Selectivity and enormous H/D isotope effects on H atom abstraction by CH3 radicals in solid methylsilane at 3.0 K-115 K

Article abstract of DOI:10.1039/b206503g

An EPR study was carried out to elucidate the hydrogen atom abstraction from methylsilane (CH₃SiH₃) by a methyl radical, CH₃SiH₃ + ^.CH₃ → CH₃SiH₂^. + CH₄, in a solid solution of CH₃SiH₃ containing 1 mol% CH₃I at the low temperatures of 3 K-115 K. The EPR spectra observed after UV-photolysis of the CH₃I at 77 K were attributed to a mixture of the CH₃SiH₂^. radical and the ^.CH₃ radical. In the CH₃SiH₃ system, the CH₃SiH₂^. radical was the major product immediately after the photolysis, while the ^.CH₃ radical was the major one in the CH₃SiD₃ system. The ^.CH₃ radicals decayed following first order kinetics in the dark in both systems. The decay rate constants for the reaction were experimentally determined to be k_(Si-H) = 3.6 × 10^-2 s^-1 and k_(Si-D) = 6.9 × 10⁶ s¹ (k_(Si-H)/k_(Si-D) = 5.2 × 10³) at 77 K; the associated apparent activation energies were E_a(Si-H) = 0.85 kJmol s¹ and E_a(Si-D) = 8.9 kJmol s¹ (E_a(Si-H/E_a(Si-D) = 1/10) above 20 K. A non-linear Arrhenius plot was obtained for the rate constant, k_(Si-H), and the rate became almost independent of the temperature below 20 K. These results suggest that the quantum mechanical tunneling effect contributes significantly to the H atom abstraction from the -SiH₃ group.

Full text of DOI:10.1039/b206503g

View Article Online / Journal Homepage / Table of Contents for this issue
Selectivity and enormous H/D isotope effects on H atom
abstraction by CH3 radicals in solid methylsilane at 3.0 K–115 K
Kenji Komaguchi,*a Yuko Ishiguri,a Hiroto Tachikawab and Masaru Shiotani*a
a
Department of Applied Chemistry, Graduate school of Engineering, Hiroshima University,
Higashi-Hiroshima 739-8527, Japan. E-mail: okoma@hiroshima-u.ac.jp;
Fax: +81-824-245494; Tel: +81-824-247870
Division of Molecular Chemistry, Graduate School of Engineering, Hokkaido University,
b
Sapporo 060-8621, Japan. Fax: +81-11-7066750; Tel: +81-11-7066750
Received 4th July 2002, Accepted 30th August 2002
First published as an Advance Article on the web 20th September 2002
An EPR study was carried out to elucidate the hydrogen atom abstraction from methylsilane (CH3SiH3) by a
methyl radical, CH3SiH3 + ꢀCH3 ! CH3SiH2ꢀ + CH4 , in a solid solution of CH3SiH3 containing 1 mol% CH3I
at the low temperatures of 3 K–115 K. The EPR spectra observed after UV-photolysis of the CH3I at 77 K were
ꢀ
ꢀ
ꢀ
attributed to a mixture of the CH3SiH2 radical and the CH3 radical. In the CH3SiH3 system, the CH3SiH2
radical was the major product immediately after the photolysis, while the ꢀCH3 radical was the major one in the
ꢀ
CH3SiD3 system. The CH3 radicals decayed following first order kinetics in the dark in both systems. The
decay rate constants for the reaction were experimentally determined to be k(Si–H) ¼ 3.6 ꢁ 10ꢂ2 sꢂ1 and
k(Si–D) ¼ 6.9 ꢁ 10ꢂ6 sꢂ1 (k(Si–H)/k(Si–D) ¼ 5.2 ꢁ 103) at 77 K; the associated apparent activation energies were
Ea(Si–H) ¼ 0.85 kJmolꢂ1 and Ea(Si–D) ¼ 8.9 kJmolꢂ1 (Ea(Si–H)/Ea(Si–D) ¼ 1/10) above 20 K. A non-linear
Arrhenius plot was obtained for the rate constant, k(Si–H) , and the rate became almost independent of the
temperature below 20 K. These results suggest that the quantum mechanical tunneling effect contributes
significantly to the H atom abstraction from the –SiH3 group.
Methylsilane (CH3SiH3) is the simplest alkylsilane, com-
1. Introduction
posed of a CH3 group and a SiH3 group. The Si–H bond
length of Si with sp3 hybridization is ca. 0.149 nm, which is
ca. 36% longer than the corresponding C–H bond (ca. 0.110
nm);11 the difference mainly comes from that in the covalent
bond radii of Si and C atoms, i.e., 0.117 and 0.0771 nm, respec-
tively.12 In addition, the dissociation energy of the Si–H bond
is more than 40 kJ molꢂ1 smaller than the C–H bond.13,14
These facts may significantly contribute to either the classical
or non-classical (i.e., tunnel) reaction kinetics, or both; the
reaction proceeds by passing over the potential barrier in the
former, but through the barrier in the latter. Hence, methylsi-
lane was chosen as a suitable molecule for studying the hydro-
gen abstraction reaction because the molecule has two groups
of CH3 and SiH3 with remarkably different reactivity expected
for each.
In the present report, H atom abstraction reactions from
methylsilane by CH3 radicals in the low temperature solid
CH3SiH3 and CH3SiD3 systems were studied by EPR spectro-
scopy. It was found that the H atoms were preferentially
abstracted from the SiH3 group, not from the CH3 group, with
a remarkable H/D isotope effect on the reaction rate as well as
the apparent activation energy. Here, we wish to demonstrate
experimentally that the quantum mechanical tunneling effect is
manifested in the H atom abstraction from the –SiH3 group.
The quantum mechanical tunneling effect on elementary reac-
tions is fundamentally important in low temperature chemis-
try. The chemical reaction via the tunneling effect is
characterized by a non-linear Arrhenius plot, a large isotope
H/D effect on the rate constant, and high selectivity of the
reaction.1,2 Four decades ago a remarkable 2D effect was
reported for the decay rate of radicals produced in protiated
and perdeuterated organic compounds at 77 K by Sullivan
and Koski.3 Since then a number of studies have been reported
on tunneling reactions in low temperature solids.4–9 For exam-
ple, in 1971, Williams and collaborators reported the H atom
abstraction from CH3CN by the CH3 radical with a low activa-
tion energy of 1.4 kcal molꢂ1 and a large ‘‘primary’’ kinetic
isotope effect, k(H)/k(D) > 140, between 77 K and 87 K.5 One
of the authors (M.S.) has clearly demonstrated a non-linear
Arrhenius plot for the reaction, CH3OH + ꢀCH3 ! ꢀCH2OH +
CH4 , in solid CH3OH at the low temperatures of 10 K–90 K.9
Recently, Hiraoka and his collaborators reported that amor-
phous silicon thin films are formed by successive H atom
abstraction from the silane (SiH4) molecule by H atoms at
the low temperature of 10 K.10 They developed a methodology
to generate H atoms in the ground state and sprinkled them on
the silane molecular layers deposited on a silicon substrate
cooled at 10 K. They proposed a reaction mechanism via quan-
tum mechanical tunneling for the formation of the amorphous
silicon. However, there is no direct experimental evidence of
paramagnetic species as reaction intermediates, because the
reaction was monitored by in situ IR spectroscopy, not by
EPR. Furthermore, no kinetic studies have ever been reported
for the H atom abstraction reaction of low temperature solid
silicon compounds.
2. Experimental
Methylsilane and methylsilane-1,1,1-d3 were synthesized from
trichloromethylsilane by a conventional method.15 The purities
of both methylsilanes were greater than 98 mol% based on
the 1H–, 13C–, 29Si-NMR analyses (JEOL Model Ex-270
5276
Phys. Chem. Chem. Phys., 2002, 4, 5276–5280
This journal is # The Owner Societies 2002
DOI: 10.1039/b206503g

View Article Online
spectrometer). For the CH3SiD3 , the degree of deuteration is
presumably 98 atom%, which is governed by that of LiAlD4
(Aldrich Chemical Co., Inc) used as the reducing reagent in
the synthesis. CH3I was purchased from Tokyo Kasei Kogyo
Co., Ltd., and used without further purification. The samples
were prepared in a Suprasil quartz tube (4 mm diameter, 20
cm length) on a vacuum line, and subjected to UV-photolysis
(high-pressure mercury lamp, 400 W) at 77 K.
EPR spectra were recorded on a Bruker ESP300E X-band
spectrometer at a microwave power level small enough to
avoid saturation in the region of 5 mW–0.6 mW. For the
experiments to measure the time-course of the radical concen-
tration at 77 K, a JEOL JES-RE1X spectrometer was used.
The sample temperature was controlled within ꢃ0.1 K using
a cryostat (Oxford ESR 900) with liquid helium as the coolant.
however, was almost independent of the illumination time,
and disappeared within a few minutes in the dark at 77 K after
the illumination.
In contrast to the results obtained for the CH3SiH3 system,
the decay rate of the CH3 radical was found to be much slower
in the CH3SiD3 system where hydrogen atoms of the SiH3
group are fully deuterated. Fig. 2 shows the EPR spectra
observed for the CH3SiD3 samples containing 1 mol% of
CH3I. Upon exposing the sample to UV-light at 77 K, the
quartet spectrum of the CH3 radical was clearly observed with
a 2.3 mT hfcc. Furthermore, it became experimentally clear
ꢀ
that the decaying CH3 radical was accompanied by the simul-
taneous growth of another quartet with the smaller hfcc of
0.8 mT at the central position. The latter quartet was attribu-
ꢀ
ted tꢀo the 1H hfcc of the CH3SiD2 radical. We repeat that the
CH3 radicals decayed in both systems, but the decay rate was
much slower in CH3SiD3 than in CH3SiH3 . The EPR results
suggest that the CH3 radical preferentially abstracts a hydro-
ꢀ
3. Results and discussion
gen atom from the SiH3 group of CH3SiH3 with an anoma-
lously large H/D isotope effect on the reaction kinetics.
3.1. EPR spectra
Fig. 1 shows the EPR spectra of a solid solution of CH3SiH3
containing 1 mol% of CH3I irradiated by UV-light at 77 K.
The spectrum consists of two components. One is a multiplet
with a total hf-coupling constant of ca. 5 mT at the central
part. The other is a quartet due to the ꢀCH3 radical as marked
by the asterisk, which is generated by the photolysis of CH3I,
but is weak in intensity. The multiplet was well reproduced by
assuming a triplet of quartets with isotropic 1H hyperfine cou-
pling constants (hfcc) of 1.2 mT (2 H) and 0.8 mT (3 H). These
EPR parameters correspond well to those reported for
3.2. Rate-determining reaction
ꢀ
ꢀ
The initial and final radicals, i.e., CH3 and CH3SiH2 , were
observed by the present EPR study. Now we shall consider
the reaction mechanism for the radical conversion. There are
ꢀ
three possible reactions (Scheme 1) for the CH3SiH2 radical
formation. One is reaction (2) where the CH3 radical abstracts
the H (or D) atom directly from the silyl group (SiH3). The
other two reactions are abstraction of the H atom from the
CH3 group, reaction (3), followed by either an inter- or
intra-molecular H atom transfer from the carbon to the silicon,
reaction (4) or (5) (Scheme 1), respectively. As mentioned
above, a large isotope effect was observed for the deuteration
of the SiH3 group, therefore, the abstraction of the H atom
from the SiH3 group either by reaction (2) or reactions (4)
and (5) is expected to be the rate-determining step. If the latter
reactions were the case, the decay of the ꢀCH2SiH3 (or
ꢀCH2SiD3) radical should be observable, accompanied by an
ꢀ
ꢀ
CH3SiH2 radicals.16,17 The concentration of CH3SiH2 radi-
cals increased with the UV-illumination time for the several
initial minutes at 77 K. The CH3 radical concentration,
Fig. 1 EPR spectra observed for CH3SiH3 containing 1 mol% of
CH3I immediately (a) and 9 min (b) after UV-photolysis for 1 min
at 77 K, and a simulated spectrum (c) using the following EPR para-
meters, aSi(H) ¼ 1.2 mT (2 H), aC(H) ¼ 0.8 mT (3 H), and line width
of 0.35 mT. The spectra were measured at 77 K in the dark. The peaks
ꢀ
due to the CH3 radicals are denoted by an asterisk. The singlet indi-
Fig. 2 EPR spectra observed for the CH3SiD3 systems containing 1
mol% of CH3I immediately (a), 77 h (b), and 143 h (c) after UV-photo-
lysis for 1 min at 77 K. The spectra were measured at 77 K in the dark.
cated by the arrow in (a) and (b) is attributable to the paramagnetic
impurities generated in the quartz sample tube by UV-light illumina-
tion.
ꢀ
The peaks due to the CH3 radicals are denoted by an asterisk.
Phys. Chem. Chem. Phys., 2002, 4, 5276–5280
5277

View Article Online
Scheme 1
ꢀ
ꢀ
ꢀ
increase in the CH3SiH2 (CH3SiD2 ), or CH2DSiD2 radical.
In the present systems, however, no EPR spectra atꢀtributable
to the ꢀCH2SiH3 (or ꢀCH2SiD3) and CH2DSiD2 radicals
were detected after the photolysis, even at the low temperature
of 4 K. Thus, reactions (4) and (5) can be ruled out, and we
ꢀ
conclude that the CH3SiH2 radicals were preferentially gener-
ated by reaction (2).
ꢀ
Fig. 3 (a) Time-courses of the number of CH3 radicals (/) and the
whole radical (S) in solid CH3SiD3 molecules containing 1 mol% of
CH3I at 77 K in the dark after UV-photolysis for 1 min at the same
temperature. (b) First-order plot of the ꢀCH3 radical decay in the
CH3SiH3 (L) and CH3SiD3 (/) systems at 77 K in the dark.
3.3. Rate constant and activation energy
We can now take only reaction (2) into account in the follow-
ing kinetic analysis. The relative number of CH3 radicals was
ꢀ
Substituting 2148 cmꢂ1 and 1552 cmꢂ1, the frequencies for the
fundamental vibrations of the Si–H and Si–D stretching modes
observed for the polycrystalline methylsilanes at 77 K,20 for
the values of nH and nD in eqn. (6) yields the classical isotope
effect of ca. 340 at most. In fact, a large H/D isotope effect of
5.2 ꢁ 103 was experimentally evaluated for the hydrogen atom
abstraction from the SiH3 group; the value is a factor of ca. 15
times greater than the maximum isotope effect due to the clas-
sical mechanism at 77 K. Here, we note that the classical iso-
tope effect on the methyl hydrogens can be several times
greater than on the silyl hydrogens because of the larger differ-
ence between the zpe of the C–H and C–D bonds than between
the Si–H and Si–D bonds.
Fig. 4 shows an Arrhenius plot of the first-order rate con-
stant, k, for the CH3 radical decay. The plot shows a nonlinear
relationship. ln k decreases with decreasing temperature, and
becomes almost temperature independent below 20 K. If a lin-
ear relation above 20 K is assumed, one can evaluate the
apparent activation energies of 0.85 kJ molꢂ1 for CH3SiH3
and 8.9 kJ molꢂ1 for CH3SiD3 ; the H/D isotope effect on
the apparent activation energy is about ten.
evaluated from the peak heights of the outer two lines
(mI ¼ ꢃ3/2), and the total number of radicals was obtained
from the double integration of the first-derivative spectra.
Fig. 3 shows the time-courses of the relative number of the
ꢀCH3 radical and the whole radical in the CH3SiD3 system at
ꢀ
77 K. The CH3 radicals decayed exponentially as shown in
Fig. 3(a), while the relative number of whole radicals remained
almost constant except for the initial several hours. The result
is consistent with the assumption of reaction (2) where the
ꢀCH3 radical decreases with corresponding increase in the
ꢀ
CH3SiH2 radical. The initial decrease in the ꢀCꢀH3 radical
may be caused by recombination reactions of the CH3 radi-
cals, and/or of the ꢀCH3 radical with an iodine atom generated
in the near vicinity. Assuming the reaction proceeds by a first
ꢀ
order process, the natural logarithm of the CH3 concentra-
tion, ln[ꢀCH3], was plotted vs. the reaction time, t. The plot
was linear, as shown in Fig. 3(b). The CH3 radicals decayed
with a half-life of t1/2 ¼ 19.3 s and 28 h for the CH3SiH3
and ꢀCH3SiD3 systems, respectively, at 77 K. The half-life of
the CH3 radicals in the CH3SiH3 system is very short, less
than one-thirtieth of those reported for organic compounds
such as CH3CN,5 CH3NC,18 and 3-methylpentane6,19 with
t1/2 ¼ 28 min, 210 min, and 11–25 min, respectively.
Theoretical calculations were performed to evaluate the
energy diagrams for both reactions (2) and (3), at a level of
MP2/6-311G(d,p) + DZPE(MP2/6-31G*) as shown in
Fig. 5.21,22 For reaction (2), the transition state was calcu-
lated to be 12.5 kJ molꢂ1 lower than that for reaction (3). This
result is consistent with the previous experimental data show-
ing that the activation energy for the reaction, SiH3SiH3 +
Based on eqn. (6) of the zero point energy (zpe) difference,
one can estimate the maximum H/D isotope effect for the reac-
tion rate constants via a classical mechanism, where h is
Planck’s constant, n is the zero point vibrational frequency,
and kB is Boltzman’s constant.18
ꢀ
ꢀCH3 ! SiH3SiH2 + CH4 , is 39 kJ molꢂ1 lower than the reac-
kH=kD ¼ 21=2expððhnH ꢂ hnDÞ=2kBTÞ
ð6Þ
tion, CH3CH3 + ꢀCH3 ! CH3CH2ꢀ + CH4 , in the gas phase at
5278
Phys. Chem. Chem. Phys., 2002, 4, 5276–5280

View Article Online
Fig. 6 The relationship between the activation energy for reactions
(2) and (3), and the angle j which the pz-orbital of the CH3 radical
forms with the Si–H (or C–H) bond of the methylsilane. The energy
was theoretically evaluated at the MP2/6-311G(d,p) level. The zero
point energy is not included here. The atomic distance was fixed as
the same as those of the transition state, r(Hꢄ ꢄ ꢄCH3) ¼ 0.1532 nm
and r(Siꢄ ꢄ ꢄH) ¼ 0.1642 nm for reaction (2), r(Hꢄ ꢄ ꢄCH3) ¼ 0.1348 nm
and r(Cꢄ ꢄ ꢄH) ¼ 0.1319 nm for reaction (3).
Fig. 4 Arrhenius plots of the rate constants, k(Si–H) (S) and k(Si–D)
(/), for reaction (2) in the temperature region between 3.0 K and
115 K.
ca. 400 K.23,24 In the present low temperature solid, this lower
potential barrier can be one reason why the tunneling H atom
abstraction proceeds more effectively for the SiH3 group than
for the CH3 group. In addition, preliminary calculations were
performed to evaluate the dependence of the activation energy
on the direction which the CH3 radical approaches to abstract
the hydrogen from the methylsilane molecule. At the transition
states, TS-I and TS-II (Fig. 5), where the pz-orbital of the CH3
radical is collinear (j ¼ 0) with the Si–H (or C–H) bond of the
methysilane, the activation energy is a minimum. When the
atomic distance was fixed as the same as those of the transition
state, r(Hꢄ ꢄ ꢄCH3) ¼ 0.1532 nm and r(Siꢄ ꢄ ꢄH) ¼ 0.1642 nm for
reaction (2), r(Hꢄ ꢄ ꢄCH3) ¼ 0.1348 nm and r(Cꢄ ꢄ ꢄH) ¼ 0.1319
nm for reaction (3), the activation energy was gradually
increased with the departure from linearity as shown in Fig. 6.
At the angles between 0ꢅ and 20ꢅ calculated here, the change in
the activation energy for the Si–H bond is relatively smaller
than for the C–H bond. These theoretical results are consistent
with the experimental ones. Consequently, the H atom abstrac-
tion with a higher selectivity and a larger rate constant is
attained at the SiH3 group, in particular via the tunneling
mechanism.
role in the rate of the H atom tunneling reaction in low tem-
perature solids. However, in the present reaction, such geome-
trical relaxation could not be that significant, because the H
atom abstractions take place at the terminal sp3 silicon. Rather
than the geometrical relaxation of the reactant molecule,
the extent to which the H atom orbital can overlap with
ꢀ
the unpaired electron orbital of the CH3 radical may play a
crucial role in the tunneling reaction.
We also examined the reaction by H-atoms, which were gen-
erated by the photolysis of HI in the solid methylsilane. It was
proved experimentally that the ꢀreaction was again highly selec-
tive, so that only the CH3SiH2 radical was observed, i.eꢀ., the
result is similar to that of the reaction initiated by the CH3
radical. The reaction proceeded too rapidly to follow the
H-atom decay by cw-EPR spectroscopy, even at 4 K. In addi-
ꢀ
tion, the CH3SiH2 radical was also preferentially generated by
g-ray radiolysis of pure CH3SiH3 at 77 K. WhenꢀCH3SiD3 was
used instead of CH3SiH3 , only the CH3SiD2 radical was
observed, i.e., no H/D isotope effect was detected for the site
of the hydrogen abstraction both for the UV-photolysis of
HI andꢀg-ray radiolysis. Furthermore, the H-atom abstraction
by the CD3 radical in the CH3SiH3 was also examined. No
Recently, Ichikawa et al.25 and Kumagai et al.26 indepen-
dently pointed out that a certain relaxation of the local geo-
metrical constraints of a reactant molecule plays an important
ꢀ
significant H/D isotope effect of the deuteration of the CH3
radical on its decayꢀrate was observed at 77 K, with the result
that the CH3SiH2 radicals were preferentially generated
immediately after the photolysis of CD3I.
4. Conclusions
ꢀ
The EPR study revealed that the CH3 radical preferentially
abstractsꢀthe H atom from the SiH3 group of CH3SiH3 to form
CH3SiH2 radicals at temperatures from 3 K to 115 K. The
ꢀCH3 radical decayed exponentially in the dark in this tem-
perature range. Based on the first-order decay of the ꢀCH3
radical, the rate constants and apparent activation energies
for reaction (2) were evaluated to be k(Si–H) ¼ 3.6 ꢁ 10ꢂ2 sꢂ1
and k(Si–D) ¼ 6.9 ꢁ 10ꢂ6 sꢂ1 (k(Si–H)/k(Si–D) ¼ 5200) at 77 K,
Ea(Si–H) ¼ 0.85 kJmolꢂ1 and Ea(Si–D) ¼ 8.9 kJmolꢂ1 (Ea(Si–H)
/
Ea(Si–D) ¼ 1/10) above 20 K, respectively. The rate con-
stants, k(Si–H) , were almost independent of the temperature
below 20 K. This, together with the observed anomalously
large isotope effect on the reaction, revealed that the tunneling
effect contributes significantly to the H atom abstraction by
Fig. 5 Energy diagrams for reactions (2) and (3) theoretically evalu-
ated at the MP2/6-311G(d,p) + DZPE(MP2/6-31G*) level. The zero
point energy is included here.
Phys. Chem. Chem. Phys., 2002, 4, 5276–5280
5279

View Article Online
K. Komaguchi, T. Kumada, Y. Aratono and T. Miyazaki, Chem.
Phys. Lett., 1997, 268, 493–497.
R. L. Hudson, M. Shiotani and F. Williams, Chem. Phys. Lett.,
1977, 48, 193–196.
ꢀCH3 radicals from the –SiH3 group, especially at tempera-
tures below 20 K. Further experimental and theoretical stu-
dies are being carried out on the H atom abstraction
reaction for a series of alkylsilanes in order to explore the
reactivity of the silyl hydrogen via tunneling.
8
9
10 K. Hiraoka, T. Sato, S. Sato, S. Hishiki, K. Suzuki, Y. Takahashi,
T. Yokoyama and S. Kitagawa, J. Phys. Chem. B., 2001, 105,
6950–6955.
11 L. Pauling, The Nature of the Chemical Bond, Cornell University
Press, 1938.
Acknowledgements
12 CRC Handbook of Chemistry and Physics, ed. D. R. Lide, CRC
Press, New York, 75th edn., 2001, pp. 9-36.
13 J. A. Seetula, Y. Feng, D. Gutman, P. W. Seakins and M. J.
Pilling, J. Phys. Chem., 1991, 95, 1658–1664.
14 J. A. Seetula, J. J. Russell and D. Gutman, J. Am. Chem. Soc.,
1990, 112, 1347–1353.
15 S. Tannenbaum, S. Kaya and G. Lewenz, J. Am. Chem. Soc.,
1953, 75, 3753–3757.
16 P. J. Krusic and J. K. Kochi, J. Am. Chem. Soc., 1969, 91,
3938–3940.
This research was supported in part by a grant from the Mini-
stry of Education, Science, and Culture of Japan. Our special
thanks are due to Dr Kumada and Dr Aratono of the Japan
Atomic Energy Research Institute, and Dr Kumagai of
Nagoya University for use of JAERI’s facilities to measure
the EPR spectra below 4.2 K. We also thank the Cryogenic
Center, Hiroshima University, for supplying the cryogen
(liquid helium).
17 J. H. Sharp and M. C. R. Symons, J. Chem. Soc. A, 1970,
3084–3087.
18 J.-T. Wang and F. Williams, J. Am. Chem. Soc., 1972, 94,
2930–2934.
19 D. W. Skelly, R. G. Hayes and W. H. Hamil, J. Chem. Phys.,
1965, 43, 2795–2798.
References
1
V. A. Benderskii, D. E. Makarov, C. A. Wight, Chemical
Dynamics at Low Temperatures, Advances in Chemical Physics
LXXXVIII, Wiley, New York, 1994.
J. E. Willard, in Chemical Kinetics of Small Organic Radicals,
ed. Z. B. Alfassi, CRC Press, Boca Raton, FL, 1988, vol. IV,
pp. 1–29.
20 I. A. Oxton, J. Mol. Struct., 1979, 56, 57–68.
21 J. B. Foresman and Æ. Frisch, Exploring Chemistry With Electro-
nic Structure Methods, Gaussian Inc., Pittsburgh, PA, 2nd edn.,
1996.
22 H. Tachikawa, N. Hokari and H. Yoshida, J. Phys. Chem., 1993,
97, 10 035–10 041.
2
3
4
5
P. J. Sullivan and W. S. Koski, J. Am. Chem. Soc., 1963, 85,
384–387.
W. G. French and J. E. Willard, J. Phys. Chem., 1968, 72,
4604–4608.
F. Williams and E. D. Sprague, J. Am. Chem. Soc., 1971, 93,
787–788.
E. D. Sprague, J. Phys. Chem., 1973, 77, 2066–2970.
T. Miyazaki, Radiat. Phys. Chem., 1991, 37, 635–642.
23 O. P. Strausz, E. Jakubowski, H. S. Sandhu and H. E. Gunning,
J. Chem. Phys., 1969, 51, 552–560.
24 J. R. McNesby and A. S. Gordon, J. Am. Chem. Soc., 1955, 77,
4719–4723.
25 T. Ichikawa, K. Kagei, Y. Ishitani, H. Tachikawa and H.
Koizumi, Radiat. Phys. Chem., 1999, 55, 529–534.
26 J. Kumagai, Y. Itagaki, M. Ishizuka, T. Kumada, A. Lund and
T. Miyazaki, J. Phys. Chem. A, 2001, 105, 3652–3657.
6
7
5280
Phys. Chem. Chem. Phys., 2002, 4, 5276–5280