Direct formation of Ge-C bonds from GeO2

Article abstract of DOI:10.1021/ja020499s

Germanium dioxide in the presence of 5% KOH reacted with dimethyl carbonate (DMC) at 250 °C to give (MeO)₄Ge. The reaction of GeO₂ and DMC is similar to that reported for SiO₂; however, the rate of reaction for germanium is much higher than that of the corresponding silicon reaction. In a side-by-side experiment using SiO₂ and GeO₂ where the surface area of the silicon dioxide was 2 orders of magnitude higher than that of the GeO₂, the base-catalyzed reaction with DMC was about an order of magnitude higher for the germanium dioxide. When GeO₂ and 5% KOH were reacted with DMC at 350 °C, two products formed: (MeO)₄Ge (70%) and MeGe(OMe)₃ (30%). Confirmation of the identity of MeGe(OMe)₃ was by GCMS, ¹H and ¹³C NMR, and comparison to an authentic sample made by reaction of MeGeCl₃ with NaOMe. Experiments to determine the mechanism of the direct formation of Ge-C from GeO₂ ruled out participation from CO, H₂, or carbon. The KOH-catalyzed reaction of other metal oxides was explored including B₂O₃, Ga₂O₃, TiO₂, Sb₂O₃, SnO₂, and SnO. Boron reacted to give unknown volatile products. Antimony reacted to give a solid which analyzed as Sb(OMe)₃. SnO reacted with DMC to give a mixture that included (MeO)₄Sn and possibly Me₃Sn(OMe).

Full text of DOI:10.1021/ja020499s

Published on Web 09/05/2002
Direct Formation of Ge-C Bonds from GeO2
Larry N. Lewis,* Kyle E. Litz, and Joseph M. Anostario
Contribution from the GE Global Research Center, Niskayuna, New York 12309
Received April 5, 2002
Abstract: Germanium dioxide in the presence of 5% KOH reacted with dimethyl carbonate (DMC) at 250
°C to give (MeO)4Ge. The reaction of GeO2 and DMC is similar to that reported for SiO2; however, the rate
of reaction for germanium is much higher than that of the corresponding silicon reaction. In a side-by-side
experiment using SiO2 and GeO2 where the surface area of the silicon dioxide was 2 orders of magnitude
higher than that of the GeO2, the base-catalyzed reaction with DMC was about an order of magnitude
higher for the germanium dioxide. When GeO2 and 5% KOH were reacted with DMC at 350 °C, two products
formed: (MeO)4Ge (70%) and MeGe(OMe)3 (30%). Confirmation of the identity of MeGe(OMe)3 was by
GCMS, 1H and 13C NMR, and comparison to an authentic sample made by reaction of MeGeCl3 with NaOMe.
Experiments to determine the mechanism of the direct formation of Ge-C from GeO2 ruled out participation
from CO, H2, or carbon. The KOH-catalyzed reaction of other metal oxides was explored including B2O3,
Ga2O3, TiO2, Sb2O3, SnO2, and SnO. Boron reacted to give unknown volatile products. Antimony reacted
to give a solid which analyzed as Sb(OMe)3. SnO reacted with DMC to give a mixture that included (MeO)4Sn
and possibly Me3Sn(OMe).
Introduction
reported the direct reaction of silicon dioxide with base and
dimethyl carbonate (DMC) to make tetramethoxysilane, eq 3.7
Rochow and Mueller discovered the copper-catalyzed direct
reaction of silicon with methyl chloride in the 1940s (eq 1,
amounts in weight percent).1-3
Si + MeCl Cu + promoters8
250-300 °C
Me2SiCl2 (D)
(83-93%)
(1)
Several related reactions of silicon dioxide to make alkoxysilanes
have been reported.8-13 Subsequent conversion of alkoxysilanes
to a Si-C-containing compound can be accomplished by using
a sacrificial alklylating agent such as that shown in eq 4.14,15
MeSiCl3 (T)
(3-10%)
Me4Si (S)
(0.01-0.5%)
(0.01-0.5%)
(0.1-5.0%)
(1.0-5.0%)
(1.0-5.0%)
Si(OMe)4 + MeMgBr ether8 Me4-xSi(OMe)x
(4)
Me2SiHCl (M2H)
MeSiHCl2 (MH)
Me3SiCl (M)
high boilers
A three-year effort was recently completed by GE researchers
called “Direct Production of Silicones from Sand”.16 No direct
catalytic method was found for production of the Si-C bond
(4) ComprehensiVe Handbook on Hydrosilylation; Marciniec, B., Ed.; Pergamon
Press: Oxford, England, 1992.
The silicon direct process is one of two main catalytic reactions
to form silicon-carbon bonds, the other being hydrosilylation
(eq 2).4-6
(5) Lewis, L. N.; Stein, J.; Gao, Y.; Colborn, R. E.; Hutchins, G. Platinum
Met. ReV. 1997, 41, 66-75.
(6) Stein, J.; Lewis, L. N.; Gao, Y.; Scott, R. A. J. Am Chem. Soc. 1999, 121,
3693.
(7) Ono, Y.; Akiyama, M.; Suzuki, E. Chem. Mater. 1993, 5, 442.
(8) Rosenheim, A.; Raibmann, B.; Schendel, G. Z. Anorg. Allg. Chem. 1931,
196, 160.
R3SiH + R′CHdCH2
9
Pt8 R3SiCH2CH2R′
(2)
(9) Laine, R. M.; Blohowiak, K. Y.; Robinson, T. R.; Hoppe, M. L.; Nardi,
P.; Kampf, J.; Uhm, J. Nature 1991, 353, 642.
A direct reaction of silicon dioxide to form molecular species
containing Si-C bonds is not known. Recently, Ono et al.
(10) Blohowiak, K. Y.; Treadwell, D. R.; Mueller, B. L.; Hoppe, M. L.; Jouppi,
S.; Kansal, P.; Chew, K. W.; Scotto, C. L. S.; Babonneau, F.; Kampf, J.;
Laine, R. M. Chem. Mater. 1994, 6, 2177.
(11) Goodwin, G. B.; Kenney, M. E. Polym. Prepr. 1986, 27, 107.
(12) Bailey, D. L. U.S. Patent 2 881 198, 1959.
* Address correspondence to this author. E-mail: lewisln@crd.ge.com.
(1) Brook, M. A. Silicon in Organic, Organometallic and Polymer Chemistry;
John Wiley & Sons: New York, 2000; p 382.
(2) Liebhafsky, H. A. Silicones Under the Monogram; John Wiley and Sons:
New York, 1978.
(3) Warwick, E. L. Forty Years of Firsts; McGraw-Hill: New York, 1990.
(13) Boudin, A.; Cerveau, G.; Chuit, C.; Corriu, R. J. P. Organometallics 1988,
7, 1165.
(14) Graefe, J.; Uzick, W.; Weinberg, U. Ger. Offen. DE 3821483; Chem. Abstr.
1990, 112, 217263.
(15) Goodwin, G. B.; Kenney, M. E. AdV. Chem. Ser. 1990, 224, 251.
9
11718
J. AM. CHEM. SOC. 2002, 124, 11718-11722
10.1021/ja020499s CCC: $22.00 © 2002 American Chemical Society

Direct Formation of Ge−C Bonds from GeO
A R T I C L E S
Table 1. PSD and Surface Area of GeO2 and Two SiO2
2
from any compound with four oxygen groups, although a new,
unexpected formation of a Si-C bond was found by stoichio-
metric reaction of tetramethoxysilane with sodium hydride.17
The chemistry of germanium is similar to that of silicon. A
germanium direct process18-25 and Ge-H26,27 addition across
double bonds are the two catalytic methods for formation of
germanium-carbon bonds. Direct catalytic formation of Ge-H
bonds has been reported via Ge(II) species.28 However, a
reaction analogous to eq 3 for germanium had not been
previously reported. In addition, the only reported method for
synthesis of (MeO)4Ge or MeGe(OMe)3 involves the reaction
of MeOH with GeX429 or MeGeX3 (X ) halide). Formation of
tetramethoxygermane is potentially interesting because of the
need for GeO2 precursors as refractive index modifiers in the
fiber-optics industry.30-33
Recently, Dupont researchers reported the first direct stan-
nylation of aromatic hydrocarbons, forming new C-Sn bonds.34
This work employed (CF3CO2)4Sn reacting reversibly with
benzene and p-xylene to make isolable aryltin compounds. This
work received considerable interest owing to the plethora of
C-C bond-forming reactions arylstannanes undergo. Nonethe-
less, direct stannylation of aromatic or alkyl hydrocarbons from
tin oxides has not yet been reported.
particle size, µM
(std dev)
2
compound
GeO2
SiO2 (Minusil)
SiO2 (silica gel)
surface area, m /g
2.2 (1.3)
2.3 (1.3)
<1 (0.5)
2.1
2.6
200
germanium was more reactive than silicon in the base-catalyzed
reaction of the oxide with DMC.
A silicon dioxide source of particle size and surface area
roughly equivalent to those of the germanium oxide was used
to compare reactivity to that of germanium. Table 1 details the
particle size and surface area of the germanium dioxide used in
this study and two different silicon dioxides investigated.
Equimolar amounts of GeO2 and SiO2 (Minusil, R-quartz) were
combined with 5% KOH by weight and reacted in the fixed-
bed reactor with DMC at 250 °C. Quantitative conversion of
all of the GeO2 in the bed occurred to give (MeO)4Ge. The
SiO2 reacted to produce only a trace of (MeO)4Si. Similarly,
Minusil alone was poorly reactive: in the presence of 5% KOH,
only trace levels of (MeO)4Si formed. The comparison of
reactivity experiment was repeated, except that silica gel was
used in place of Minusil. Silica gel is amorphous and has a
high surface area. The amorphous silicon dioxide sources were
the most reactive of all the many SiO2 compounds investigated.16
Nevertheless, germanium dioxide was much more reactive than
silicon dioxide (silica gel), despite the fact that the silicon
dioxide used had a surface area 2 orders of magnitude higher
than that of the germanium oxide used (Figure 1). All of the
GeO2 in the bed was consumed in less than 2 h, while almost
6 h was needed to react all of the silicon dioxide present in the
bed. Both germanium and silicon dioxides were quantitatively
consumed to form M(OMe)4 (M ) Ge and Si).
The base-catalyzed reaction of germanium dioxide was
repeated at higher temperature. When the reaction was per-
formed at 350 °C, a second product was formed in about 25%
selectivity with a lower gas chromatographic (GC) retention time
than the Ge(OMe)4. Analysis of the new product by gas
chromatography/mass spectroscopy (GCMS) was consistent with
formation of MeGe(OMe)3.35 1H and 13C NMR analysis of the
product mixture from reaction of 5% base and GeO2 with DMC
showed the presence of new peaks in the methyl and methoxide
regions. An authentic sample of MeGe(OMe)3 was prepared
following the procedure of West et al.,36 and mass spectroscopic
and 1H and 13C NMR analysis confirmed that the new product
was indeed MeGe(OMe)3.37
This report describes (1) the facile base-catalyzed reaction
of germanium oxide and DMC toward (MeO)4Ge, (2) a
comparison of the reactivity of SiO2 with that of GeO2, and (3)
the unexpected direct synthesis of MeGe(OMe)3, the first direct
reaction from GeO2 to form molecular Ge-C-containing
species.
Results and Discussion
Germanium dioxide with 5% KOH was reacted with DMC
in a fixed-bed reactor at 250 °C, resulting in virtually quantita-
tive formation of (MeO)4Ge. As expected, the chemistry of
germanium dioxide was similar to that found for reaction with
silicon dioxide, eq 3. It was of interest to determine if
(16) Lewis, L. N.; Schattenmann, F. J.; Jordan, T. M.; Carnahan, J. C.; Flanagan,
W. P.; Wroczynski, R. J.; Lemmon, J. P.; Anostario, J. M.; Othon, M. A.
Inorg. Chem. 2002, 41, 2608.
(17) Schattenmann, F. J.; Ligon, W. V.; Donahue, P.; Grade, H.; Abatto, K. J.
Am. Chem. Soc., in press.
(18) Zueva, G. Ya.; Luk’yankina, N. V.; Ponomarenko, V. A. IzV. Akad. Nauk
SSSR, Ser. Khim. 1967, 192-194; Chem. Abstr. 1967, 66, 95140.
(19) Moedritzer, Kurt. U.S. Patent 3 446 824, 1969; Chem. Abstr. 1969, 71,
50227.
(20) Belyi, A. P.; Gorbunov, A. I.; Golubtsov, S. A.; Fel’dshtein, N. S. J.
Organomet. Chem. 1969, 17, 485-489.
(21) Chong, T.; Skaates, J. M. J. Catal. 1973, 28, 20-32.
(22) Lee, M. E.; Bobbitt, K. L.; Lei, D.; Gaspar, P. P. Synth. React. Inorg. Met.-
Org. Chem. 1990, 20, 77-81.
The origin of formation of (MeO)4Ge from the base-catalyzed
reaction with DMC is likely similar to that proposed for the
silicon analogue (Scheme 1).16 Analysis of the gases from the
base-catalyzed reactions of both silicon and germanium dioxide
with DMC by GCMS showed primarily formation of CO2. The
route in Scheme 1 is further supported by previous work on
dissolution of silicon dioxide and base.38 In addition, the higher
reactivity of germanium vs silicon is consistent with the fact
that germanium is more electropositive than silicon, and the
(23) Gar, T. K.; Mironov, V. F. Metallorg. Khim. 1993, 6, 285; Chem. Abstr.
1994, 120, 164268.
(24) Schlecht, S. Angew. Chem., Int. Ed. 2002, 41, 1178.
(25) Okamoto, M.; Asano, T.; Suzuki, E. Catal. Lett. 2002, 78, 33.
(26) Bulten, E. J. MTP Int. ReV. Sci., Inorg. Chem. Ser., 2 1975, 4, 241.
(27) Lesbre, M.; Mazerolles, P.; Satge, J. The Organic Compounds of Germa-
nium; Wiley: London, 1971.
(28) Litz, K. E.; Bender, J. E.; Banaszak Holl, M. M. Angew. Chem., Int. Ed.
1997, 36, 496.
(29) Mehrotra, R. C.; Gupta, V. D.; Srivastava, G. ReV. Silicon, Germanium,
Tin Lead Compd. 1975, 1, 299-377.
(30) Yokogawa, K.; Kamiya, K.; Omura, T. Jpn. Kokai Tokkyo Koho, JP
60151242 A2 19850809, 1985; Chem. Abstr. 1986, 104, 94045.
(31) Susa, K.; Matsuyama, I. Jpn. Kokai Tokkyo Koho, JP 61035847 A2
19860220, 1986; Chem. Abstr. 1986, 105, 106232.
(32) Konishi, S. Jpn. Kokai Tokkyo Koho, JP 62197328 A2 19870901, 1987;
Chem. Abstr. 1987, 107, 181973.
(35) Dube, G. Z. Chem. 1969, 9, 316-317.
(36) West, R.; Hunt, H. R.; Whipple, R. O. J. Am. Chem. Soc. 1954, 76, 310.
(37) Drake, J. E.; Glavincevski, B. N.; Humphries, R. E.; Majid, A. Can. J.
Chem. 1979, 57, 1426-1430.
(33) Nishimoto, H. Jpn. Kokai Tokkyo Koho, JP 09133827 A2 19970520, 1997;
Chem. Abstr. 1997, 127, 72786.
(34) Grushin, V. V.; Marshall, W. J.; Thorn, D. L. AdV. Synth. Catal. 2001,
(38) Cheng, H.; Tamaki, R.; Laine, R. L.; Babonneau, F.; Chujo, Y.; Treadwell,
D. R. J. Am. Chem. Soc. 2000, 122, 10063.
143, 433.
9
J. AM. CHEM. SOC. VOL. 124, NO. 39, 2002 11719

A R T I C L E S
Lewis et al.
Figure 1. Relative rate of formation of M(OMe)4 vs time (M ) Ge and Si, silica gel).
Scheme 1. Possible Scheme for KOH-Catalyzed Formation of
(MeO)4Ge from GeO2 and DMC
Scheme 2. Possible Scheme for Formation of MeGe(OMe)3 from
Reduced Ge Species and DMC
Ghosh40 and of Datta and Hati41 and the experimental evidence
cited therein, which support the statement that germanium is
less electronegative (and thus more electropositive) than silicon.
A more meaningful argument on our part would be based upon
relative Ge-O/Si-O bond enthalpies (659 vs 800 kJ/mol
according to CRC Handbook of Chemistry and Physics42), where
the increased reactivity arises from the relatively weaker Ge-O
bonds.
The base-catalyzed reactions of GeO2 and DMC at 250 °C
resulted in an off-white solid residue, while the reactions run
at 350 °C, e.g., when Ge-C bond formation occurred, resulted
in black bed residues. We ruled out catalytic processes due to
the adventitious presence of other elements in the germanium
oxide. In addition, XRD of our GeO2 used revealed the structure
to be hexagonal. We considered three possibilities for formation
of the Ge-C bond. Carbon monoxide was considered as a
possible reducing agent. Formation of CO, and its possible role
as an oxygen acceptor, was supported by a reaction between
5% KOH, GeS2, and DMC at 250 °C. While reaction of
germanium disulfide did not result in formation of germanium
proposed scheme would predict that germanium would be more
reactive than silicon. However, there is no clear route for
formation of the Ge-C bond. One possibility may be via
alkoxycarbonylgermanium intermediates, formed by reaction of
DMC with reduced Ge on the surface, which decarboxylate to
yield Me-Ge species (Scheme 2). Experiments were conducted
to test this hypothesis.
A reviewer is thanked for noting that there remains a
continuing debate on the relative electronegativity of group 14
elements.39 We are persuaded by the calculations of Ghanty and
(40) Ghanty, T. K.; Ghosh, S. K. J. Phys. Chem. 1996, 100, 17429.
(41) Datta, D.; Hati, S. J. Phys. Chem. 1995, 99, 10742.
(42) CRC Handbook of Chemistry and Physics, 71st ed.; CRC Press: Boca
Raton, FL, 1991-2; pp 9-105-9-113.
(39) Cotton, F. A.; Wilkinson, G.; Murillo, C. A.; Bochmann, M. AdVanced
Inorganic Chemistry, 6th ed.; Wiley-Interscience: New York, 1999; p 259.
9
11720 J. AM. CHEM. SOC. VOL. 124, NO. 39, 2002

Direct Formation of Ge−C Bonds from GeO
A R T I C L E S
2
methoxides, GCMS analysis of the gas products from this
reaction did show OdCdS. The observation of carbonyl sulfide
suggests that if CO can remove S from GeS2, then possibly
CO can remove an oxygen from GeO2. However, when the base-
catalyzed reaction of GeO2 and DMC was carried out in a gas
stream containing 10 vol % CO, no Ge-C bond formation
occurred at 250 °C, nor was there observed an increase in Ge-C
formation at the typical temperature of 350 °C.
for the tin-based product by NMR. The 119Sn NMR showed
primarily one peak at +126 ppm. The literature values of 119Sn
NMR for the series (Me4-nSn(OMe)n (n ) 0-4) are consistent
with our 126 ppm peak assigned to Me3Sn(OMe).43 Further-
1
more, the H NMR showed two methoxide resonances at 3.30
(major) and 3.26 (minor) ppm. In addition, the 1H NMR showed
a peak at 0.32 ppm with (presumed) Sn-HJ coupling (satellites)
of 58 Hz. The 3.26/0.32 ppm peaks are consistent in location,
relative intensity, and coupling constant with Me3Sn(OMe),
while the single 3.30 ppm peak could be assigned to (MeO)4Sn,
as suggested by the GC/GCMS data.43-44 It is possible that the
(MeO)4Sn and Me3Sn(OMe) (if present) coeluted in the GC
and thus gave rise to the single peak there and a complicated
mixture in the mass spectrum obtained in the GCMS analysis.
Summary. In this report, we describe the facile base-
catalyzed reaction of germanium oxide and DMC toward
(MeO)4Ge in high GeO2 conversion (80%) and selectivity
(>98%). A comparison of the reactivity of SiO2 with that of
GeO2 revealed Ge to be significantly more reactive to base-
catalyzed reaction with DMC than Si, despite 2 orders of
magnitude difference in their surface areas. Increasing the
temperature from 250 to 350 °C resulted in the unexpected direct
synthesis of MeGe(OMe)3 with (MeO)4Ge in a 1:3 ratio, with
a very slight drop in overall GeO2 conversion (72%). This is
the first reported direct reaction from GeO2 to molecular Ge-
C-containing species. Apparently, other metal oxides in com-
bination with base will react with DMC to give metal alkoxides.
Some evidence exists, at least for tin, that direct metal-carbon
bond formation can occur for a metal other than germanium.
A second possible reducing agent for Ge-C formation was
H2. Like the CO example above, addition of H2 to the carrier
gas stream in the base-catalyzed reaction of GeO2 with DMC
did not result in extra Ge-C bond formation. Finally, we
considered that carbon could act as an oxygen acceptor. Carbon
deposition from nonproductive DMC decomposition likely
occurred via C-H bond breakage.16 Therefore, base-catalyzed
reaction of GeO2 with DMC was performed in the presence of
activated carbon. No additional formation of Ge-C bonds was
caused by the presence of activated carbon, and no additional
species were formed. Thus, at this point, the mechanism of
formation of MeGe(OMe)3 from base-catalyzed reaction of
GeO2 and DMC is not known.
Other Metals. We investigated the reaction of some other
metal oxides in a manner similar to that described for GeO2.
When Ga2O3, TiO2, or SnO2 was used with 5 wt % KOH in
place of GeO2 in the reaction with DMC, no condensed product
containing the appropriate metal was obtained, nor was there
significant weight loss from the bed. These results suggest that
virtually no reaction of these metal oxides with DMC occurred
from 250 to 350 °C. B2O3 was combined with 5 wt % KOH
and then reacted with DMC from 250 to 350 °C. No boron-
containing product was condensed. However, the bed lost almost
all of its weight. A volatile, noncondensed material was made
from this reaction, but its identity is not known at this time.
Experimental Section
Measurements. GC measurements were made using a Hewlett-
Packard model 6890 instrument with a thermal conductivity detector
and a 30 m HP 1 column (0.32 mm i.d. and 0.25 µM film thickness).
GCMS data were recorded using a JEOL SX 102 high-resolution,
double-focusing magnetic sector instrument with a 30 m DB 5 capillary
column. NMR spectra were recorded in CDCl3 with TMS as an internal
standard on a GE QE-300 instrument, 1H and 13C NMR at 300.15 and
75.48 MHz, respectively. XRD measurements were performed on a
Siemans D500 diffractometer, which utilized a Bragg-Brentano vertical
Θ:Θ goniometer. The instrument used a Cu anode tube, (wavelength
for Cu KR1 ) 1.54056 Å), and sample was measured at the following
parameters:
When Sb2O3 was used with 5% KOH and contacted with
DMC, a reaction did occur. There were no antimony-containing
products in the condensables, but a solid was formed in the
heated elbow region that exits the reactor (past the glass frit
holding the solids). Thus, some volatile material exited the bed
and deposited as a solid on the elbow. The solid, brown, hard
material was insoluble in common organic solvents and was
not volatile enough to give a mass spectrum. Elemental analysis
showed a reasonable match consistent with Sb(OMe)3. In one
experiment performed at 350 °C, the bed lost 85% of its weight,
consistent with virtually complete conversion of the antimony
oxide to volatile products. A likely possibility is that the
antimony methoxide formed as a monomer that was volatile
enough to escape the reactor but immediately polymerized on
the elbow.
power generation
instrument setup
kV ) 45
mA ) 40
2Θ range ) 10-90°
step size ) 0.05°
dwell time ) 2 s
slit size: Siemans D500 ) 1°
Finally, the reaction of SnO (as opposed to SnO2 as described
above) was performed with 5% KOH and DMC. Up to 60%
weight loss was noted when the reaction was carried out at 350
°C; less weight loss occurred at lower temperatures. The identity
of the main tin-containing product is equivocal. GC analysis
showed a relatively clean mixture composed of unreacted DMC,
a new peak with slightly higher retention time than that of
(MeO)4Si, and the internal standard, cyclooctane. GCMS
analysis of the product solution was consistent with formation
of (MeO)4Sn. While no library mass spectrum of (MeO)4Sn was
available, a computer-generated pattern closely matched the
observed pattern we found. Contradictory evidence was obtained
Particle size measurements (10 replicates) were made as follows. Small
samples of powder were removed and placed on a glass slide. A small
droplet of dispersion fluid with a refractive index that provided good
contrast between the powder and background was applied to a glass
cover slip. The cover slip was placed on the powder sample and
dispersed using mechanical agitation. The slide was then placed on a
transmitted light microscope (Zeiss Photomicroscope III) and observed
at sufficient magnification to distinguish the size and shape of the power.
The microscope was outfitted with a high-resolution camera (Kodak
ES1), which was interfaced to a computer in order to provide a digital
(43) Kennedy, J. D. J. Mol. Struct. 1976, 31, 207.
(44) Kennedy, J. D. J. Chem. Soc., Perkin Trans. 2 1977, 242.
9
J. AM. CHEM. SOC. VOL. 124, NO. 39, 2002 11721

A R T I C L E S
Lewis et al.
Table 2. Analysis of Elements in GeO2, Minusil, and Silica Gel
argon) above the upper heating zone through a septum via motor-driven
syringe. The reactor downstream was collected using a water-chilled
condenser.
element
GeO , ppm
Minusil, ppm
silica gel, ppm
2
Al
B
<10
<20
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
major
<10
<10
<10
<10
<10
<10
<10
<10
<10
<10
75
780
10
13
78
<10
20
<10
783
<10
<10
<10
<10
38
Typical Procedure. The GeO2 source was stirred in an aqueous
solution of KOH. The suspension was heated to dryness, and the solid
obtained was further dried overnight at 115 °C. The material was ground
and charged into the fixed-bed flow reactor. The solid (typically 1.0
g) was kept at the reaction temperature in the argon stream for 1 h.
DMC (typically 20 mmol/h ) 1.68 mL/h; a total of 10 mL) was fed
into the carrier gas stream (20 mL/min) using a motor-driven syringe.
The reaction products downstream were collected in fractions using a
water-chilled condenser and analyzed by gas chromatography. Forma-
tion of Ge(OMe)4 and MeGe(OMe)3 was confirmed by GCMS and 1H
and 13C NMR. The GeO2-utilization values were calculated from the
collected amounts of Ge(OMe)4 and MeGe(OMe)3. The weight differ-
ence in the GeO2 bed before and after reaction was also determined;
however, a black carbon-based deposit rendered a precise weight
determination impossible.
Ba
Be
Ca
Cd
Co
Cr
Cu
Fe
Ga
Mg
Mn
Mo
Ni
P
<10
212
<10
<10
<10
<10
290
<10
36
<10
<10
<10
51
<10
<10
10
140
<10
<10
23
<10
<25
<10
<10
<10
<10
<10
<10
<25
<25
<25
<10
<10
<10
123
<10
<10
<10
13
<10
<10
<10
112
Pb
Sn
Sr
Ti
V
The GeO2/KOH mixture was added to the fixed-bed reactor (1 g)
and then heated to 250 °C with DMC injection and the Ar carrier gas.
More than 80% of the GeO2 reacted to make a single product, Ge-
<10
<10
41
Zn
Zr
As
Bi
Ge
Hf
Hg
In
Nb
Sb
Ta
Te
Th
U
1
(OMe)4. The product was confirmed by GC, GCMS, and H and 13C
<10
<25
<10
<10
<10
<10
<10
<10
<25
<25
<25
<10
<10
NMR analysis. GC analysis showed a single peak with retention time
just below that of the cyclooctane internal standard. GCMS analysis
showed that the mass spectrum of the observed peak had a peak
envelope centered around 196 amu. Fragments indicated loss of OMe
groups. Ge has five naturally occurring isotopes giving rise to the
1
envelope pattern. H NMR showed a single resonance at 3.59 ppm
(CDCl3), and 13C NMR showed a single resonance at 53.96 ppm.
The experiment was repeated, except that the temperature was 350
°C. GC analysis showed that Ge(OMe)4 and MeGe(OMe)3 were
produced in about a 3:1 ratio. Total GeO2 conversion was 72%.
Confirmation of the formation of MeGe(OMe)3 was accomplished with
GCMS (EI) and GCMS (CI, H+). The latter technique showed the parent
ion center around 179 amu. Fragments were observed for M+ - CH3
and M+ - OMe. 1H NMR showed the presence of new methoxy
resonances at 3.53 ppm and Ge-CH3 methyl resonance at 0.59 ppm,
while 13C NMR showed the presence of a new methoxy carbon
resonance at 52.01 ppm and a CH3-Ge methyl resonance at -6.2 ppm;
W
Si
XRD analysis showed that all of the GeO2 was in the hexagonal phase.
image for further analysis. Image analysis software (Clemex Vision)
was used to quantify the particle sizes.
Materials. GeO2 was obtained from Aldrich. Elemental analysis was
done by Galbraith. Minusil was obtained from GE Silicones (Waterford,
NY), and silica gel was obtained from JT Baker (60-200 mesh). Table
2 gives elemental analysis data for the metal oxides used.
Procedures. Reactions were performed in a fixed-bed reactor as
described previously.45 The reactor was a glass tube (20 cm long; 1
cm i.d.) with a porous glass grid to hold the GeO2 sample. The reactor
was centered vertically in a Nichrome ribbon-wound glass tube (5 cm
o.d.). Two pairs of electrodes were fitted to the Nichrome to create
two heated zones. The top heating zone was used to preheat the DMC/
carrier gas mixture, and the bottom section was for the reactor itself.
The Nichrome-wound tube was centered in a quartz glass tube (6.4
cm o.d.) for insulation and safety purposes. The carrier gas stream was
regulated by using a mass flow controller (MKS Instruments type
1179A) connected to a four-channel readout device (MKS Instruments
type 247D). The DMC was added into the carrier gas stream (typically
1
reference H and 13C values for Ge-CH3 were 0.51 and -7.43 ppm,
respectively.31
Synthesis of MeGe(OMe)3.36 To a three-necked round-bottom flask
equipped with an argon inlet, reflux condenser, and magnetic stir bar
was added MeGeCl3 (9.5 g, 49 mmol, Gelest). NaOMe was prepared
by slow addition of Na (3.6 g, 156 mmol) to MeOH (35 mL) under
Ar. The NaOMe/MeOH solution was slowly added to MeGeCl3, first
at 0 °C and then at reflux (2 h), with stirring. A white precipitate was
removed by filtration. The MeOH was removed by ambient pressure
distillation, and then the product was purified through a short path
column at 135 °C at ca. 100 mmHg. About 2.5 g of colorless liquid
was collected.
Acknowledgment. Stanlee Buddle carried out the BET
surface area measurements. Jim Grande performed the particle
size measurements. Michelle Othon performed the XRD mea-
surements. John Leman is thanked for helpful discussions. This
work was supported by U.S. Department of Energy Contract
DE-FC02-98CH10931.
(45) (a) Ward, W. J.; Lewis, L. N.; Bablin, J. M. In Silicon for the Chemical
Industry V; Oye, H. A., Rong, H. M., Nygaard, L., Schussler, G., Tuset, J.
K., Eds.; Norwegian University of Science & Technology: Trondheim,
Norway, 2000; p 309. (b) Lewis, L. N.; Ward, W. J. Ind. Eng. Chem. Res.
2002, 41, 397-402.
JA020499S
9
11722 J. AM. CHEM. SOC. VOL. 124, NO. 39, 2002