Observation of unique pressure effects in the combination reaction of benzyl radicals in the gas to liquid transition region

Article abstract of DOI:10.1039/b305954e

The combination reaction of two benzyl radicals has been studied in a wide range of pressures (0.01-1000 bar) and temperatures (250-400 K) in various bath gases (Ar, N₂ and CO₂). The measured second-order combination rate constants of benzyl radicals were independent of the pressure and the bath gas below 1 bar, as expected for a limiting "high pressure" rate constant of a termolecular combination process. However, the reaction becomes steadily faster when the pressure in Ar is further raised until it finally starts to decrease when densities corresponding to diffusion controlled kinetics are reached. Such a unique pressure dependence was more strongly accentuated in CO₂ and at lower temperatures. Our results seem to provide first clear indications for a contribution of the radical-solvent interaction in the combination reaction kinetics of such large radicals as benzyl in the gas to liquid transition range.

Full text of DOI:10.1039/b305954e

View Article Online / Journal Homepage / Table of Contents for this issue
Observation of unique pressure effects in the combination reaction of
benzyl radicals in the gas to liquid transition region
Kawon Oum,* Kentaro Sekiguchi, Klaus Luther and J u¨ rgen Troe
Institut f u¨ r Physikalische Chemie, Universit a¨ t G o¨ ttingen, Tammannstraße 6, D-37077 G o¨ ttingen,
Germany. E-mail: koum@gwdg.de; Fax: +49 551 393150; Tel: +49 551 3912598
Received 29th May 2003, Accepted 9th June 2003
First published as an Advance Article on the web 12th June 2003
1
,2
The combination reaction of two benzyl radicals has been studied
in a wide range of pressures (0.01–1000 bar) and temperatures
diffusion controlled range is expected following eqn. (2):
g
rec
k
kdiff
(
250–400 K) in various bath gases (Ar, N and CO ). The mea-
2
krec
ꢀ
ð2Þ
2
g
krec þ kdiff
sured second-order combination rate constants of benzyl radicals
were independent of the pressure and the bath gas below 1 bar, as
expected for a limiting ‘‘high pressure’’ rate constant of a termo-
lecular combination process. However, the reaction becomes
steadily faster when the pressure in Ar is further raised until it
finally starts to decrease when densities corresponding to diffu-
sion controlled kinetics are reached. Such a unique pressure
Here kg denotes the recombination rate constant at a given
density ‘‘in the absence of diffusion control’’, and kdiff is
given by the Smoluchowski equation for diffusion-controlled
encounters. In liquid solutions, the recombination of benzyl
radicals occurs close to the diffusion limit. In the gas phase,
Fenter et al. and later Boyd et al. found that the reaction
rec
3
4
5
dependence was more strongly accentuated in CO and at lower
2
rates are at the limiting high pressure values in 760 Torr N2 ,
and almost independent of temperature between 400–450 K
4
temperatures. Our results seem to provide first clear indications
for a contribution of the radical–solvent interaction in the com-
bination reaction kinetics of such large radicals as benzyl in
the gas to liquid transition range.
5
and 435–519 K. A mild temperature dependence of the limit-
ing high pressure rate constant, k1 (T ), in the range of
,1
700–1500 K, was also observed in a shock wave study.
Reaction (1) was also investigated by Brennecke’s group
6
in supercritical CO
(
2
(70–170 bar) and supercritical ethane
ꢁ
7
50–120 bar) at two temperatures of 35 and 50 C. A rapid
Introduction
increase of the recombination rates was clearly observed when
the pressure was lowered from an initially higher value
towards the critical pressure. The authors interpreted their
data only on the basis of diffusion-controlled kinetics by utiliz-
ing spin-statistical factors and concluded that there is no indi-
cation of an additional increase of the recombination rates due
to solvent effects in reaction (1). However, in the present study,
we have observed a further increase of the recombination rates
in the gas phase at higher pressures of Ar, N and CO , when
raising the pressure from 1 bar into the critical range or even
much higher. In addition a pronounced temperature depen-
dence of such a unique solvent effect in CO was found. In
the light of our new results presented here, earlier data from
Brennecke and co-workers have to be reinterpreted. Our
results strongly suggest that benzyl radical–solvent molecule
interactions have a profound influence on kinetics in the gas
to liquid transition region.
Combination reactions of radicals in the gas phase show a
characteristic pressure dependence, also well known for the
reverse process, i.e. unimolecular dissociation: Linear pressure
dependence of the rate constants at ‘‘low’’ pressures is fol-
lowed by the ‘‘fall-off’’ transition regime leading into the
‘‘high’’ pressure range with a constant, limiting value, k ,
1
independent of the density and nature of the bath gas. For
large radicals this constant value ‘‘k ’’ of the so-called energy
transfer mechanism is often reached at pressures far below
ET
1
2
2
1 bar. Thus a broad plateau over orders of magnitude in pres-
ET
sure is expected with k equal to k until densities are so high
2
1
that diffusion of the reactants begins to play a role and slow
down the reaction rate. The combination reactions of large
radicals therefore provide an excellent testing ground for
additional bath gas (solvent) effects on the reaction kinetics,
beyond that of energy transfer in Markovian binary collisions
responsible for the pressure dependence mentioned above.
Basic features may be studied under pure gas phase conditions
at temperatures far above the critical value and densities reach-
ing far into the range of typical liquids. Suitable choice of bath
media/solvents and temperature should allow one to identify
additional effects occurring only under typical supercritical
conditions or really close to the critical point. As a prototype
of such studies, the combination reaction of benzyl radicals
Experimental
In this study, benzyl radicals were generated from the bimole-
cular reaction of chlorine atoms with excess amounts of
toluene, following the photolysis of molecular chlorine at
308 nm using an excimer laser:
˙
6 5 2
(C H CH ) has been chosen:
ꢂ
Cl2 þ hn ð308 nmÞ ! 2 Cl
Cl þ C6H5CH3 ! C6H5CH2 þ HCl
ð3Þ
ð4Þ
_
_
ꢂ
C6H5CH2 þ C6H5CH2 ! C14H14
k1
ð1Þ
_
If there is no significant solvent effect in reaction (1), a tran-
sition of the pseudo-second order recombination rate constant
The temporal behaviour of the benzyl radicals was monitored
using the UV absorption technique at 253 nm on time scales of
a few milliseconds. A 200 W HgXe lamp was used as a detec-
tion light source and the light level was measured by a
krec ( ¼ k ) from termolecular low-pressure gas-phase beha-
1
viour, via a broad high-pressure plateau, into the dense fluid
DOI: 10.1039/b305954e
Phys. Chem. Chem. Phys., 2003, 5, 2931–2933
This journal is # The Owner Societies 2003
2931

View Article Online
monochromator-photomultiplier arrangement through the
high pressure cell (with a spectral band width of ca. 2 nm).
Premixed gas mixtures of Cl , toluene, and the bath gas were
1 bar. By averaging the rate constants below 1 bar, a value of
ET
1; 1
ðTÞ ¼ ð4:1ꢅ0:3Þꢃ10ꢄ11ðT=300 KÞꢄ0:3 cm molecule
3
ꢄ1 ꢄ1
k
s
2
compressed in an oil-free diaphragm compressor, and then
flowed through the high pressure cell (path length 10 cm and
optical diameter 0.9 cm). For experiments below 1 bar, a glass
flow cell was used in a separate setup (path length 52 cm and
optical diameter 3 cm). Flow rates were controlled by flow
meters (Tylan FM361 and FM362) at rates such that reagents
and products were removed from the observation volume
between laser pulses. Total pressures were measured with high
pressure meters (Burster, Model 8201). Two platinum resis-
tance thermometers were directly attached to the front and
back side of the cell to measure the temperature. Impurities
in the bath gases, especially oxygen, were removed by a series
of gas cleaning adsorber (Oxisorb, Messer-Griesheim) and
dust filters. All chemicals were purified in a pump–thaw–
freezing cycle prior to use. Typical concentration of benzyl
ð5Þ
was determined. A modest negative temperature dependence of
ET
1;1
k
was measured in Ar at 5 different temperatures between
1
0
2
75–400 K.
Because of the second order kinetics of reaction (1), the
measured reaction rates are sensitive to the absolute initial
concentration of benzyl radicals and therefore the value of
their absorption coefficients (sbenzyl). Uncertainties in s
benzyl
ET
1;1
may be one reason for the scatter of k
in previous
4
–6
reports.
Experimental consistency is thus better checked
ET
1;1
by comparison of reduced values, k /sbenzyl . A value of
ꢄ16
2
ꢄ1
sbenzyl ¼ 1.3 ꢃ 10
cm molecule was used in this study
after wavelength calibration and comparison with a recent
1
1
gas phase redetermination of sbenzyl(l) of the benzyl band
with its steep peak near 253 nm, also reported earlier. Possi-
1
2
1
3
ꢄ3
radicals was (1–5) ꢃ 10 molecule cm . Detailed information
8
,9
ble influences of density- and temperature-dependences of
1
on our experimental setup can be found elsewhere.
0
s
benzyl have also to be considered carefully. However, under
our experimental conditions, we could not see any unusual
behaviour in the UV absorption signals due to spectral shift
or unusual changes of the absorption intensities.
Results
10
Fig. 1 shows steadily increasing rate constants k
up to 100 bar in Ar and N , followed by decreasing rate con-
stants above 100 bar as the reaction becomes diffusion-
controlled. In CO , a much larger increase was seen when
approaching this pressure regime. In general, the degree of
such a unique pressure effect follows a tendency of CO
> He. A salient temperature dependence was obser-
1
from 1 bar
1
We measured second-order combination rate constants (k ) at
2
various combinations of reaction parameters, such as pressure,
temperature, and bath gases. Fig. 1 shows pressure and bath
gas dependent rate constants (k ) at room temperature in a
2
1
very wide pressure range of 0.01–1000 bar. The observed time
dependent UV absorption traces of benzyl radicals at 253 nm
were well explained by the second-order kinetic expression of
2
>
Ar ꢀ N
2
ved on this pressure effect, with an enhancement at a lower
temperature of 275 K and a correspondingly smaller effect at
reaction (1) and thus allowed the derivation of k . On a longer
1
time scale, the appearance of residual absorptions from pro-
ducts like bibenzyl and benzyl chloride is expected, however
their contribution is negligible due to their much smaller
absorption coefficients compared to the predominant absorp-
tion signal from benzyl radicals. Other side reactions were
350 K, as shown in Fig. 2. Our preliminary data confirm such
a unique pressure effect also in CF H. A full report on the
completed results of this study will follow.
3
1
0
1
0
found to be insignificant.
Rate constants were found to be at the limiting high pressure
Discussion
ET
;1
value of the pure energy transfer (ET) mechanism (k1
)
It is evident from our current results that there is a solvent-
induced pressure and temperature effect in radical combination
reaction rates, even in large aromatic radicals like benzyl (see
the solid line in Fig. 1, expected when no solvent effect is
already at 10 mbar, showing a constant value regardless of
the pressure and the nature of the bath gas up to about
Fig. 1 Density dependent second-order recombination rate constants
at 300 K. The solid line illustrates the expected
recombination rates when considering no solvent effects.
(k
1
) in CO
2
, Ar and N
2
Fig. 2 Density dependent second-order recombination rate constants
(k ) in CO at three different temperatures (275, 300 and 350 K).
1 2
2
932
Phys. Chem. Chem. Phys., 2003, 5, 2931–2933

View Article Online
considered). It has been believed, especially in the field of
supercritical fluids, that radical combination reactions obey
diffusion-controlled kinetics without any appreciable solvent
Prozesse’’) are gratefully acknowledged. Helpful discussions
with T. Lenzer and V. Ushakov are also acknowledged.
1
3–16
effect.
a clear knowledge of the pure gas-phase recombination rate
However, this can not be sufficiently judged without
References
constant ‘‘in the absence of diffusion control’’ (kg in
rec
eqn. (2)). Experimental data obtained only in the regime of
dense fluids do not allow one to tell whether a transition from
gas to liquid is smooth or deviates from the expected curve
following eqn. (2).
1
2
3
B. Otto, J. Schroeder and J. Troe, J. Chem. Phys., 1984, 81, 202.
J. Troe, J. Phys. Chem., 1986, 90, 357.
J. Sobek, R. Martschke and H. Fischer, J. Am. Chem. Soc., 2001,
1
23, 2849.
4
5
F. F. Fenter, B. Noziere, F. Caralp and R. Lesclaux, Int. J. Chem.
Kinet., 1994, 26, 171.
A. Boyd, B. Noziere and R. Lesclaux, J. Phys. Chem., 1995, 99,
10 815.
W. M u¨ ller-Markgraf and J. Troe, J. Phys. Chem., 1988, 92, 4899.
C. B. Roberts, J. Zhang, J. E. Chateauneuf and J. F. Brennecke,
J. Am. Chem. Soc., 1993, 115, 9576.
H. Stark, PhD Thesis, G o¨ ttingen University, 1999.
K. Luther, K. Oum and J. Troe, J. Phys. Chem. A, 2001, 105,
Because rate constants for large radicals such as benzyl are
already at the high pressure limit of the energy transfer
mechanism below 1 bar, the additional increase of rate con-
stants in the medium density region (before the diffusion-
controlled regime) must come from other dynamic features.
A most likely candidate, the so-called radical complex mechan-
ism, has been known for a long time and made responsible for
unusual gas phase high pressure behaviour in the atom recom-
6
7
8
9
5
535.
1
7–19
20
21
bination of I
O .
2
as well as in O + O2 , Cl + O2 and F +
So far it was considered to be relevant only in small
10 K. Oum, K. Sekiguchi, K. Luther and J. Troe, to be published.
11 N. Fay, PhD Thesis, G o¨ ttingen University, 1997.
2
2
2
1
1
1
2
3
4
N. Ikeda, N. Nakashima and K. Yoshihara, J. Phys. Chem., 1984,
88, 5803.
C. B. Roberts, J. Zhang, J. F. Brennecke and J. E. Chateauneuf,
J. Phys. Chem., 1993, 97, 5618.
J. Brennecke and J. E. Chateauneuf, Chem. Rev., 1999, 99, 433.
molecular systems. Only very recently ‘‘turn-ups’’ from con-
ET
stant values k at increasing high gas densities have been
found in combination reactions of slightly larger radicals
1
8
ClO + ClO, CCl
23
3
+ Br and CCl
3
+ CCl
3
.
The influence of
a radical complex mechanism as well as the possible role of
collisional quenching of excited electronic states are discussed
in ref. 23. A more detailed analysis of the analogous situation
for benzyl as the first case of really large radicals is
15 Y. Arai, T. Sako and Y. Takebayashi, Supercritical Fluids—
Molecular Interactions, Physical Properties, and New Applications,
Springer-Verlag, Berlin Heidelberg, 2002.
1
6
T. Arita and O. Kajimoto, J. Phys. Chem. A, 2003, 107, 1770.
17 H. Hippler, K. Luther and J. Troe, Chem. Phys. Lett., 1972,
6, 174.
H. Hippler, K. Luther and J. Troe, Ber. Bunsen-Ges. Phys. Chem.,
973, 77, 1020.
H. Hippler, K. Luther and J. Troe, Ber. Bunsen-Ges. Phys. Chem.,
973, 77, 1104.
1
0
underway.
1
1
8
9
1
1
1
Acknowledgements
2
2
0
1
H. Hippler, R. Rahn and J. Troe, J. Chem. Phys., 1990, 93, 6560.
S. Baer, H. Hippler, R. Rahn, M. Siefke, N. Seitzinger and
J. Troe, J. Chem. Phys., 1991, 95, 6463.
Financial support by the Alexander von Humboldt Founda-
tion (‘‘Sofja Kovalevskaja Program’’) and the Deutsche
Forschungsgemeinschaft through the Sonderforschungs-
bereich 357 (‘‘Molekulare Mechanismen unimolekularer
2
2
P. Campuzano-Jost, A. E. Croce, H. Hippler, M. Siefke and
J. Troe, J. Chem. Phys., 1995, 102, 5317.
23 K. Oum, V. Ushakov, K. Luther and J. Troe, in preparation.
Phys. Chem. Chem. Phys., 2003, 5, 2931–2933
2933