electron
Diffraction; Hargittai, I., Hargittai. M., Eds.; VCH Publishers: New
York, 1988.
(24) (a) Csa´kva´ri, B.; Wagner, Z.; Go¨mo¨ry, P.; Rozsondai, B.; Mijlhoff,
F. C.; Hargittai, I. J. Organomet. Chem. 1976, 107, 287. (b) Borisenko,
K. B.; Rozsondai, B.; Hargittai, I. J. Mol. Struct. 1997, 406, 137.
(25) Vilkov, L. V.; Tarasenko, N. A. Zh. Strukt. Khim./J. Struct. Chem.
(Engl. Transl.) 1969, 10, 1102/979.

5326 Inorganic Chemistry, Vol. 38, No. 23, 1999
Mitzel et al.
Table 2. Important Geometry Parameter Values (Å/deg) for
Ge-O-N angles are much wider than in ClH2SiONMe2 [anti
conformer 79.7(1)° in the crystal, 87.1(9)° in the gas phase]8
and in H3SiONMe2 [102.6(1)° in the crystal],6 but is smaller
than in Me3SiON(CF3)2 [113.4(19)° in the gas phase].9 This
can be interpreted as a weak attractive interaction (â-donor-
acceptor interaction) between Si and N atoms. It is weaker than
in H3SiONMe2 because of the reduced ability of a SiMe3 group
to act as an acceptor compared with a SiH3 group, but stronger
than in Me3SiON(CF3)2, because an N(CF3)2 group is a weaker
donor than a Me2N group.
A comparison of the E-O-N angles of Me3SiONMe2 and
Me3GeONMe2 with that of the homologous compound Me3-
SnONMe2 [102.7(8)°]26 documents the better acceptor ability
of tin than its lighter homologues, which also allows this
compound to form additional intermolecular Sn‚‚‚O contacts
in the solid state.
Resulting from the weak E‚‚‚N interactions, the geometries
of the Me3EO groups are distorted, i.e., a slight widening of
the two symmetry equivalent out-of-plane angles O-E-C to
110.9(7)° and 110.3(14)° and a compression of the angles
O-E-C in the plane of symmetry to 102.1(6)° and 101.7(6)°
for Me3SiONMe2 and Me3GeONMe2. This deformation is
slightly less than in the tin compound Me3SnONMe2, which
has an in-plane angle of 99.6(10)°.26
The similarity between the Si-O-N and Ge-O-N angles
and the distortions of the Me3EO group in Me3SiONMe2 and
Me3GeONMe2 substantiates the electronic similarity of Me3Si
and Me3Ge groups in their behavior as acceptors. By comparing
the experimental data with the ab initio predictions it becomes
obvious how inadequately the MP2/6-31G* level describes the
situation. Although the geometry of Me3SiONMe2 is generally
well reproduced by the calculations at this level, the predicted
Ge-O-N angle is substantially less [102.9°] than the experi-
mentally observed one [108.9(7)°]. Our earlier work documented
that the strength of â-donor interactions is slightly overestimated
at MP2/6-31G* and a better description is achieved with larger
basis sets (in particular MP2/6-311G**). Because of the problem
size our present computing resources did not allow us to improve
on the applied degree of theoretical sophistication. Apart from
this GeON fragment the geometry of Me3GeONMe2 is repro-
duced satisfactorily by the calculations.
Me3SiONMe2, Me3GeONMe2, and Reference Compounds in
Comparisona
sum of
compound/method
E-O-N
E‚‚‚N
Bartell’s radii
Me3SiONMe2/GED
Me3GeONMe2/GED
Me3SnONMe2/GED
Me3SiON(CF3)2/GED
MesF2SiON(SiMe3)2/XRD
ClH2SiONMe2 (anti)/XRD
ClH2SiONMe2 (anti)/GED
ClH2SiONMe2 (gauche)/GED 104.7(11) 2.468(25)
H3SiONMe2/XRD
107.9(6)
108.9(7)
102.5(8)
113.4(19) 2.66
117.66(11) 2.692
79.7(1)
87.1(9)
2.566(8)
2.682(11)
2.731(14)
2.69
2.72
3.02
2.69
2.69
2.69
2.69
2.69
2.69
2.69
2.028(1)
2.160(7)
102.6(1)
95.2(av) 2.318(av)
2.453(1)
H2Si(ONMe2)2/XRD
a av ) averaged value, GED values all ra.
Å], which is consistent with a less electrophilic Me3Si group
compared with H3Si or ClH2Si units.
The experimental Ge‚‚‚N distance in Me3GeONMe2 is only
1.4% shorter than the sum of Bartell’s radii of Ge and N (4.6%
in Me3SiONMe2). On this basis we can conclude only a weak
attraction between Ge and N atoms if any. The ab initio
predicted Ge‚‚‚N distance is 2.595 Å, which is 4.6% shorter
than the sum of Bartell’s radii, the same value as for the silicon
analogue. In the tin analogue the Sn‚‚‚N distance is 9.6% shorter
than the sum of the one-angle radii, which means that
germanium seems to have exceptionally low electrophilicity in
this series of compounds, a fact that is not reproduced
satisfactorily by theory at the MP2/6-31G* level.
We used our calculated geometries at the MP2/6-31G* level
of theory to perform a natural bond orbital analysis (NBO),
which provided us with a coarse description of electron
delocalization by second-order perturbation theory analysis of
the Fock matrix in the NBO basis. In both compounds, Me3-
SiONMe2 and Me3GeONMe2, the largest contributions for
orbital interactions through electron delocalization stem from
electron density in the lone pair of electrons at nitrogen and
oxygen. There are interactions between the oxygen lone pairs
and the σ*(E-C) orbitals belonging to the methyl groups in
gauche position relative to the E-O-N plane. Such interactions
are established contributions in E-O compounds and this type
of negative hyperconjugation has been claimed to be responsible
for the low basicity of silylated oxygen atoms and the wide
bond angles of silylated oxygen. Concerning the interaction
between the nitrogen donor and silicon or germanium acceptor
atoms, the NBO calculations reveal a lp(N)fσ*(E-C) interac-
tion with the σ*(E-C) orbital of the methyl group in the
E-O-N plane. The NBO analysis provides us with stabilization
energies gained through electron delocalization, which are
neither physically observable nor very meaningful in their
absolute values, but give an estimate of the relative strength of
such interactions. The lp(N)fσ*(E-C) interaction energy in
Me3SiONMe2 is about as large as in H3SiONMe2 [lp(N)f σ*-
(Si-H), SiON 102.6(1)° in the crystal], but only 32% of the
lp(N)fσ*(Si-Cl) interaction in the anti conformer of ClH2-
SiONMe2, which results in a much smaller SiON angle in the
latter [87.1(9)° in the gas phase]. According to the ab initio
prediction that the GeON angle in Me3GeONMe2 is smaller than
the SiON angle in Me3SiONMe2, the lp(N)fσ*(Ge-C) interac-
tion in Me3GeONMe2 has been predicted to be stronger by 40%
than the lp(N)fσ*(Si-C) interaction in Me3SiONMe2.
The E‚‚‚N donor-acceptor interactions can thus be described
as weak type of remote negative hyperconjugation, although
other alternative interpretations are possible. In particular the
Coulomb attraction between electropositive silicon and germa-
Oberhammer et al. have already pointed out the advantage
of assessing the strength of a possible E‚‚‚N interaction between
geminal atoms by comparing the E-N distance with Bartell’s
one-angle radii,9 describing the repulsion between the spheres
of two atoms A and B bound to the same center Y. The angle
A-Y-B can be calculated with known distances A-Y and
B-Y. A selection of compounds containing E-O-N fragments
with their angles E-O-N, the distance E‚‚‚N, and the sum of
the corresponding Bartell’s radii are listed in Table 2 for
comparison.
None of the Si‚‚‚N distances of the SiON compounds in Table
2 significantly exceeds the sum of Bartell’s radii at 2.69 Å.
Klingebiel’s compound MesF2SiON(SiMe3)2 has the widest
angle Si-O-N [117.66(11)°] and a distance Si‚‚‚N of 2.692
Å. This indicates the absence of a significant attraction between
N and Si atoms, probably because of the low basicity of the
silylated nitrogen atom and despite the two electronegative F
substituents at silicon. At 2.566(8) Å the Si‚‚‚N distance of Me3-
SiONMe2 indicates the presence of an attraction weaker than
in H3SiONMe2 [2.453(1) Å] or trans-ClH2SiONMe2 [2.160(7)
(26) Mitzel, N. W.; Losehand, U.; Richardson, A. Organometallics 1999,
18, 2610.

Molecular Structures of Me3SiONMe2 and Me3GeONMe2
Inorganic Chemistry, Vol. 38, No. 23, 1999 5327
Table 3. Experimental Conditions (Camera Distances [mm], Electron Wavelengths [Å], Nozzle and Sample Temperatures [°C]), Data Ranges
and Weighting Functions [Å-1], Correlation Parameters, Scale Factors, and Final R Factors for the GED Experiments and Refinements of the
Compounds Me3SiONMe2 and Me3GeONMe2
T
correlation
compd/data set
camera dist wavelength nozz samp ∆s smin
s1
4.0 14.0 16.4
0.4 8.0 10.0 30.4 35.6
0.2 2.0 4.0 14.0 16.4
0.4 8.0 10.0 30.8 36.0
s2
smax
param
scale factor
R1
Rg
Me3SiONMe2/1
Me3SiONMe2/2
Me3GeONMe2/1
Me3GeONMe2/2
746.37
300.84
746.37
299.76
0.04894
0.04894
0.04894
0.04894
25
24
25
23
13
13
4
0.2 2.0
-0.0211
0.1485
0.4779
0.1220
0.958(4)
0.880(10)
1.007(8)
0.937(14)
0.0269 0.0314
0.0446
0.0712 0.0581
0.0458
17
°C), and (N,N-dimethylhydroxylamino)trimethylsilane (2.6 g, 19.5
mmol, 69%) was isolated as a colorless liquid by fractionation through
a series of cooled traps (-20, -78, -96, -196 °C) with the product
nium and electronegative nitrogen centers contributes a great
deal to this kind of secondary bonding.
1
retained in the -78 °C trap. H NMR δ ) 0.2 (s, 9H, H3C), 2.38 (s,
Conclusion
1
6H, H3C). 13C NMR δ ) -1.1 (q, JCH ) 118.3 Hz (CH3)3Si), 50.4
With the determination of the gas-phase structures of Me3-
SiONMe2 and Me3GeONMe2 we have established structural data
of a series of compounds, which allow us to assess the acceptor
ability of Me3E groups (E ) Si, Ge, Sn) toward donor atoms
in the â-position. Tin is clearly the best acceptor, followed by
silicon, whereas almost no interaction is detectable for germa-
nium. The high ability of oxygen centers to adopt a wide range
of different angles if bound to silicon substituents is documented
once more. As expected Me3SiONMe2 shows a â-donor-
acceptor interaction, which is intermediate in strength between
H3SiONMe2 (higher electrophilicity at Si) and Me3SiON(CF3)2
(lower nucleophilicity at N). The NBO analyses show that the
interaction of the lone pair at nitrogen with the E substituent
(E ) Si or Ge) in the E-O-N plane is the most important
contribution of electron delocalization concerning the secondary
bonding in the E-O-N unit. The strength of the interaction
increases with the nature of this E substituent in the series Me
< H , Cl < F.
In an earlier contribution we discussed a mechanism for the
catalytic action of hydroxylamines on the alcoholysis of Si-H
functions involving â-donor-acceptor interactions between Si
and N atoms in Si-O-N systems,27 leading to partially
hypercoordinate silicon centers and thus accelerating SN2
reactions. This postulate was based on structure determinations
of Si(ONR2)4 and Si-hydrogenated compounds. With the present
example we provide evidence that such interactions may also
be important even for trialkylsilylhydroxylamines or for reac-
tions involving highly alkylated silicon groups, which are
systems more similar to those actually applied in praxis.
(qq, 1JCH ) 135.4 Hz, 3JCNCH ) 6.0 Hz, (CH3)2N), 15N{1H} NMR δ )
-247.8 (s), 17O{1H} NMR δ ) -13.1 (s), 29Si NMR δ ) -17.18
(dec, 2JSiCH ) 6.5 Hz). IR (gas): 2967 cm-1 s (νCH3). MS(Cl): m/z )
133.0 (M+), 118.0 (M+ - CH3), 75.0 (M+ - ONME2). Anal. H15C5-
NOSi (133.27 g/mol). Calcd: C, 45.1; H, 11.3; N, 10.5. Found: C,
44.9; H, 11.3; N, 10.7.
(N,N-Dimethylhydroxylamino)trimethylgermane. At -50 °C, 0.9
g of n-butyllithium (14 mmol, 1.6 M in hexane) was added dropwise
to a solution of 1.0 mL of N,N-dimethylhydroxylamine (0.9 g, 14 mmol)
in 20 mL of pentane. The mixture was stirred for 1 h at ambient
temperature, and solvents were removed in vacuo. At -196 °C ca. 20
mL of dimethyl ether and 1.0 g of trimethylchlorogermane (6.5 mmol)
were condensed onto the remaining salt. The mixture was stirred at
-96 °C for 1 h, then slowly warmed to -30 °C and stirred for another
1.5 h. All volatile products were condensed into a trap (-196 °C) and
(N,N-dimethylhydroxylamino)trimethylgermane (1.0 g, 5.9 mmol, 90%)
was isolated as a colorless liquid by fractionation through a series of
cooled traps (-20, -78, -96, -196 °C) with the product retained in
1
the -78 °C trap. H NMR δ ) 0.32 (s, 9H, H3C), 2.47 (s, 6H, H3C).
1
1
13C NMR δ ) -0.44 (q, JCH ) 125.9 Hz, CH3), 50.8 (q q, JCH
)
134.0 Hz, JCNCH ) 5.8 Hz, CH3). 15N{1H} NMR δ ) -250.6 (s).
17O{1H} NMR δ ) 146.4 (s). 73Ge{1H} NMR δ ) 522 (s). IR (gas):
2992 cm-1 s (νCH3). MS(Cl): m/z ) 178.7 (M+ - 1, 74Ge), 118.7
(M+ - 1 - ONMe2, 74Ge). Anal. H15C5GeNO (177.79 g/mol). Calcd:
C, 33.8; H, 8.40; N, 7.90. Found: C, 34.0; H, 8.49; N, 7.58.
3
Electron Diffraction Experiments. GED Data. Electron-scattering
intensity data for Me3SiONMe2 and Me3GeONMe2 were recorded on
Kodak Electron Image film using the Oregon State University diffrac-
tion apparatus operating at 60 kV acceleration voltage. Two data sets
at different camera distances were recorded for each compound from
three exposures. Diffraction patterns of CO2 were recorded concurrently
for wavelength calibration. Further experimental conditions and general
parameters concerning the refinements are listed in Table 3. The least-
squares refinements were carried out using the program ED9628 with
the scattering factors established by Fink and co-workers.29 The refined
molecular parameters, their definition, and the applied restraints, a list
with selected interatomic distances including vibrational amplitudes and
applied restraints, are listed in Table 4.
Experimental Section
General. The experiments were carried out using a standard Schlenk
line or a vacuum line with greaseless PTFE stopcocks, which is attached
directly to the gas cell in an FTIR spectrometer (Midac Prospect FTIR).
All NMR spectra were recorded at 21 °C on a JEOL JNM-LA400
spectrometer in sealed tubes with C6D6 as a solvent directly condensed
onto the sample from K/Na alloy.
(N,N-Dimethylhydroxylamino)trimethylsilane. At -50 °C, 1.8 g
of n-butyllithium (28 mmol, 1.6 M in hexane) was added dropwise to
a solution of 2.0 mL of N,N-dimethylhydroxylamine (1.7 g, 28 mmol)
in 25 mL of pentane. The mixture was stirred for 1 h at ambient
temperature, and solvents were removed in vacuo. At -196 °C ca. 25
mL of dimethyl ether and 3.8 mL of trimethylchlorosilane (3.1 g, 28
mmol) were condensed onto the remaining salt. The mixture was stirred
at -96 °C for 1 h, then slowly warmed to -30 °C, and stirred for
another 1.5 h. All volatile products were condensed into a trap (-196
GED Model. The geometrical models for Me3SiONMe2 and Me3-
GeONMe2 were defined in Cs symmetry. The atom numbering scheme
is provided in Figures 5 and 6. While fixing the differences between
the parameters defining hydrogen atom positions (C-H distances,
angles, and torsion angles) to calculated values and treating the silicon/
(29) Ross, A. W.; Fink, M.; Hilderbrandt, R. International Tables for X-ray
Crystallography; Wilson, A. J. C., Ed.; Kluwer Academic Publish-
ers: Dordrecht, Boston, 1992; Vol. C, p 245.
(30) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.; Johnson,
B. G.; Robb, M. A.; Cheeseman, J. R.; Keith, T.; Petersson, G. A.;
Montgomery, J. A.; Raghavachari, K.; Al-Laham, M. A.; Zakrzewski,
V. G.; Ortiz, J. V.; Foresman, J. B.; Cioslowski, J.; Stefanov, B. B.;
Nanayakkara, A.; Challacombe, M.; Peng, C. Y.; Ayala, P. Y.; Chen,
W.; Wong, M. W.; Andres, J. L.; Replogle, E. S.; Gomperts, R.;
Martin, R. L.; Fox, D. J.; Binkley, J. S.; Defrees, D. J.; Baker, J.;
Stewart, J. P.; Head-Gordon, M.; Gonzalez, C.; Pople J. A. Gaussian
94, Revision C.2; Gaussian, Inc.: Pittsburgh, PA, 1995.
(27) Mitzel, N. W.; Blake, A. J.; Rankin, D. W. H. J. Am. Chem. Soc.
1997, 119, 4143.
(28) Mitzel, N. W.; Brain, P. T.; Rankin, D. W. H. ED96, Version 2.0,
1998. A program developed on the basis of formerly described ED
programs: Boyd, A. S. F.; Laurenson, G. S.; Rankin, D. W. H. J.
Mol. Struct. 1981, 71, 217.

5328 Inorganic Chemistry, Vol. 38, No. 23, 1999
Mitzel et al.
restraint
a
Table 4. Distances, Amplitudes, and Restraints for the GED Refinements of the Compounds Me3SiONMe2 and Me3GeONMe2
Me3SiONMe2
amplitude
Me3GeONMe2
amplitude
no.
atom pair
dist
restraint
dist
d1
d2
d3
d4
d5
d6
d7
d8
E1-O2
E1-C6
1.688(2)
1.870(4)
1.872(2)
1.481(6)
1.452(4)
1.108(4)
2.566(8)
2.769(12)
2.935(11)
2.324(6)
3.110(13)
3.030(45)
2.418(18)
3.531(6)
4.112(10)
3.240(14)
4.888(11)
4.569(12)
3.681(21)
2.493(19)
2.056(7)
2.005(7)
2.071(7)
3.495(26)
4.426(10)
3.731(30)
0.048(1)
0.049(4)
0.047(2)
0.059(5)
0.047(3)
0.074(1)
0.077(6)
0.094(16)
0.085(13)
0.084(6)
0.075(12)
0.076(16)
0.069(12)
0.115(4)
0.090(9)
0.106(13)
0.172(11)
0.225(21)
0.326(39)
0.137(22)
0.067(18)
0.067(20)
0.071(18)
0.248(46)
0.115(13)
0.182(32)
1.812(2)
1.949(3)
1.950(2)
1.475(6)
1.460(4)
1.107(3)
2.682(11)
2.918(13)
3.088(26)
2.335(6)
3.264(16)
3.136(85)
2.410(16)
3.664(6)
4.258(13)
3.386(22)
5.050(13)
4.725(15)
3.843(36)
2.546(8)
2.083(11)
2.029(11)
2.084(11)
3.495(22)
4.485(17)
4.104(30)
0.047(3)
0.048(6)
0.054(3)
0.047(3)
0.047(2)
0.075(2)
0.090(7)
0.078(12)
0.076(11)
0.068(6)
0.075(11)
0.092(20)
0.066(12)
0.122(4)
0.100(14)
0.184(25)
0.180(20)
0.182(25)
0.268(40)
0.099(9)
0.091(17)
0.091
1.18(12)a
1.00(10)b
1.18(12)a
1.00(10)b
E1-C7
O2-N3
N3-C4
C4-H9
E1‚‚‚N3
O2‚‚‚C6
O2‚‚‚C7
O2‚‚‚C4
C6‚‚‚C7
C7‚‚‚C8
C4‚‚‚C5
E1‚‚‚C4
N3‚‚‚C6
N3‚‚‚C7
C6‚‚‚C4
C7‚‚‚C4
C7‚‚‚C5
E1‚‚‚H15
N3‚‚‚H9
N3‚‚‚H10
N3‚‚‚H11
E1‚‚‚H9
E1‚‚‚H10
E1‚‚‚H11
1.03(10)c
1.03(10)c
0.082(16)
0.093(19)
0.088(18)
0.064(13)
0.102(20)
0.101(20)
0.067(13)
0.112(22)
0.089(18)
0.172(34)
0.149(30)
0.184(37)
0.233(47)
0.120(24)
0.101(20)
0.100(20)
0.102(20)
0.208(42)
0.178(36)
0.150(30)
0.082(16)
0.093(19)
0.088(18)
0.064(13)
0.102(20)
0.101(20)
0.067(13)
0.112(22)
0.089(18)
0.172(34)
0.149(30)
0.184(37)
0.233(47)
0.120(24)
0.101(20)
tied u32
d9
d10
d11
d12
d13
d14
d15
d16
d17
d18
d19
d20
d21
d22
d23
d24
d25
d26
0.091
tied u32
0.219(42)
0.130(22)
0.173(24)
0.209(43)
0.178(36)
0.150(30)
b
c
a Restraints defined by ratios are marked: au2/u1, u3/u2, u5/u4. Distances and amplitudes are given in Å.
of 10%, absolute restraints 20% of the calculated value. These
amplitudes and restraints are given in Table 3.
Ab Initio Calculations. Ab initio molecular orbital calculations were
carried out using the Gaussian 94 program.30 Geometry optimizations
and vibrational frequency calculations were performed from analytic
first and second derivatives at the SCF and MP2 levels of theory.
Calculations were undertaken at the SCF level using the standard
3-21G*31-33 and 6-31G*34-36 basis sets, whereas the larger basis set
was used for calculations at the MP2 level of theory. Vibrational
amplitudes for GED restraints were calculated from MP2/6-31G* force
fields by means of the program ASYM40.37
Figure 5. Molecular structure of Me3SiONMe2 in the gas phase as
determined by electron diffraction.
Acknowledgment. This work was supported by the Bay-
erisches Staatsministerium fu¨r Wissenschaft, Forschung und
Kunst (Bayerischer Habilitationsfo¨rderpreis 1996 for N.W.M.),
the Deutsche Forschungsgemeinschaft, the Fonds der Chemis-
chen Industrie, the Leonhard-Lorenz-Stiftung, and the National
Science Foundation under grant CHE95-23581. Generous sup-
port from Prof. K. Hedberg (Corvallis) and Prof. H. Schmidbaur
(Garching) is gratefully acknowledged.
Supporting Information Available: Correlation matrix elements
and calculated vibrational frequencies for Me3SiONMe2 and Me3-
GeONMe2. This material is available free of charge via the Internet at
http://pubs.acs.org.
Figure 6. Molecular structure of Me3GeONMe2 in the gas phase as
determined by electron diffraction.
germanium- and nitrogen-bound methyl groups independently, a total
of 18 parameters was used to refine the structures. The parameter
definitions are listed in Table 1. Eight geometrical restraints based on
the ab initio calculated values (MP2/6-31G*) were used to reduce
correlation between parameters of similar nature (differences) or those
describing hydrogen positions (absolute restraints). Their definition and
values with assigned uncertainties based on an educated guess (experi-
ence with related structures) are also given in Table 1.
All vibrational amplitudes were refined, which belong to a pair of
scatterers contributing more than 5% of the most important pair (Si-
O, Ge-O). Restraints for these amplitudes were calculated from
harmonic force fields obtained at the MP2/6-31G* level of theory,
which were transformed into amplitudes by means of the program
ASYM40.28 Ratios between amplitudes were assigned an uncertainty
IC990689C
(31) Binkley, J. S.; Pople, J. A.; HehreW. J. J. Am. Chem. Soc. 1980, 102,
939.
(32) Gordon, M. S.; Binkley, J. S.; Pople, J. A.; Pietro, W. J.; HehreW. J.
J. Am. Chem. Soc. 1982, 104, 2797.
(33) Pietro, W. J.; Francl, M. M.; Hehre, W. J.; Defrees, D. J.; Pople, J.
A.; Binkley J. S. J. Am. Chem. Soc. 1982, 104, 5039.
(34) Hehre, W. J.; Ditchfield, R.; PopleJ. A. J. Chem. Phys. 1972, 56, 2257.
(35) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28, 213.
(36) Gordon, M. S. Chem. Phys. Lett. 1980, 76, 163.
(37) Hedberg, L.; Mills, I. M. ASYM20, ASYM40, Programs for Force
Constants and Normal Coordinate Analysis, Version 3.0; June 1994.
See also: Hedberg, L.; Mills, I. M. J. Mol. Spectrosc. 1993, 160, 117.