Reactions of silicon atoms with methane and silane in solid argon: A matrix-spectroscopic study

Article abstract of DOI:10.1002/1521-3765(20021004)8:19<4383::AID-CHEM4383>3.0.CO;2-1

The reaction of silicon atoms with methane (1) and silane (7) in argon at 10 K has been studied. Methane (1) is found to be inert to silicon atoms, but reaction occurs upon photochemical excitation under formation of methylsilylene (2). On the contrary, silane (7) is spontaneously converted into a mixture of silylsilylene (10) and disilene (12). Subsequent irradiation generates the butterfly-type disilyne Si₂H₂ (14), together with a third photoproduct, which we assume to be the doubly bridged Si₂H₄ isomer 13. The structural elucidation of the new species is based on the comparison of the experimental IR and UV/Vis spectra with data from density functional theory calculations. The results are supported by isotopic labeling studies.

Full text of DOI:10.1002/1521-3765(20021004)8:19<4383::AID-CHEM4383>3.0.CO;2-1

FULL PAPER
Reactions of Silicon Atoms with Methane and Silane in Solid Argon:
A Matrix-Spectroscopic Study**
G¸nther Maier,* Hans Peter Reisenauer, and Jˆrg Glatthaar[a]
Abstract: The reaction of silicon atoms with methane (1) and silane (7) in argon at
10 K has been studied. Methane (1) is found to be inert to silicon atoms, but reaction
occurs upon photochemical excitation under formation of methylsilylene (2). On the
contrary, silane (7) is spontaneously converted into a mixture of silylsilylene (10) and
disilene (12). Subsequent irradiation generates the butterfly-type disilyne Si2H2 (14),
together with a third photoproduct, which we assume to be the doubly bridged Si2H4
isomer 13. The structural elucidation of the new species is based on the comparison of
the experimental IR and UV/Vis spectra with data from density functional theory
calculations. The results are supported by isotopic labeling studies.
Keywords:
density functional calculations
matrix isolation ¥ photoisomerization
co-condensation
¥
¥
Introduction
Results and Discussion
During the past five years, we have studied the reactions of
thermally generated silicon atoms with low molecular weight
reactants in an argon matrix. The reaction products were
identified by IR and UV/Vis spectroscopy, aided by compar-
ison with calculated spectra. The method turned out to be very
versatile and successful.[2] The selected substrates were
molecules with isolated, conjugated, or aromatic p bonds,
compounds containing p bonds and at the same time free
electron pairs, and molecules possessing n electrons only in
combination with s bonds. These reactions can be understood
considering the basic features of atomic silicon. First, it has a
triplet ground state and thus a diradical type of reaction can
be anticipated. According to the law of spin conservation the
primary reaction product should be a triplet molecule.
Second, the silicon atom has an empty 3p orbital, and as a
consequence strong electrophilic behavior can be expected.
No wonder that all molecules with p and n electrons are
excellent partners.[2] In continuation of our strategy we
wanted to learn something about the reactivity of pure s
systems. For instance, which reactions will occur if simple
substrate molecules such as methane (1) or silane (7) come in
contact with silicon atoms? The answer is given herein.
Reactions of methane: The selection of methane (1) has not
only an academic aspect but also practical relevance. There
are reports that porous silicon can be a dangerous material in
the presence of oxygen or nitrogen, depending on the grain
size. Lump silicon represents one end of the scale, silicon
atoms the other extreme. Silicon powder lies in between.
Applying our standard procedure[2] we condensed methane
(1) as a gaseous mixture with argon onto a spectroscopic
window at 10 K. The concentration varied from pure methane
to a methane/argon ratio of 1:1000. The silicon atoms,
generated by resistive heating of a silicon rod to a temper-
ature of about 13808C, were separately cocondensed at the
same time. The relative amount of 3Si atoms was estimated by
using a quartz microbalance. In all cases only the IR spectrum
of the starting material could be registered. There was not
even an indication for the existence of a complex 4 between
methane (1) and a silicon atom. This was in our hands the first
case that the substrate molecule did not react with silicon
atoms.
If one rationalizes how a reaction can be enforced in such a
case, one has to keep in mind that the silicon atom in the gas
phase shows a weak UV transition at 220 nm and another,
strong absorption at 251 nm. This suggests that irradiation
with light of wavelength 254 nm can be a way to activate
silicon atoms.
Indeed, irradiation of the co-condensate of methane (1) and
silicon atoms with short wavelengths (l ˆ 185 or 254 nm)
leads to methylsilylene (2). Evidently, photoexcitation en-
[a] Prof. Dr. G. Maier, Dr. H. P. Reisenauer, Dr. J. Glatthaar
Institut f¸r Organische Chemie
Justus-Liebig-Universit‰t Giessen
Heinrich-Buff-Ring 58, 35392 Giessen (Germany)
Fax: (49)641-99-34309
À
forces the insertion of a silicon atom into the C H bond of
methane (1). The intensity of the IR bands of 2 arising during
E-mail: guenther.maier@org.chemie.uni-giessen.de
[**] Hetero p Systems, Part 32: Part 31 see: ref. [2h]. See also ref. [1]
Chem. Eur. J. 2002, 8, No. 19
¹ 2002 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/02/0819-4383 $ 20.00+.50/0
4383

G. Maier et al.
FULL PAPER
Table 1. Theoretical (B3LYP/6 311 ‡‡ G**) IR absorptions (unscaled;
wavenumbers incmÀ1) of CH4Si isomers 2, 3, 5, 6, and connecting transition
state TS2,3. Absolute intensities [kmmolÀ1] are given in parentheses.
Species Point State IR Absorptions (Intensity), Symmetry
group
2
Cs
C2v
Cs
Cs
C1
1A'
1A1
3A''
3A''
1A1
152.4 (1) a'', 572.5 (5) a'', 628.9 (10) a', 662.5 (57)
a', 950.7 (84) a', 1264.3 (27) a', 1434.4 (6) a',
1449.2 (15) a'', 2009.3 (332) a', 2990.5 (4) a',
3042.5 (15) a'', 3103.1 (13) a'
454.1 (25) b1, 485.1 (6) b2, 729.1 (0) a2, 780.6 (59)
b1, 841.9 (81) b2, 941.6 (26) a1, 994.2 (29) a1,
1399.1 (10) a1, 2270.0 (32) a1, 2293.2 (96) b2,
3142.1 (0) a1, 3230.6 (0) b2
117.9 (0) a'', 555.8 (8) a', 659.2 (12) a', 791.5 (13)
a'', 881.4 (71) a', 1267.9 (0) a', 1449.9 (10) a',
1453.8 (8) a'', 2164.1 (82) a', 3019.8 (10) a', 3091.9
(1) a'', 3108.3 (4) a'
116.1 (41) a'', 329.5 (32) a', 632.0 (46) a'', 640.3
(48) a', 796.3 (29) a', 943.5 (66) a', 946.1 (52) a'' ,
948.2 (210) a' , 2196.2 (118) a'', 2199.4 (79) a',
2224.2 (82) a', 3261.8 (2) a'
3
5
irradiation was relatively weak. Nevertheless, its structural
elucidation has to be taken for granted, since it is possible to
establish a photoequilibrium between 2 and the isomeric
silaethene (3). With l > 400 nm the equilibrium is shifted to
silaethene (3), with l ˆ 254 nm to the side of methylsilylene
(2). The IR spectra of 2 and 3 and the mutual interconversion
have been studied by us before.[3]
6
TS2,3
À 1101.2 (155) a1, 593.3 (3) a1, 788.7 (37) a1,
808.9 (7) a1, 892.6 (46) a1, 994.6 (60) a1, 1008.6
(94) a1, 1438.6 (5) a1, 1891.6 (10) a1, 2097.4 (208)
a1, 3070.7 (16) a1, 3164.7 (3) a1
A thermal equilibrium between 2 and 3 should be on the
side of silaethene (3). According to calculations (B3LYP/6-
311 ‡ G**), carried out in comparison to the silane system
(see below), methylsilylene (2) should be 2.5 kcalmolÀ1 higher
in energy (Scheme 1; Table 1). The connecting transition state
TS2,3 is 36.3 kcalmolÀ1 above silaethene (3). The starting
3
components Si and methane (1) lie 36.8 kcalmolÀ1 higher
than silaethene (3). The singlet/triplet gap of the two silylenes
2 and 5 is calculated to be
22.6 kcalmolÀ1. Triplet silylcar-
bene (6) is even higher in en-
ergy than the starting com-
3
pounds Si atoms plus methane
(1) and therefore cannot be
expected as
a product. At-
tempts to calculate a triplet
complex 4 between a silicon
atom and methane (1) failed.
Reactions of silane
Prediction: On the basis of two
arguments it is possible to pre-
dict that the behavior of silane
(5) might be completely differ-
ent from that of methane (1).
On the one hand Skell and
Owen described already in
1967[4] that thermally vaporized
silicon atoms insert into the
À
Si H bond of trimethylsilane,
which has been deposited on a
liquid nitrogen cooled wall. On
the other hand it can be shown
by calculation that methane (1)
and silane (5) behave quite
differently when attacked by a
silicon atom. Gordon et al.[5]
calculated in 1989 that the in-
Scheme 1. Calculated (B3LYP/6 311 ‡ G**) relative energies and geometries of some stationary points on the
CH4Si potential energy surface (PES); zero-point vibrational energies included. Distances are given in ä and
angles in degrees.
3
À
sertion of Si atoms into a Si H
bond of silane needs only a
4384
¹ 2002 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/02/0819-4384 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, No. 19

Reaction of Si Atoms with Methane and Silane
4383 4391
small barrier. For the corresponding reaction with methane
(1) no clear answer could be given. Our own calculations
substantiate the big difference in reactivity between silane (7)
and methane (1) (Figure 1).
together with the corresponding vibrational spectra were
calculated with the 6 311 ‡ G** basis set and the B3LYP
functional. The results are presented in Table 2. Scheme 2
shows the calculated relative energies of some minima and the
connecting transition states.
Table 2. Theoretical (B3LYP/6 311 ‡ G**) IR absorptions (unscaled;
wavenumbers incmÀ1) of Si2H4 isomers 8 13, disilyne (14), silylene (15),
and connecting transition states. Absolute intensities [kmmolÀ1] are given
in parentheses.
Species Point State IR Absorptions (Intensity), Symmetry
group
8
C1
3A1
3A''
1A'
1A1
1Ag
1Ag
1A1
99.2 (1) a1, 109.8 (0) a1, 400.1 (16) a1, 777.6 (27)
a1, 867.4 (497) a1, 927.6 (69) a1, 945.2 (60) a1,
1069.6 (88) a1, 1827.6 (280) a1, 2238.3 (79) a1,
2267.3 (44) a1, 2282.2 (63) a1
108.0 (0) a'', 416.2 (4) a', 428.2 (21) a', 524.4 (8)
a'', 648.8 (18) a', 887.6 (358) a', 937.8 (50) a'',
949.4 (63) a', 2162.1 (87) a', 2199.9 (79) a', 2206.1
(91) a'', 2234.6 (90) a'
67.9 (12) a'', 369.1 (9) a', 389.0 (33) a'', 428.9 (27)
a', 717.1 (63) a', 868.1 (224) a', 932.8 (73) a', 957.2
(36) a'', 2031.3 (208) a', 2180.5 (78) a', 2189.5
(112) a'', 2217.5 (119) a'
397.4 (7) a1, 447.4 (3) a1, 477.7 (27) a1, 636.2 (7)
a1, 703.1 (37) a1, 858.5 (94) a1, 961.4 (40) a1,
1009.9 (350) a1, 1625.7 (113) a1, 2042.7 (182) a1,
2195.2 (148) a1, 2220.6 (126) a1
324.2 (0) ag, 348.0 (23) au, 446.1 (32) bu, 524.1 (0)
au, 562.1 (0) ag, 615.7 (0) bg, 919.5 (182) bu, 955.8
(0) ag, 2224.5 (114) bu, 2228.9 (0) ag, 2245.7 (0)
bg, 2257.6 (150) au
281.5 (3) bu , 376.7 (0) ag, 708.3 (17) au, 833.1
(156) bu, 845.1 (0) ag, 849.6 (0) bg, 1287.7 (32) au,
1428.6 (0) bg, 1453.1 (978) bu, 1633.0 (0) ag,
2043.6 (0) ag, 2060.9 (454) bu
9
Cs
10
11
12
13
14
Cs
C1
C2h
C2h
C2v
Figure 1. Calculated changes of the potential energy during the approach
of a ground state Si atom (triplet) to a methane molecule (1; upper curve)
or a silane molecule (7; lower curve); B3LYP/6 311 ‡ G**; full optimi-
zation at each step.
If a silicon atom in its triplet ground state approaches
methane (1) the energy is continuously raised. There is no
indication of any bonding interaction. On the contrary, the
reaction coordinate for the approach between a 3Si atom and
silane (7) descends steadily until complex 8 is reached.
522.7 (1) a1, 961.0 (50) a1, 1081.8 (0) a2, 1172.5
(382) b2, 1535.4 (10) b1, 1617.2 (5) a1
15
TS8,9
C2v
C1
1A1
3A1
1024.4 (110) a1, 2041.8 (303) b2, 2042.5 (291) a1
À 1020.8 (33) a1, 272.3 (0) a1, 291.7 (28) a1, 501.9
(11) a1, 542.0 (4) a1, 884.2 (356) a1, 936.9 (53) a1,
939.7 (107) a1, 1792.5 (242) a1, 2183.4 (57) a1,
2236.1 (73) a1, 2248.5 (76) a1
Background: Simple silicon hydrides are of great interest for
several reasons. They play an important role in silicon
chemical vapor deposition (CVD) processes, which are of
significance to the semiconductor industry. A detailed study
on the reaction of silicon atoms with silane (7) will help to
understand the mechanisms of these reactions. Another
appeal is that starting with the first isolation of a disilene by
West et al.[6] a new chapter in silicon chemistry was opened.
But the isolation and identification of the parent disilene is
still missing. Last but not least silicon hydrides are excellent
target molecules to demonstrate the unique bonding charac-
teristics of silicon compared to carbon, resulting very often in
surprising ™bridged∫ structures. These fascinating aspects
explain the numerous experimental[7] and theoretical[8] studies
covering silicon hydrides SiHn and Si2Hn.
TS10,11 C1
TS11,12 C1
1A1
1A1
À 320.1 (35) a1, 406.9 (13) a1, 457.1 (8) a1, 521.0
(2) a1, 703.9 (54) a1, 832.1 (178) a1, 877.6 (40) a1,
959.1 (96) a1, 1946.1 (72) a1, 2042.7 (191) a1,
2211.7 (114) a1, 2231.0 (118) a1
À 630.9 (4) a1, 331.8 (11) a1, 457.2 (3) a1, 508.2
(11) a1, 651.3 (8) a1, 682.9 (15) a1, 940.7 (87) a1,
1001.9 (204) a1, 1854.3 (94) a1, 2070.1 (140) a1,
2153.8 (171) a1, 2203.1 (128) a1
The global minimum is disilene (12). Already twenty years
ago[8o,p] the trans-bent geometry was suggested. The stabiliza-
tion energy compared with the two components triplet silicon
atom and silane (7) is 43.2 kcalmolÀ1. Besides 12 three other
isomers 10, 11, and 13, which should be formed in exothermic
3
reactions from Si atoms and 7, can be expected as reaction
products (Scheme 2). In the earlier papers the discussion was
concentrated on disilene (12) and silylsilylene (10). Dibridged
isomers of type 13 were for the first time mentioned by
Trinquier.[8i] According to our calculations the isomer 11 with
a monobridged structure has also to be taken into account. On
the triplet energy hypersurface the starting components form
without an activation barrier a loose complex 8. Its stabiliza-
Calculations: For the structural identification of the expected
species it was necessary to obtain the calculated vibrational
spectra. In addition to the earlier theoretical treatments[8] we
carried out some calculations on our own using the Gaussian
package of programs.[9] To get an overview of the Si2H4
potential energy surface (PES), several stationary points
Chem. Eur. J. 2002, 8, No. 19
¹ 2002 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/02/0819-4385 $ 20.00+.50/0
4385

G. Maier et al.
FULL PAPER
Scheme 2. Calculated (B3LYP/6 311 ‡ G**) relative energies and geometries of some stationary points on the H4Si2 PES; zero-point vibrational energies
included. Distances are given in ä and angles in degrees.
tion energy amounts to 5 kcalmolÀ1. In a subsequent step the
primary complex 8 is transformed (activation energy of
5.9 kcalmolÀ1) into triplet silylsilylene (9), followed by
intersystem crossing to singlet silylsilylene (10) (T/S gap
14.1 kcalmolÀ1). On the singlet energy hypersurface now
isomerization of 10 to the thermodynamically more stable
(DE ˆ 5.7 kcalmolÀ1) 12 takes place. The monobridged iso-
mer 11 is only 1.2 kcalmolÀ1 higher in energy than 10 and
13.4 kcalmolÀ1 more stable than the dibridged species 13. Two
calculated transition structures TS10,11 and TS11,12 are
found on the isomerization of 10 to 12. The barriers leading
from 10 to the monobridged species 11 (1.8 kcalmolÀ1) and
from 11 to 12 (6.6 kcalmolÀ1) are remarkably low. For
comparison: In case of the transformation of 2 to 3 the
calculated barrier amounts to 33.8 kcalmolÀ1.
Matrix experiments: A gaseous
mixture of silane (7) and argon
was deposited together with
silicon atoms, generated by re-
sistive heating of a silicon rod to
a temperature of about 13808C,
onto a CsI window at 10 K.
FTIR and UV/Vis spectra were
taken of the matrices. Besides
the very intense absorptions of
unreacted SiH4 and small bands
of unavoidable side products
(SiN2,[2h] SiH2 (15), Si2H2
(14),[2a] Si2H6) additional ab-
sorptions in the spectral range
around
900
and
1950
2200 cmÀ1 were registered.
Some of these bands overlap-
ped with the strong absorptions
of the starting material SiH4.
The additional new bands could
4386
¹ 2002 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/02/0819-4386 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, No. 19

Reaction of Si Atoms with Methane and Silane
4383 4391
Table 3. Theoretical (B3LYP/6 311 ‡ G**) and experimental (Ar, 10 K)
IR absorptions of disilene (12).
be revealed by producing difference spectra during subse-
quent irradiation of the matrices with different wavelengths
(Figures 2 and 3). The analysis of the spectra (see below)
shows that the reaction of silicon atoms with silane (7) leads to
a mixture of silylsilylene (10) and disilene (12). Upon
irradiation of the matrix with long wavelengths (l > 570 nm)
10 is isomerized to 12. The back-reaction can be enforced by
using shorter wavelengths. With l ˆ 334 nm 12 regenerates 10.
Parallel to this isomerization partial dehydrogenation of 12 is
initiated and one observes the bands of disilyne (14)[2a] with its
typical butterfly-type geometry.[10] In addition a single band at
1355.4 cmÀ1 appears, which we tentatively attribute to the
dibridged isomer 13. If wavelengths longer than 310 nm
(especially 505 nm) are used the hydrogen attached to disilyne
(14) in the same matrix cage is recaptured again (as products
could be identified 12 (UV spectroscopy) and 10 (IR
spectroscopy)).
Symmetry
Mode
ncalcd
[cm-1]
Icalcd
[km mol-1]
nexp
[cm-1]
n4
n7
n12
n6
n3
n9
n11
n2
n10
n1
n8
n5
ag
au
bu
au
ag
bg
bu
ag
bu
ag
bg
au
ds bend
das rock
das bend
das twist
ns Si-Si str.
das rock
das scissor
ds scissor
nas SiH str.
ns SiH str.
nas SiH str.
nas SiH str.
324.2
348.0
446.1
524.1
562.1
615.7
919.5
955.8
2224.5
2228.9
2245.7
2257.6
(0.0)
(22.9)
(31.6)
(0.2)
(0.0)
(0.0)
(182.2)
(0.0)
(113.5)
(0.0)
(0.0)
(150.0)
904.3
(s)
2180.2
(w)[a]
2207.8
(w)[a]
[a] Additional IR absorptions at 2163.2, 2191.0, and 2212.3 cmÀ1 of a second
matrix site were also observed.
Based on the experimental and theoretical findings the
mechanism of the reaction of 3Si atoms with silane (7) can be
summarized as follows: Via the triplet complex 8 triplet
silylsilylene (9) is formed. Both are too short-lived to be
detected. Intersystem crossing gives singlet silylsilylene (10).
The reaction leading to 10 delivers enough energy to surpass
the two barriers on the pathway from 10 to 12 even at 10 K.
Dependent on the wavelength mutual interconversions with
the third Si2H4 isomer 13 and disilyne (14) can be photo-
initiated.
Another observation is of importance in relation to the
standard procedure to prepare highly hindered disilenes, that
is the dimerization of the corresponding silylenes.[6] We found
no indication that the parent silylene (13), which is present in
the matrix–or can be generated independently in high yield
by cocondensation of silicon atoms with hydrogen–dimerizes
upon annealing of the matrix (30 35 K) to disilene (12).
by us before for the SiH stretching vibrations of 3.[3c]
Additional structural proof for 12 stems from the measured
isotopic shifts in the tetradeuterated isotopomer [D4]12
(Table 4, Figure 3). An even better help in the elucidation of
Table 4. Theoretical (B3LYP/6 311 ‡ G**) and experimental (Ar, 10 K)
IR absorptions of [D4]disilene ([D4]12).
Symmetry
Mode
ncalcd
[cm-1]
Icalcd
[km mol-1]
nexp
[cm-1]
n7
n4
n12
n6
n9
n3
n11
n2
n10
n1
n8
n5
au
ag
bu
au
bg
ag
bu
ag
bu
ag
bg
au
das rock
ds bend
das bend
das twist
248.5
255.8
326.9
371.5
468.9
513.4
665.0
713.2
1594.3
1600.2
1629.2
1638.4
(11.6)
(0.0)
(16.7)
(0.1)
(0.0)
(0.0)
(94.6)
(0.0)
(63.3)
(0.0)
(0.0)
(80.5)
das rock
ns Si-Si str.
das scissor
ds scisssor
nas SiD str.
ns SiD str.
nas SiD str.
nas SiD str.
658.3
(s)
1573.4
(m)
Spectroscopic identification: Recent efforts in the experi-
mental search for Si2H4 species resulted in the observation
that compounds of this elemental composition can exist.
Neutral Si2H4 was first detected as a transient species in
photoionization mass spectrometric studies.[7d] Quite recently
Stafast et al.[7b] showed that Si2H4 compounds are intermedi-
ates in the laser chemical vapor deposition of silane. Koshi
et al.[7a] were able to observe Si2H4 species produced in the
thermal decomposition of silanes by using time-of-flight mass
spectrometry coupled with VUV laser photoionization.
In our opinion the structural elucidation of disilene 12 and/
or its isomers was still missing. Matrix isolation spectroscopy
is the method of choice to fill this gap.
1614.7
(m)
disilene (12) can be expected from a study of the less
symmetrical trideuterated isotopomer [D3]12, in which even
the band for the Si Si stretching vibration should be IR-
active (Table 5). Effectively, this band was too low in intensity
to be identified in the complex experimental spectrum.
In the spectrum of silylsilylene (10) all vibrations are IR-
active (Table 6). A characteristic feature of silylenes with a
hydrogen atom at the divalent silicon is the low frequency
(1900 2010 cmÀ1) of the SiH stretching vibration.[3a] The
sensitivity of these species to matrix effects is notorious. For
instance, methylsilylene (2) shows remarkable splittings in the
SiH stretching range. Analogously, silylsilylene (10) reveals
three SiH stretches at 1963, 1974, and 2013 cmÀ1(Table 6,
Figure 2), originating from different matrix sites. The normal
SiH valence stretch vibrations of the SiH3 group could not be
registered as nicely, since they overlapped with the strong
absorptions of unreacted silane (7). A very good correlation
between the calculated and experimental value exists for the
second dominant band, which has to be assigned to the SiH3
deformation vibration (calcd 868.1, exp 860.6 cmÀ1). An
Tables 3 8–together with Figures 2 and 3–summarize
the calculated and experimental IR spectra of the isomers 10,
12, and 13.
Because of the C2h symmetry of disilene 12 only three IR
bands with a sufficient intensity can be expected (Table 3) to
be detectable, two SiH stretching vibrations and one SiH2
wagging vibration. All three can be detected in the difference
spectrum (Figure 2). The agreement between the calculated
and experimental spectra is acceptable. The two SiH stretches
are additionally split by matrix effects. There are at least two
different matrix sites. Similar phenomena have been observed
Chem. Eur. J. 2002, 8, No. 19
¹ 2002 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/02/0819-4387 $ 20.00+.50/0
4387

G. Maier et al.
FULL PAPER
Figure 2. IR spectra of disilene (12) and silylsilylene (10): Top and bottom: Calculated at the B3LYP/6 311 ‡ G** level of theory. Middle: Difference spectrum
of the photoisomerization 12 ! 10 in argon at 10 K. Positive bands increase, negative decrease during the irradiation with light of wavelength l ˆ 334 nm.
analogous analysis is also possible for the isotopomer [D4]10
(Table 7, Figure 3). In case of [D3]10 with its Cs symmetry
three different positional isomers can be anticipated. Accord-
ingly, the experimental spectrum of [D3]10 was very compli-
cated and gave no additional information.
According to calculation (Table 8) the dibridged isomer 13
should show an intense asymmetric Si2H stretching vibration
at 1453.1 cmÀ1. In the experiment a corresponding band can
be observed at 1355.4 cmÀ1 (Figure 2). We are aware of the
fact that it is dangerous to argue on the basis of a single band.
Figure 3. IR spectra of [D4]disilene [D4]12 and [D4]silylsilylene [D4]10: Top and bottom: Calculated at the B3LYP/6 311 ‡ G** level of theory. Middle:
Difference spectrum of the photoisomerization [D4]12 ! [D4]10 in argon at 10 K. Positive bands increase, negative decrease during the irradiation with light
of wavelength l ˆ 334 nm. x: Hydroxysilylene.
4388
¹ 2002 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/02/0819-4388 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, No. 19

Reaction of Si Atoms with Methane and Silane
4383 4391
Table 5. Theoretical (B3LYP/6 311 ‡ G**) and experimental (Ar, 10 K)
IR absorptions of [D3]disilene ([D3]12).
Table 8. Theoretical (B3LYP/6 311 ‡ G**) and experimental (Ar, 10 K)
IR absorptions of dibridged isomer 13 and [D4]13.
Symmetry
Mode
ncalcd
[cm-1]
Icalcd
[km mol-1]
nexp
[cm-1]
Symmetry Mode
Isotop- ncalcd
omer
Icalcd
nexp
[cm-1] [km mol-1] [cm-1]
n12
n11
n10
n9
n8
n7
n6
n5
n4
n3
n2
n1
a1
a1
a1
a1
a1
a1
a1
a1
a1
a1
a1
a1
ds rock
ds bend
das bend
ds twist
262.8
267.5
341.0
425.1
490.4
545.2
688.9
833.3
1597.0
1615.0
1634.3
2245.8
(8.0)
(5.3)
(14.5)
(2.5)
(1.0)
(5.5)
(64.2)
(50.4)
(36.8)
(21.1)
(48.4)
(67.3)
n12 bu
das SiH
281.5
202.0
376.7
367.0
708.3 (16.8)
508.6 (8.3)
833.1 (155.5)
601.5 (77.1)
845.1
623.9
849.6
605.8
1287.7 (32.3)
926.9 (17.9)
1428.6
1018.2
1453.1 (987.2)
1041.9 (513.7)
1633.0
1157.8
2043.6
1470.3
(3.4)
(1.7)
(0.0)
(0.0)
D4
D4
D4
D4
D4
D4
D4
D4
D4
D4
D4
D4
n4 ag
n6 au
n11 bu
n3 ag
n8 bg
n5 au
n7 bg
n10 bu
n2 ag
n1 ag
n9 bu
ns Si-Si str.
dsoop SiH
das SiH
das rock
ns Si-Si str.
das scissor
ds scisssor
nas SiD str.
ns SiD str.
nas SiD str.
ns SiH str.
681.8
827.5
1575.5
1591.5
1609.9
2191.9
(s)
(m)
(w)
(w)
(w)
(m)
ds SiH
(0.0)
(0.0)
(0.0)
(0.0)
dsoop Si2H
ns Si2H str.
ns Si2H str.
nas Si2H str.
ns Si2H str.
ns SiH str.
nas SiH str.
(0.0)
(0.0)
Table 6. Theoretical (B3LYP/6 311 ‡ G**) and experimental (Ar, 10 K)
IR absorptions of silylsilylene (10).
1355.4 (vw)
999.9 (vw)
Symmetry
Mode
ncalcd
[cm-1]
Icalcd
[km mol-1]
nexp
[cm-1]
(0.0)
(0.0)
(0.0)
(0.0)
n12
n8
n11
n7
n6
n5
n4
n10
n3
n2
n9
n1
a''
a'
a''
a'
a'
a'
a'
a''
a'
a'
a''
a'
das twist
67.9
369.1
389.0
428.9
717.1
868.1
932.8
957.2
2031.3
2180.5
2189.5
2217.5
(12.0)
(8.5)
ns Si-Si str.
das SiH def.
ds SiH def.
ds SiH def.
ds SiH3 def.
ds SiH3 def.
das SiH3 def.
ns SiH str.
ns SiH str.
nas SiH str.
ns SiH str.
(33.1)
(27.1)
(63.4)
(224.1)
(73.3)
(36.2)
(208.3)
(77.6)
(112.4)
(219.4)
2060.9 (453.7)
1482.3 (239.5)
707.3
860.6
927.2
(w)
(s)[a]
(w)
(w)
(s)[b]
(m)
(m)
(m)
947.5
Additional support for the structural identification of 10
and 12 comes from the UV-spectroscopic measurements.
Calculations predict for 12 a very strong UV transition (1Bu
1Ag) at 322 nm. The experimental value is 329 nm and fits very
well to that of other disilenes substituted by four small alkyl
groups (Me or Et: 345 nm[6b]). As expected, the characteristic
feature for 10 is a long wavelength absorption of low intensity.
The difference spectrum taken during the photoisomerization
[D4]10 ! [D4]12 with light of wavelenth l > 570 nm (Figure 4)
is in accordance ((1A'' 1A'); l ˆ 582 nm) with theory (l ˆ
669 nm). In a similar manner the difference spectrum of the
photoisomerization 14 ‡ H2 ! 12 ‡ 10 with light of the
wavelength l ˆ 405 nm (Figure 5) can be regarded as a strong
hint for the postulated reaction. The agreement between the
calculated (408 nm) and experimental (409 nm) absorption of
disilyne (14) is remarkable.
1963.3
2111.9
2153.2
2166.7
[a] Additional IR absorptions (other matrix sites) were observed at 868.8
(m) and 879.5 (m)cmÀ1. [b] Additional IR absorptions (other matrix sites)
were observed at 1974.5 (m) and 2013.2 (w)cmÀ1
.
Table 7. Theoretical (B3LYP/6 311 ‡ G**) and experimental (Ar, 10 K)
IR absorptions of [D4]silylsilylene ([D4]10).
Symmetry
Mode
ncalcd
[cm-1]
Icalcd
[km mol-1]
nexp
[cm-1]
n12
n11
n8
n7
n6
n5
n4
n10
n3
n2
n9
n1
a''
a''
a'
a'
a'
a'
a'
a''
a'
a'
das twist
49.5
282.7
290.4
378.3
533.3
641.7
671.3
686.2
1461.2
1554.6
1581.2
1599.1
(6.2)
(17.8)
(10.0)
(16.4)
(43.3)
(76.9)
(62.7)
(20.1)
(106.9)
(40.3)
(65.5)
(75.6)
das SiD def.
ds SiD def.
ns Si-Si str.
ds SiD def.
ds SiD3 def.
ds SiD3 def.
das SiD3 def.
ns SiD str.
ns SiD str.
nas SiD str.
ns SiD str.
526.7
(w)
(s)
642.4
[a]
681.3
1427.7
1529.5
1552.1
1581.2
(w)
Conclusion
(s)[b]
(w)
The reactions of silicon atoms in an argon matrix at 10 K with
methane (1) and silane (7) belong to two completely different
worlds. With methane (1) no thermal reaction occurs and the
insertion of the Si atoms into the C H bond needs photo-
activation. In contrast, insertion into the Si H bond of silane
a''
a'
(w)[c]
(w)[c]
[a] Hidden underneath SiD4 def. absorptions. [b] Additional absorptions
(matrix effect) were observed at 1434.0, 1436.0, 1448.0, and 1462.3 cmÀ1
[c] Broad signal.
3
À
.
À
(7) takes place without an activation barrier and a mixture of
silylsilylene (10) and disilene (12) can be detected.
However, for [D4]13 the same kind of band with the expected
isotopic shift (calcd 1041.9, exp 999.9 cmÀ1) is observed
(Table 8, Figure 3). This observation can be taken as an
indication that the structural assignment might be correct. For
[D3]13 both types of absorption should occur. But owing to
the small intensities of these bands an experimental verifica-
tion was not possible.
Experimental Section
The cryostat for matrix isolation was a closed-cycle compressor unit RW2
with coldhead base unit 210 and extension module ROK from Leybold.
The matrix window was CsI and the spectrometer was an IFS85 FTIR
Chem. Eur. J. 2002, 8, No. 19
¹ 2002 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/02/0819-4389 $ 20.00+.50/0
4389

G. Maier et al.
FULL PAPER
instrument from Bruker. The light sour-
ces used were a mercury high-pressure
lamp (HBO200 from Osram) with a
monochromator (Bausch and Lomb)
and a mercury low-pressure spiral lamp
with a Vycor filter (Gr‰ntzel).
For the production of silicon atoms a rod
with the dimensions 0.7 Â 2 Â 22 mm
was cut out from a highly doped silicon
wafer and heated resistively by using a
current of 10 A at 10 V. Under these
conditions the surface temperature
amounted to 1350 13808C.
Silane (7) was purchased from Aldrich
(puriss. 99.99%) and was used without
further purification.
[D4]Silane ([D4]7) was prepared by
treatment of tetrachlorosilane with lith-
ium aluminum tetradeuteride by using a
slightly modified version of the proce-
dure described in literature[11]
:
A
Schlenk tube, filled with a solution of
LiAlD4 (210 mg, 5.00 mmol; Aldrich
98%-D) in dry methyl tert-butyl ether
(25 mL) was connected to a vacuum line
and degassed by four freeze pump
thaw cycles. The degassed solution was
frozen (À1968C) again, and freshly
distilled and degassed SiCl4 (850 mg
5.00 mmol) was condensed onto it. The
reaction mixture was allowed to warm
up slowly. At about 08C a lively evolu-
tion of gas was observed. The reaction
products were condensed into a second
trap (À1968C). The trap, containing the
crude [D4]silane, was then warmed up to
À1208C, and the silane was condensed
into a third trap at À1968C to remove
small traces of the solvent. By repeating
this procedure two times, pure SiD4
([D4]7) (110 mL, 180 mg, 4.98 mmol)
was isolated (99% yield).
Figure 4. UV/Vis spectra of [D4]disilene ([D4]12) and [D4]silylsilylene ([D4]10): Difference spectrum of the
photoisomerization [D4]10 ! [D4]12 in argon at 10 K. Positive bands increase, negative bands decrease during
the irradiation with light of wavelength l > 570 nm.
In the same manner [D3]silane ([D3]7)
was prepared by treatment of trichloro-
silane with lithium aluminum tetradeu-
teride: HSiD3 (175 mg, 4.98 mmol, 99%
yield) was isolated from the reaction
of HSiCl3 (670 mg, 5.00 mmol) with
LiAlD4 (210 mg) in methyl tert-butyl
ether (25 mL).
Acknowledgements
Support by the Deutsche Forschungsge-
meinschaft and the Fonds der Chem-
ischen Industrie is gratefully acknowl-
edged. We thank the Wacker-Chemie
GmbH, Burghausen, for a gift of silicon
wafers.
[1] Presented as a Poster during the
European Silicon Days in Munich,
September 6 7, 2001.
[2] a) G. Maier, H. P. Reisenauer, A.
Meudt, H. Egenolf, Chem. Ber./
Recueil 1997, 130, 1043 1046;
b) G. Maier, H. P. Reisenauer, H.
Egenolf in Organosilicon Chemis-
Figure 5. UV/Vis spectra of disilene (12) and disilyne (14): Difference spectrum of the photoisomerization 14
‡ H2 ! 12 ‡ 10 in argon at 10 K. Positive bands increase, negative bands decrease during the irradiation with
light of wavelength l ˆ 405 nm.
4390
¹ 2002 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/02/0819-4390 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, No. 19

Reaction of Si Atoms with Methane and Silane
4383 4391
try III :From Molecules to Materials (Eds.: N. Auner, J. Weis), VCH,
Weinheim, 1998, pp. 31 35; c) G. Maier, H. P. Reisenauer, H.
Egenolf, J. Glatthaar, Eur. J. Org. Chem. 1998, 1307 1311; d) G.
Maier, H. P. Reisenauer, H. Egenolf, Eur. J. Org. Chem. 1998, 1313
1317; e) G. Maier, H. P. Reisenauer, H. Egenolf, Monatshefte f¸r
Chemie 1999, 130, 227 235; f) G. Maier, H. P. Reisenauer, H. Egenolf,
Organometallics 1999, 18, 2155 2161; g) G. Maier, H. P. Reisenauer,
H. Egenolf in Organosilicon Chemistry IV :From Molecules to
Materials (Eds.: N. Auner, J. Weis), Wiley-VCH, Weinheim, 2000,
pp. 64 69; h) G. Maier, H. P. Reisenauer, J.Glatthaar, Organometal-
lics 2000, 19, 4775 4783; i) summary: G. Maier, A. Meudt, J. Jung, H.
Pacl in The chemistry of organic silicon compounds, Vol. 2 (Eds.: Z.
Rappoport, Y. Apeloig), Wiley, New York, 1998, Chapter 19,
pp. 1143 1185; k) G. Maier, H. P. Reisenauer, H. Egenolf, J. Glat-
thaar in Organosilicon Chemistry V : From Molecules to Materials,
Wiley-VCH, Weinheim, in press.
M¸ller, J. Am. Chem. Soc. 1995, 117, 5363 5364; e) M C. Ernst, A. F.
Sax, J. Kalcher, Chemical Physics Lett. 1993, 216, 189 193; f) L. A.
Curtiss, K. Raghavachari, P. W. Deutsch, J. A. Pople, J. Chem. Phys.
1991, 95, 2433 2444; g) A. F. Sax, J. Kalcher, J. Phys. Chem. 1991, 95,
1768 1783; h) J. A. Boatz, M. S. Gordon, J. Phys. Chem. 1990, 94,
7331 7323; i) G. Trinquier, J. Am. Chem. Soc. 1990, 112, 2130 2137;
j) J.-P. Malrieu, G. Trinquier, J. Am. Chem. Soc. 1989, 111, 5916 5921;
k) G. Trinquier, J.-P. Malrieu, J. Am. Chem. Soc. 1987, 109, 5303 5315;
l) M. S. Gordon, T. N. Truong, E. K. Bonderson, J. Am. Chem. Soc.
1986, 108, 1421 1427; m) P. Ho, M. E. Coltrin, J. S. Binkley, C. F.
Melius, J. Phys. Chem. 1986, 90, 3399 3406; n) H. J. Kˆhler, H.
Lischka, J. Am. Chem. Soc. 1982, 104, 5884 5889; o) H. Lischka, H.-J.
Kˆhler, Chemical Physics Lett. 1982, 85, 467 471; p) R. A. Poirier,
J. D. Goddard, Chemical Physics Lett. 1981, 80, 37 41.
[9] Gaussian 98, Revision A. 7, M. J. Frisch, G. W. Trucks, H. B. Schlegel,
G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A.
Montgomery, Jr., R. E. Stratmann, J. C. Burant, S. Dapprich, J. M.
Millam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi,
V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo,
S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui, K.
Morokuma, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B.
Foresman, J. Cioslowski, J. V. Ortiz, A. G. Baboul, B. B. Stefanov, G.
Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R. L.
Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A.
Nanayakkara, C. Gonzalez, M. Challacombe, P. M. W. Gill, B.
Johnson, W. Chen, M. W. Wong, J. L. Andres, C. Gonzalez, M.
Head-Gordon, E. S. Replogle, J. A. Pople, Gaussian, Inc., Pittsburgh
PA, 1998.
[3] a) G. Maier, G. Mihm, H. P. Reisenauer, D. Littmann, Chem, Ber.
1984, 117, 2369 2381 (in Table 1 the wavenumbers 2035 and
2014 cmÀ1 for silaethene in a N2 matrix have to be corrected to 2235
and 2214 cmÀ1); b) G. Maier, G. Mihm, H. P. Reisenauer, Chem. Ber.
1984, 117, 2351 2368; c) R. Zetzmann, Dissertation, Universit‰t
Giessen, 2000.
[4] P. S. Skell, P. W. Owen, J. Am. Chem. Soc. 1967, 89, 3933 3934.
[5] S. Sakai, J. Deisz, M. S. Gordon, J. Phys. Chem., 1989, 93, 1888 1893.
[6] a) R. West, J. Fink, J. Michl, Science 1981, 214, 1343; b) summary: G.
Raabe, J. Michl in The Chemistry of Organic Silicon Compounds
(Eds.: S. Patai, Z. Rappoport), Wiley, New York, 1989, Part 2,
pp. 1015 1142.
[7] a) K. Tonokura, T. Murasaki, M. Koshi, Chemical Physics Lett. 2000,
319, 507 511; b) F. Falk, G. Mollekopf, H. Stafast, Appl. Phys. A,
1998, 67, 507 512; c) B. Ruscic, J. Berkowitz, J. Chem. Phys. 1991, 95,
2407 2415; d) B. Ruscic, J. Berkowitz, J. Chem. Phys. 1991, 95, 2416
2432.
[8] a) C. Pak, J. C. Rienstra-Kiracofe, H. F. Schaefer, III, J. Phys. Chem.
A, 2000, 104, 11232 11242; b) M. T. Swihart, R. W. Carr, J. Phys.
Chem. A, 1998, 102, 785 792; c) G. Katzer, M. C. Ernst, A. F. Sax, J.
Kalcher, J. Phys. Chem. A, 1997, 101, 3942 3958; d) Y. Apeloig, T.
[10] M. Bogey, H. Bolvin, C. Demuynck, J.-L. Destombes, Phys. Rev. Lett.
1991, 66, 413 416.
[11] a) D. Norman, W. A. Yolly, J R. Webster, Inorg. Synth. 1968, 11, 170;
b) N. Auner, U. Klingebiel in Synthetic Methods of Organometallic
and Inorganic Chemistry (Ed.: W. A. Herrmann), Thieme, Stuttgart,
1996, pp. 144 145.
Received: February 22, 2002 [F3898]
Chem. Eur. J. 2002, 8, No. 19
¹ 2002 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/02/0819-4391 $ 20.00+.50/0
4391