Mercury-sensitized photolysis of Me2SiH2 the disproportionation reactions of the Me2SiH radical

Article abstract of DOI:10.1039/a605762d

Mercury-sensitized photolysis of Me₂SiH₂ yields in the primary step an Me₂HSi radical and an H atom with a quantum yield of one. The Me₂HSi radicals undergo a combination reaction [k(2)] as well as two kinds of disproportionate reactions leading to dimethylsilylene [k(3)] and 1-methylsilaethene [k(4)]. The following branching ratios have been determined: k(3)/[k(2) + k(3) + k(4)] = 0.64 ± 0.10 and k(4)/[k(2) + k(3) + k(4)] ≥ 0.007. For Me₃Si radicals the branching ratio for disproportionation was also determined and a value of 0.063 was obtained.

Full text of DOI:10.1039/a605762d

View Article Online / Journal Homepage / Table of Contents for this issue
Mercury-sensitized photolysis of Me SiH
2
2
The disproportionation reactions of the Me SiH radical
2
C. Kerst and P. Potzinger
Ž
Max Planck Institut f ur Stromungsforschung, 37073 Gottingen, Germany
Mercury-sensitized photolysis of Me SiH yields in the primary step an Me HSi radical and an H atom with a quantum yield of
2
2
2
one. The Me HSi radicals undergo a combination reaction [k(2)] as well as two kinds of disproportionation reactions leading to
2
dimethylsilylene [k(3)] and 1-methylsilaethene [k(4)]. The following branching ratios have been determined: k(3)/
[k(2) ] k(3) ] k(4)] \ 0.64 ^ 0.10 and k(4)/[k(2) ] k(3) ] k(4)] P 0.007. For Me Si radicals the branching ratio for dispro-
3
portionation was also determined and a value of 0.063 was obtained.
The bimolecular reaction of two Me Si radicals has been the
subject of a great number of studies.1h5 Of special interest is
FID detection of the photolysis products were assumed to be
proportional to the number of carbons in the product in ques-
tion.20 Most of the photolysis products were identiÐed by
coupled gas chromatography mass spectrometry (HP 5971A).
The computer program KINAL21 was used to simulate the
but-2-yne and MeOH experiments.
3
the extent of disproportionation which leads to an SixC
double bonded species. The published values for the rate con-
stant ratio of disproportionation to combination varies over a
large range3,6h12 and only recently has the ratio determined
by di†erent groups converged.13,14 For the Me HSi radical a
2
further disproportionation path is possible, i.e., the formation
Results
of silylene. The extent of such a reaction has been studied for
The mercury-sensitized photolysis of dimethylsilane yielded
six products: H ; one disilane, HMe SiSiMe H (HSI2H);
SiH .15 In this paper we report the branching ratio of the
3
di†erent self-reaction channels of the Me SiH radical.
2
2
2
2
The Me SiH radical was generated by mercury-sensitized
2
three trisilanes, of which two have been identiÐed
HMe SiSiMe SiMe H (HSI3H) and HMe SiSiH(Me)-
photolysis of Me SiH , which has been studied by Nay et
2
2
2
2
tetrasilane,
2
2
CH SiMe H
(HSISIHCSIH);
Me HSi(SiMe ) SiMe H (HSI4H). At higher conversion a
pentasilane, which we tentatively assign the following struc-
ture Me HSi(SiMe ) SiMe H (HSI5H), was also detected. In
the presence of additives like NO and MeOH additional pro-
ducts were detected.
and
one
al.16 They reported two main products,
H
and
2
2
2
Me HSiSiMe H, with quantum yields of 0.81 and 0.71, as well
as a trimer with the formula Me HSiSiMe SiMe H. The pro-
2
2 2
2
2
2
2
2
2
ducts were explained by an SiwH bond splitting in the
primary process followed by H abstraction and radical com-
bination. The trimer was thought to be the result of a second-
ary reaction. In this paper we give a more detailed account of
the mechanism and also report evidence for secondary pro-
cesses, which are important at rather small conversion levels.
2
2 3
2
The time dependence of the photolysis products is shown in
Fig. 1. The photoproducts H , HSI2H and HSISIHCSIH
2
show the typical behaviour of a primary product that is
destroyed by secondary processes at higher conversion or one
whose rate of formation is diminished by a polymer coating of
the light entrance window. The plots of concentration of
HSI3H and HSI4H vs. time are S-shaped, indicating that both
depletion and formation are due to secondary processes. For
the primary products the number of molecules formed per
cm3, n, as a function of time, t, was Ðtted to the expression
n \ aI t ] b(I t)2. I represents the number of photons
Experimental
The experimental details concerning gas handling, actinom-
etry, photolysis conditions, and end-product analysis were by
and large the same as described previously,14 therefore only a
short account will be given. Static photolyses were performed
at room temperature and carried out in a cylindrical quartz
cell attached to a conventional vacuum line. The light source
was a thermostatted low-pressure mercury lamp, operating in
dc mode. All photolysis experiments were performed with a
mercury concentration of 3.6 ] 1013 cm~3. The light intensity
absorbed by the Hg atoms was determined by two actinom-
eters: Hg-sensitized cis-butene to trans-butene conversion
abs
abs
abs
absorbed per unit time and unit volume. The coefficient a is
taken equal to the quantum yield. From the data shown
in Fig.
1
the following quantum yield values are
derived /(H ) \ 0.92 ^ 0.01, /(HSI2H) \ 0.75 ^ 0.01 and
2
/(HSISIHCSIH) \ (6 ^ 1) ] 10~3. If one performs the
same experiment at a higher Me SiH concentration, a sub-
2
2
stantial increase in the quantum yields of the primary prod-
[/(trans-butene) \ 0.517,18] and Hg-sensitized N O photoly-
ucts /(H ) \ 0.98 ^ 0.04, /(HSI2H) \ 0.98 ^ 0.04 and
2
2
sis [/(N ) \ 1.019].
/(HSISIHCSIH) \ (7 ^ 1) ] 10~3 (Fig. 2) is observed. The
2
All substances were of commercial origin and of the highest
dependence of the quantum yields on Me SiH pressure is
2
2
purity available. The gas chromatographic purity of Me SiH
shown in Fig. 3; in these experiments the ratio of substrate
pressure to photolysis time was kept constant. The quantum
yield of HSISIHCSIH decreases with increasing Me SiH
2
2
was determined to be 99.6%. Special care had to be exercised
in our experiments with MeOH. The total duration of the
experiment, mixing times, temperatures in the freezeÈthaw
cycles and in the analysis, etc., had to be kept constant to
obtain reproducible results with MeOH as the trapping agent.
End-product analyses were undertaken by gas chromatog-
raphy (HP 5980, fused silica capillary column OV1). C H
2
2
pressure, while /(HSI4H) seems to increase. Variation of the
light intensity (Fig. 4) has a signiÐcant e†ect on the quantum
yield of HSISIHCSIH only. Its dependence on the absorbed
intensity can be Ðtted to a function /(HSISIHCSIH) \
(aI )/(bI ] c), which provide a limiting high intensity
2
6
abs
abs
was used as an internal standard and the response factors for
quantum yield of /(HSISIHCSIH) B 7 ] 10~3.
J. Chem. Soc., Faraday T rans., 1997, 93(6), 1071È1078
1071

View Article Online
Fig. 1 Dependence of the product concentrations on exposure time in the Hg-sensitized photolysis of 1.60 ] 1018 cm~3 Me SiH ; I \ 5.69
2
2
abs
] 1013 cm~3 s~1
In the presence of a large excess of H , H atoms are formed
in the primary process, and the same products were observed
as above, with the exception of the unidentiÐed trimer. Again
HSI2H and HSISIHCSIH turned out to be of primary origin,
while, HSI3H and HSI4H were of secondary origin (Fig. 5).
The quantum yields derived from Fig. 5 are as follows:
NO was added as a radical scavenger. As can be seen from
Fig. 7, HSI2H is totally scavenged in the limit of high NO
concentration. All the other Si containing products disappear
completely at smaller NO concentrations. Besides the two
2
main products of the reactions with NO, Me HSiOH and
2
Me HSiOSiMe H (HSIOSIH), also shown in Fig. 7, a number
2
2
/(HSI2H/H ) \ 0.73 ^ 0.04,
/(HSISIHCSIH/H ) \ (3.6
of other unidentiÐed products of small concentration were
also observed.
2
2
^ 0.3) ] 10~3. / (product/H ) denotes the quantum yield of a
2
product in the presence of H . At longer irradiation times the
products behave as shown in Fig. 6. Under these experimental
MeOH was added to get information on the two dispro-
2
portionation pathways of the Me SiH radical, i.e., the forma-
2
conditions the decrease of Me SiH was also measured, and
tion of MeHSixCH and Me Si. For both intermediates
2
2
2
2
found to be /([Me SiH /H ) \ 0.86. In these experiments
Me HSiOMe is the expected product of the scavenging pro-
cesses. We measured the number of Me HSiOMe molecules
formed as a function of time for di†erent MeOH concentra-
2
2
2
2
we looked speciÐcally for higher silanes, but with the excep-
tion of a small amount of pentasilane no additional products
could be found.
2
tions. The slope of the straight lines was then plotted as a
Fig. 2 Dependence of the product concentrations on exposure time in the Hg-sensitized photolysis of 4.35 ] 1018 cm~3 Me SiH ; I \ 5.14
abs
2
2
] 1013 cm~3 s~1
1072
J. Chem. Soc., Faraday T rans., 1997, V ol. 93

View Article Online
Fig. 3 Dependence of the product quantum yields on substrate pressure in the Hg-sensitized Me SiH photolysis; [Me SiH ]/t \ const., con-
2
2
2
2
version B1%
function of [MeOH]. Exactly the same experiments were per-
close to one restricts the number of possible primary processes
formed for Me SiH (Fig. 8).
To check speciÐcally for the disproportionation reaction
to two: an SiwH bond split (I) and/or molecular H elimi-
3
2
nation (II)
leading to Me Si we added pentamethyldisilane and but-2-yne
as e†ective scavengers for silylene. The experiments with
2
Hg* ] Me SiH ] H ] Me HSi ] Hg
(I)
2
2
2
Me Si H failed because Me SiH abstracts hydrogen from
] H ] Me Si ] Hg
(II)
5
2
2
Me Si H quite efficiently.22 The experiments with but-2-yne
2
2
5
2
were hampered by radical addition to the triple bond. As can
be seen from Fig. 9, Me Si disappears just about as fast as
HSI2H.
Our experience with similar systems14,20 suggests that the
following reactions will take place in connection with primary
process (I)
6
2
Discussion
H ] Me SiH ] H ] Me SiH
(1)
(2)
(3)
2
2
2
2
Primary processes
2Me SiH ] Me HSiSiMe H
2
2
2
The observation of an almost perfect material balance, a 1 : 1
relation of the two main products and a quantum yield very
] Me Si ] Me SiH
2
2
2
Fig. 4 Dependence of the product quantum yields on light intensity in the Hg-sensitized photolysis of 1.62 ] 1018 cm~3 Me SiH ; I ] t \
abs
2
2
const., conversion B3%
J. Chem. Soc., Faraday T rans., 1997, V ol. 93
1073

View Article Online
Fig. 5 Dependence of the product concentrations on exposure time in the Hg-sensitized photolysis of 1.72 ] 1019 cm~3 H and 1.60 ] 1018
2
cm~3 Me SiH at low conversion; I \ 5.54 ] 1013 cm~3 s~1
2
2
abs
] MeHSixCH ] Me SiH (4)
MeHSiwCH SiH(Me)Cz H , which is formed in a fast bimolecu-
2 2
2
2
2
lar reaction,23 suggests that the stationary concentration of
Me Si ] Me SiH ] Me HSiSiMe H
(5)
the Me SiH radical must be high. Other arguments exist that
2
2
2
2
2
2
MeHSixCH ] 2Me SiH ] Me HSiSiH(Me)CH SiMe H
show the contribution of (II) to the overall mechanism must
2
2
2
2
2
be small. Addition of NO leads to an almost complete disap-
(6)
pearance of HSI2H (Fig. 7). If Me Si does not react faster
2
If primary process (II) occurs, it would be simply followed
by reaction (5). Primary process (I) together with reactions.
(1)È(6) is able to explain all our experimental Ðndings,
however, such is not the case for (II), even if we postulate a
third primary process (III)
with NO than SiH ,24 only 12% of the Me Si should be scav-
2
2
enged at the highest NO concentration.
In the presence of an excess of H , H atoms are formed in
2
the primary process followed by reactions (1)È(6). In this case
the quantum yields of the primary products are somewhat
reduced, possibly due to the involvement of HgH and
Hg* ] Me SiH ] H ] MeHSixCH ] Hg
(III)
Me HSiHgH,25 but the product ratios are pretty much the
2
2
2
2
2
The formation of the trimer Me HSiSiH(Me)SiMe H
same as in the sensitization of Me SiH . This is the strongest
2
2
2
2
cannot be explained in the absence of Me SiH radicals. Fur-
argument for the predominance of process (I) in the Hg-
sensitized photolysis of Me SiH .
2
thermore, the absence of the silaethylene dimer,
2
2
Fig. 6 Dependence of the product concentrations on exposure time in the Hg-sensitized photolysis of 1.17 ] 1019 cm~3 H and 2.40 ] 1017
2
cm~3 Me SiH at high conversion; I \ 2.88 ] 1013 cm~3 s~1
2
2
abs
1074
J. Chem. Soc., Faraday T rans., 1997, V ol. 93

View Article Online
Fig. 7 Dependence of the product concentrations on NO concentration in the Hg-sensitized photolysis of 2.40 ] 1017 cm~3 Me SiH ; I
\
2
2
abs
2.88 ] 1013 cm~3, t \ 60 s
At this point one should mention that there is no proof of
another primary process, such as (IV) which has been
comparison to (I).
observed with Me SiH14
Secondary processes
3
Hg* ] Me SiH ] H ] MeH SiCH ] Hg
(IV)
Quite unexpectedly, secondary products were formed with
very high efficiencies. At conversions as low as 0.1% HSI3H
was detectable. Such behaviour was not observed in studies of
Me Si,20 Me SiH14 and Me Si H,26 and must be attributed
2
2
2
2
Process (IV) could still be occuring if the silyl-substituted
methyl radical reacts with Me SiH exclusively by dispro-
portionation
2
4
3
5 2
to the availability of a weak SiwH bond in the disilane
product. Of the four major active intermediates Hg*, H,
Me HSi and Me Si, only Me HSi will be able to react with
MeH SiCH ] Me SiH ] Me SiH ] Me Si
(7)
2
2
2
2
2
2
Our experiments do not allow us to exclude (II), (III)
and (IV) completely, but they are certainly negligible in
2
2
2
the disilane product in the early stage of photolysis. From the
Fig. 8 InÑuence of the MeOH concentration on the amount of methoxysilane formed (a) in the Hg-sensitized photolysis of 1.68 ] 1019 cm~3 H
2
and 1.31 ] 1017 cm~3 Me SiH , I \ 2.39 ] 1013 cm~3 s~1: (a ) top line [MeOH] \ 1.57 ] 1016 cm~3, middle line [MeOH] \ 1.04 ] 1016
abs
cm~3, bottom line [MeOH] \ 0.53 ] 1016 cm~3; (a ) slope of [Me HSiOMe] vs. t plots vs. [MeOH]. (b) In the Hg-sensitized photolysis of
2
2
1
2
2
1.68 ] 1019 cm~3 H and 9.33 ] 1017 cm~3 Me SiH, I \ 2.39 ] 1013 cm~3 s~1: (b ) top line [MeOH] \ 2.27 ] 1016 cm~3, middle line
2
3
abs
1
[MeOH] \ 1.46 ] 1016 cm~3, bottom line [MeOH] \ 0.53 ] 1016 cm~3; (b ) slope of [Me SiOMe] vs. t plots vs. [MeOH].
2
3
J. Chem. Soc., Faraday T rans., 1997, V ol. 93
1075

View Article Online
and possibly a third one, the formation of disilene.
Me SiH ] Me HSiSiMe ] Me SiH ] Me SixSiMe
2
2
2
2
2
2
2
(11)
H atoms undergo not only abstraction reactions, but also sub-
stitution reactions,27,28 which will transform higher molecular
weight products to lower molecular weight products. Our
results, especially the low quantum yield for Me SiH con-
2
2
sumption, suggest that destruction of the higher silanes is
indeed an important process at high conversion. No product
of higher molecular weight than HSI5H was observed, and
HSI2H seems to reach a photostationary state.
Radical–radical reactions
Discussion of our attempts to determine the branching ratio
for the bimolecular self reaction of dimethylsilyl radicals is as
follows. From the quantum yields at low conversion and the
proposed reactions (1)È(6) we easily derive k(4)/
[k(2) ] k(3) ] k(4)] \ (7 ^ 1) ] 10~3. This value is only a
lower limit because of the possible occurrence of dispro-
portionation reactions: the formation of HSISIHCSIH is
more complicated than described by reaction (6). Addition of
an Me HSi radical to the SixC double bond may occur at
2
the Si or at the C end of the double bond. If the reactivity is
similar to the trimethylsilyl case,20 then preferential addition
at the Si end is predicted
MeHSixCH ] Me HSi ] Me HSiSiH(Me)CH
(12)
2
2
2
2
The silyl substituted methyl radical will react with another
Me HSi radical and the following three reactions can take
2
place:
Me HSiSiH(Me)CH ] Me HSi ] Me HSiSiHMe ] Me Si
2
2
2
2
2
2
(13)
] Me HSiSi(Me)xCH ] Me SiH
(14)
(15)
2
2
2
2
] Me HSiSiH(Me)CH SiMe H
2
2
2
From the results below and from our investigations of the
Fig. 9 InÑuence of the but-2-yne concentration on the amount of
disilane formed (a) in the Hg-sensitized photolysis of 1.53 ] 1019
cm~3 H and 1.70 ] 1018 cm~3 Me SiH; I \ 5.19 ] 1013 cm~3
reactions of Me HSi with Me Si 22 we can conclude that
2
5 2
reactions (13) and (14) are the most important of these three
reactions. The substituted silene formed in (14) will react
further to yield tri- or tetra-silanes. In both cases the tri- or
tetra-silanes are only a small fraction of the same products,
otherwise formed, and it is not possible to associate them with
2
3
abs
s~1, t \ 300 s; curve; simulation with k(19) \ 5.5 ] 10~16 cm3 s~1;
(b) in the Hg-sensitized photolysis of 1.50 ] 1019 cm~3 H and
2
1.73 ] 1018 cm~3 Me SiH ; I \ 6.15 ] 1013 cm~3 s~1, t \ 300 s;
2
2
abs
curve in (b ) simulation with k(3) \ 0.0 cm3 s~1; top curve in (b )
1
2
simulation with k(3) \ 1 ] 10~11 cm3 s~1, bottom curve in (b ) simu-
2
reaction (4). If the Me HSi radical adds to the C side of the
lation with k(3) \ 2 ] 10~11 cm3 s~1
2
double bond either the trimer, HSiSiHCSiH, or an isomeric
disilane, MeH SiCH SiHMe , should be formed. The isomer-
2
2
2
ic disilane was not observed. The decrease of /(HSISIHCSIH)
results in ref. 22 we know that silyl radicals are able to
abstract H atoms from the SiwH bond of disilanes
with increasing Me SiH pressure (Fig. 3) can be explained if
2
2
one assumes that the substituted methyl radical produced in
reaction (12) is able to abstract a hydrogen atom from
Me SiH .
The true branching ratio, k(4)/[k(2) ] k(3) ] k(4)], might be
larger than given above, but it would still be in agreement
with our observations made so far, that disproportionation
reactions leading to SixC double bonds are utterly unimpor-
tant in the self reaction of methylated silyl radicals.
Me SiH]Me HSiSiMe H]Me SiH ]Me HSiSiMe (8)
2
2
2
2
2
2
2
2
2
Reaction (8) competes only with reactions (2)È(4), and comes
therefore into play at low HSI2H concentrations. For Hg*, H
and Me Si the reactant HSI2H has to compete with
2
Me SiH , which is present in very large excess.
2
2
At higher conversion, H and Me Si will also contribute to
secondary product formation. Although reactions with Me Si
will always extend the Si chain length by one unit, this might
2
To determine the extent of reaction (3) we used a number of
2
scavengers whose rate constants with Me Si are known.
2
But-2-yne is a very efficient scavenger of Me Si,29 but unfor-
2
not necessarily be the case for H and Me HSi. The disilyl
2
radical formed in reaction (8) will react mainly with Me SiH,
2
tunately reacts with silyl radicals as well.30 To investigate the
the radical with the highest stationary concentration. Two
interference of silyl radicals we studied the system
reactions must be taken into account
HgÈH ÈMe SiHÈMeCCMe under the same conditions as the
2
3
dimethylsilane system. The product spectrum is complicated
and the Ðnal products are unknown. We therefore concen-
trated on the decrease of Me Si only. The error bars given in
Me SiH ] Me HSiSiMe ] HMe SiSiMe SiMe H
(9)
2
2
2
2
2
2
] Me Si ] Me HSiSiMe H (10)
2
2
2
6 2
1076
J. Chem. Soc., Faraday T rans., 1997, V ol. 93

View Article Online
Fig. 9 mirror our GC reproducibility and our estimates of
other error sources. We tried to simulate the decrease of
Me Si with increasing but-2-yne concentration by utilizing a
In Fig. 8 the Me HSiOMe concentration is plotted as a
function of t. Straight lines were obtained whose slopes as well
as intercepts depend on the MeOH concentration. Plotting
2
6
2
very simple mechanism [(16)È(20)].
the rate of Me HSiOMe formation as a function of [MeOH]
2
we get again a straight line, as predicted by the mechanism.
H ] Hg* ] 2H ] Hg
(16)
(17)
(18)
(19)
(20)
The slope of this straight line is given by
2
H ] Me SiH ] H ] Me Si
k(3)
k(24)
I
3
2
3
abs
\ (2.8 ^ 0.3) ] 10~5
k(2) ] k(3) ] k(4) k(5) [Me SiH ]
2Me Si ] Me Si
2
2
3
6 2
Dividing by I \ 2.4 ] 1013 cm~3 s~1 and the known rate
Me Si ] MeCCMe ] R
abs
3
constant ratio k(24)/k(5) \ 0.2,29,32,33 and multiplying by
Me Si ] R ] P
[Me SiH ] \ 1.3 ] 1017 cm~3 we obtain
3
2
2
We varied only k(19), since all other rate constants are
known, k(20) was estimated to be 5 ] 10~11 cm3 s~1. For
k(19) \ 5.5 ] 10~16 cm3 s~1 we Ðnd a minimum in the s2
values, and the line in Fig. 9 shows the calculated values. The
k(19) value found is appreciably lower than the absolute rate
constants for similar systems in the liquid phase.31
k(3)
\ 0.64 ^ 0.10
k(2) ] k(3) ] k(4)
This value depends very crucially on the rate constant ratio
k(24)/k(5). There also exists a value for k(24) in the literature,
k(24) \ 1.5 ] 10~11 cm3 s~1, measured in the liquid phase,34
which would reduce the branching ratio by more than a factor
of ten. Computer simulations show however that this high
value for k(24) cannot be reconciled with our data.
For the Me SiH system the mechanism consists of the reac-
2
tions (1), (2), (3), (5), (16) and
Me SiH ] MeCCMe ] R
(21)
(22)
(23)
The intercept is given by k(4)/[k(2) ] k(3) ] k(4)] ] I
\
2
abs
(3.0 ^ 3.3) ] 10~10 cm3 s~1.
R ] Me HSi ] P
2
Dividing by I
we obtain for the branching ratio, k(4)/
abs
Me Si ] MeCCMe ] P@
[k(2) ] k(3) ] k(4)] \ (1.25 ^ 1.30) ] 10~3. This rather impre-
cise value is lower than the value given above, but agrees
insofar that this analysis also yields a negligibly small value
for k(4).
2
The rate constants k(1), k(5) and k(23) are known, k(2)/k(3) is
varied subject to the condition k(2) ] k(3) \ k(18), all other
rate constants are taken from the Me SiH system. If we
If our analysis is correct, we expect for Me SiH the follow-
3
3
assume that no disproportionation is taking place in the
ing dependence of Me SiOMe on photolysis time
3
Me SiH system, the reduced s2 value is larger than in the
2
2
Me SiH system by a factor of 6. If all points in Fig. 9 are
k(28)
3
[Me SiOMe] \
I
t ] [Me SiOMe] (27)
taken into account we obtain a best Ðt for the branching ratio
3
abs
3
d
k(18) ] k(28)
k(3)/[k(2) ] k(3)] \ 0.33. Discarding the high but-2-yne value,
where reaction (28) runs as follows
2Me Si ] Me SixCH ] Me SiH
because in the Me SiH system a similar restricted concentra-
3
tion range was covered, we arrive at a value for k(3)/
(28)
[k(2) ] k(3)] \ 0.66 (Fig. 9). The but-2-yne experiments were
only taken as an indication that process (3) is actually taking
place.
A quantitative Ðgure for the branching ratio was expected
from the scavenging experiments with MeOH. MeOH is
usually applied for silaethylenes, but it is also a possible scav-
3
2
2
3
In deriving reactions (26) and (27) we have assumed that
even at the smallest MeOH concentration employed, silaethyl-
ene is completely scavenged by MeOH.
The independence of Me SiOMe formation on MeOH con-
centration is borne out by experiment (Fig. 8). From the inter-
3
enger for Me Si despite its much lower rate constant in com-
cept in Fig. 8 we arrive with the same value for I as above
2
abs
parison to but-2-yne, because no reactions with silyl radicals
k(28)/[k(18) ] k(28)] \ 0.063. This value is in very good agree-
are taking place. The rate constant for reaction
ment with our previous result.14
In our treatment of the data we have assumed that Me Si is
Me Si ] MeOH ] Me HSiOMe
(24)
2
2
2
formed in a direct reaction. There also exists another pathway
to Me Si, combination of two Me HSi radicals (2) and
decomposition of the vibrationally excited dimer. We reckon
that the contribution of this reaction path to the Me Si for-
has been determined32 and it holds true that under our
2
2
experimental conditions k(5)[Me SiH ] A k(24)[MeOH]. We
2
2
expect a dependence for Me HSiOMe formation on MeOH
2
2
mation can be largely excluded, due to the large number of
concentration, however, MeOH has one severe drawback: it
internal degrees of freedom of the dimer molecule and the
high pressure used in our experiments.
reacts with Me SiH in the dark to form Me HSiOMe.
With a mechanism comprising reactions (1)È(5), (16), (24)
2
2
2
and (25)
MeHSixCH ] MeOH ] Me HSiOMe
(25)
Conclusion
2
2
we arrive at the following expression for Me HSiOMe as a
function of photolysis time and MeOH concentration
The reaction of Hg (3P ) with Me SiH is characterized by an
2
1
2
2
abstraction reaction from the SiwH bond. Possibilities for
other processes occurring like abstraction from the CwH
[Me HSiOMe] \
bond or molecular H elimination have been examined but
2
2
could not be established, however, very small contributions
G
k(4)I
k(3)I
abs
k(24) [MeOH]
H
abs
]
t
can not be totally excluded. The quantum yield of the primary
process is one but decreases with increasing conversion. The
reason for this result is the high reactivity of the weak SiwH
bond of polysilanes towards all the reactive intermediates in
our system. Our results suggest that the secondary reactions
do not build up molecular weight products higher than penta-
silanes, but lead to a stationary state of the di-, tri- and tetra-
k(2) ] k(3) ] k(4) k(2) ] k(3) ] k(4) k(5) [Me SiH ]
2
2
] [Me HSiOMe]
(26)
2
d
where [Me HSiOMe]
methoxysilane formed in the dark, which, to a good approx-
is the amount of dimethyl-
2
d
imation, is independent of the photolysis time t.
J. Chem. Soc., Faraday T rans., 1997, V ol. 93
1077

View Article Online
16 M. A. Nay, G. N. C. Woodall, O. P. Strausz and H. E. Gunning,
J. Am. Chem. Soc., 1965, 87, 179.
17 G. R. De Mare, J. Photochem., 1977, 7, 101.
18 M. Termonia and G. R. De Mare, Chem. Phys. L ett., 1974, 25,
402.
silanes. This must mean that substitution reactions are
important.
The Me HSi radical undergoes two di†erent kinds of dis-
2
proportionation reactions: formation of a silaethylene and
formation of silylene. The latter is the most important step in
19 R. J. Cvetanovic, W. E. Falconer and K. R. Jennings, J. Chem.
the self-reaction of the Me SiH radical.
Phys., 1961, 35, 1225.
2
20 M. Ahmed, P. Potzinger and H. Gg. Wagner, J. Photochem. Pho-
tobiol. A., Chem., 1995, 86, 33.
References
21 T. Turanyi, Comput. Chem., 1990, 14, 253.
22 C. Kerst, P. Potzinger and H. Gg. Wagner, Z. Naturforsch. A,
Phys. Sci., 1996, 51, 102.
1
2
G. B. Watts and K. U. Ingold, J. Am. Chem. Soc., 1972, 94, 491.
P. P. Gaspar, A. D. Haizlip and K. Y. Choo, J. Am. Chem. Soc.,
1972, 94, 9032.
23 Th. Brix, N. L. Arthur and P. Potzinger, J. Phys. Chem., 1989, 93,
8193.
3
4
5
6
P. Cadman, G. M. Tilsey and A. F. Trotman-Dickenson, J.
Chem. Soc., Faraday T rans. 1, 1972, 68, 1849.
24 J. O. Chu, D. B. Beach, R. D. Estes and J. M. Jasinski, Chem.
Phys. L ett., 1988, 143, 135.
N. Shimo, N. Nakashima and K. Yoshihara, Chem. Phys. L ett.,
1986, 125, 303.
Th. Brix, E. Bastian and P. Potzinger, J. Photochem. Photobiol.
A., Chem., 1989, 49, 287.
O. P. Strausz, L. Gammie, G. Theorakoupolos, P. G. Meyer and
I. G. Csizmadia, J. Am. Chem. Soc., 1976, 98, 1622.
S. K. Tokach and R. D. Koob, J. Phys. Chem., 1979, 83, 774.
E. Bastian, P. Potzinger, A. Ritter, H. P. Schuchmann, C. von
Sonntag and G. Weddle, Ber. Bunsen-Ges. Phys. Chem., 1980, 84,
56.
25 C. Kerst, P. Potzinger and H. Gg. Wagner, J. Photochem. Photo-
biol. A, Chem., 1995, 92, 141.
26 C. Kerst, P. Potzinger and H. Gg. Wagner, Z. Naturforsch. A,
Phys. Sci, 1996, 51, 105.
27 T. L. Pollock, H. S. Sandhu, A. Jodhan and O. P. Strausz, J. Am.
Chem. Soc., 1973, 95, 1017.
7
8
28 L. Fabry, P. Potzinger, B. Reimann, A. Ritter and H. P. Steenber-
gen, Organometallics, 1986, 5, 1231.
29 J. E. Bagott, M. A. Blitz, H. M. Frey, P. D. Lightfoot and R.
Walsh, J. Chem. Soc., Faraday T rans. 2, 1988, 84, 515.
30 A. Alberti and G. Pedulli, Rev. Chem. Intermed., 1987, 8, 207.
31 C. Chatgilialoglu, K. U. Ingold and J. C. Sciano, J. Am. Chem.
Soc., 1971, 93, 1709.
9
L. Gammie, I. Safarik, O. P. Strausz. R. Robertge and C.
Sandorfy, J. Am. Chem. Soc., 1980, 102, 378.
10 S. K. Tokach and R. D. Koob, J. Am. Chem. Soc., 1980, 102, 376.
11 B. J. Cornett, K. Y. Choo and P. P. Gaspar, J. Am. Chem. Soc.,
1980, 102, 377.
32 J. E. Bagott, M. A. Blitz, H. M. Frey, P. D. Lightfoot and R.
Walsh, Int. J. Chem. Kinet., 1992, 24, 127.
12 D. J. Doyle, S. K. Tokach, M. S. Gordon and R. D. Koob, J.
Phys. Chem., 1982, 86, 3626.
33 J. E. Bagott, M. A. Blitz, H. M. Frey and R. Walsh, J. Am. Chem.
Soc., 1990, 112, 8337.
13 I. Safarik, A. Jodhan, O. P. Strausz and T. N. Bell, Chem. Phys.
L ett., 1987, 142, 115.
34 H. Shizuka, H. Tamaha, K. Tonokura, K. Murata, H. Hiratsuka,
J. Ohshita and M. Ishikawa, Chem. Phys. L ett., 1988, 143, 225.
14 C. Kerst, P. Potzinger and H. Gg. Wagner, J. Photochem. Photo-
biol. A., Chem., 1995, 90, 19.
15 B. Reimann, A. Matten, R. Laupert and P. Potzinger, Ber.
Bunsen-Ges. Phys. Chem., 1977, 81, 500.
Paper 6/05762D; Received 19th August, 1996
1078
J. Chem. Soc., Faraday T rans., 1997, V ol. 93