Heavy ion radiolysis of liquid benzene

Article abstract of DOI:10.1021/jp021401z

The mechanisms responsible for the main products in liquid benzene radiolysis (biphenyl, molecular hydrogen, and phenyl radical) are probed with protons, helium ions, and carbon ions of a few to 30 MeV energy. Phenyl radical yields have been examined using iodine scavenging techniques. The results are combined with similar data for γ-rays and suggest that phenyl radicals mainly react with benzene to give a long-lived adduct, which leads to polymer formation. Iodine can react with this adduct to give enhanced yields of biphenyl. Biphenyl is the predominant single hydrocarbon product in the radiolysis of neat benzene with a yield of 0.075 molecule/100 eV. Its yield is nearly independent of radiation type and energy suggesting that its formation in neat benzene is due to a fast ion-molecule process and not due to phenyl radicals. The total yield of 0.7 radicals/100 eV is almost entirely due to phenyl radicals and H atoms. A reexamination of the fluorescence from the singlet excited state of benzene suggests that this state is the precursor for molecular hydrogen and acetylene, whereas the triplet excited state decays to phenyl radicals and H atoms. Most of the excited states formed in the γ-radiolysis of benzene seem to decay to ground without formation of any product.

Full text of DOI:10.1021/jp021401z

11408
J. Phys. Chem. A 2002, 106, 11408-11413
Heavy Ion Radiolysis of Liquid Benzene
Jay A. LaVerne* and M. S. Araos
Radiation Laboratory, UniVersity of Notre Dame, Notre Dame, Indiana 46556
ReceiVed: June 12, 2002; In Final Form: September 17, 2002
The mechanisms responsible for the main products in liquid benzene radiolysis (biphenyl, molecular hydrogen,
and phenyl radical) are probed with protons, helium ions, and carbon ions of a few to 30 MeV energy.
Phenyl radical yields have been examined using iodine scavenging techniques. The results are combined
with similar data for γ-rays and suggest that phenyl radicals mainly react with benzene to give a long-lived
adduct, which leads to polymer formation. Iodine can react with this adduct to give enhanced yields of biphenyl.
Biphenyl is the predominant single hydrocarbon product in the radiolysis of neat benzene with a yield of
0.075 molecule/100 eV. Its yield is nearly independent of radiation type and energy suggesting that its formation
in neat benzene is due to a fast ion-molecule process and not due to phenyl radicals. The total yield of 0.7
radicals/100 eV is almost entirely due to phenyl radicals and H atoms. A reexamination of the fluorescence
from the singlet excited state of benzene suggests that this state is the precursor for molecular hydrogen and
acetylene, whereas the triplet excited state decays to phenyl radicals and H atoms. Most of the excited states
formed in the γ-radiolysis of benzene seem to decay to ground without formation of any product.
Introduction
magnitude greater than the yield of biphenyl. Most of the
benzene decomposition is thought to lead to “polymer” forma-
Liquid benzene is generally considered to be radiation inert
because the observed product yields are much lower than
corresponding aliphatic compounds; however, the reason for this
stability is not fully understood. Benzene, and by analogy other
aromatic compounds, were postulated to offer protection from
radiolytic decay when mixed with aliphatic hydrocarbons
1
,4
tion consisting of multiples of C6 units. The phenyl radical is
believed to have a major role in the formation of almost all of
the major hydrocarbon products from the radiolysis of benzene.
Therefore, a thorough examination of its chemistry would give
a significant amount of information on the radiolysis of liquid
benzene. Iodine scavenging techniques have successfully ex-
1
2
because of this presumed stability. For such a common
1
3-15
amined primary radical formation in cyclic alkanes
and
molecule, there is still considerable uncertainty in the mechanism
of benzene decomposition. The phenyl radical is thought to be
the main precursor of the hydrocarbon products, but until
3
more recently in the γ-radiolysis of benzene. These techniques
are expanded in this work to include different types of radiation
in order to provide fundamental information on the fate of the
phenyl radical in the radiolysis of liquid benzene. Variation in
particle LET (linear energy transfer, ) -dE/dx) influences
radical decay kinetics, which can be useful to distinguish
between different mechanisms.
1
recently, the evidence has been circumstantial. A recent study
using selective scavenging techniques has directly identified the
phenyl radical formation in liquid benzene and determined that
its yield can make up a substantial fraction of the total radical
3
yield. The work presented here extends that study to include
The major observed product from benzene radiolysis at high
LET is molecular hydrogen, see ref 16 and references therein.
Burns and co-workers have long proposed that second-order
reactions of the singlet excited state of benzene lead to the
an examination of the production of phenyl radicals in the
radiolysis of benzene with different types of radiation. This
fundamental knowledge of the radiolysis of benzene is extremely
important because it can be used as a basis of models for other
aromatic systems.
8,17,18
formation of molecular hydrogen.
Recent studies attempted
to examine the fluorescence induced by the radiolysis of benzene
with different types of radiation in order to directly compare
excite state yields with those of molecular hydrogen.19 Unfor-
tunately, it was later discovered that an artifact of the experi-
mental setup was contributing to the observed fluorescence at
high LET. These experiments have been repeated using a
slightly different experimental radiolysis cell, and the results
are reported here. Correlation of a given product with its
precursor state allows one to predict radiation chemical yields
with various types of radiation and for compounds difficult to
examine directly.
The main hydrocarbon product observed in the radiolysis of
neat liquid benzene is biphenyl with a yield of about 0.07
4-8
molecules/100 eV. It is believed that excited benzene formed
by the radiation initially decomposes to a phenyl radical and a
1
hydrogen atom. Spin trapping of the radicals in the radiolysis
9
of solid benzene gives evidence for the phenyl radical. Early
scavenging studies in the liquid could not identify the phenyl
radical, and it was proposed that its addition to benzene was
4
too fast for it to be observed. Labeling studies suggest that the
phenyl radical adds to benzene to give the phenylcyclohexa-
dienyl radical, which leads to biphenyl and lesser amounts of
4
,10
other dimers.
However, experiments with scavengers have
In this work, the radiolysis of benzene and its solutions with
iodine were examined with protons, helium ions, and carbon
ions and are compared to that found in γ-radiolysis. Very low
doses have been used in the scavenging studies to avoid
secondary reactions and to enable the use of low iodine
show that the total radical production in benzene is about 0.7
molecules/100 eV.4
,11,12
This value is almost an order of
*
To whom correspondence should be addressed.
1
0.1021/jp021401z CCC: $22.00 © 2002 American Chemical Society
Published on Web 10/24/2002

Heavy Ion Radiolysis of Liquid Benzene
J. Phys. Chem. A, Vol. 106, No. 46, 2002 11409
concentrations. Iodine concentrations were varied from 0.1 to
Radiation chemical yields, G values, are given in units of
molecules (radicals) formed (or decayed) per 100 eV of total
energy absorption.
30 mM. The dependences of biphenyl and iodobenzene yields
on iodine concentration are used to examine the temporal
variation of the phenyl radical. Fluorescence yields were
measured and compared to previous results on the yield of
molecular hydrogen formation. A general scheme for the overall
radiolysis of benzene is presented.
Results and Discussion
Selective ion monitoring techniques using a gas chromato-
graph/mass spectrometer have allowed the usage of doses down
to 1 kGy for accurate determination of the biphenyl yield in
liquid benzene. The radiation chemical yield of biphenyl in neat
liquid benzene has been found to be linear with dose in the
Experimental Section
Particle irradiations were performed using H, He, and 12C
1
4
3
range of 1-10 kGy in γ-radiolysis. Linearity was also observed
ions obtained from the 10 MV FN Tandem Van de Graaff of
the University of Notre Dame Nuclear Structure Laboratory.
The window assembly and irradiation procedure were essentially
for the heavy ions used in these studies, suggesting that dose
effects on secondary reactions are insignificant at the doses used
in this work.
the same as previously reported.1
6,20
Particle energy was
determined by magnetic analysis, and energy loss to the
Radical Production. The employment of a number of
different radical scavengers has shown that the total radiation
chemical yield, G value, of radicals in the γ-radiolysis of liquid
benzene is 0.72 radicals/100 eV of energy absorbed. 4 Radical
identification and their specific yields have proven to be more
problematic. The radiolysis of liquid benzene is thought to lead
to an excited state that decomposes to a phenyl radical and a
21
windows was calculated using standard stopping power tables.
Absolute dosimetry in the scavenging experiments was obtained
by collecting and integrating the charge from the sample cell
in combination with the particle energy. Beam currents were 5
2
0
nA, and total energy deposited was usually (1-5) × 10 eV
in 20 mL of sample to give a total dose of about 100-500 Gy.
The beam diameter was 6.35 mm, and completely stripped ions
were used. The sample cell for particle irradiation was made of
1
hydrogen atom:
•
•
2
C H ' C H + H
(1)
(2)
(3)
Pyrex with a mica window of about 5 mg/cm . The sample was
6
6
6
5
purged with ultrahigh purity helium before and during the
irradiation.
•
•
H + C H f C H
6
6
6
7
Irradiations with γ-rays were made using a Gammacell 220
Co-γ source having a dose rate of 16 Gy/min. The sample
6
0
•
•
C H + C H f C H -C H
6
6
5
6
6
6
5
6
cell was made from a quartz cuvette and contained about 4 mL
of sample. The sample was purged with ultrahigh purity helium
and the sample cell sealed with a rubber septum before the
radiolysis. Dosimetry was made in the same sample cell with
the Fricke dosimeter.22 A correction due to the relative electron
densities of the two media was applied. The yields reported
here were performed at total doses of 2500 Gy. Product yields
were found to be independent of total dose from 1 to 10 kGy.
Fluorescence experiments were performed in exactly the same
manner as previously reported using single photon counting
techniques.19 The present experiments utilized a titanium cell
window instead of the mica used in the previous studies. Mica
appears to scintillate strongly in the region of interest when
irradiated with very high LET particles. The sample cell, optics,
and photomultipliers were located in a light-tight box capable
Hydrogen iodide formation from iodine solutions demonstrates
the production of the H atom, and quantitative analysis gives
the HI yield in liquid benzene as 0.7 molecule/100 eV. This
23
value corresponds well with the total iodine, G(-I2), loss of
11
0
.33. Apparently, the radicals produced in reactions 2 and 3
7
react with iodine to give HI as previously suggested. Reaction
9
-1 -1
2
occurs with a rate constant of about 1.1 × 10 M s , so in
24,25
liquid benzene, the lifetime of the H atom is about 80 ps.
The cyclohexadienyl radical produced in reaction 2 has been
26
identified by ESR studies, but very little is known about its
chemistry.
Phenyl radical formation in the radiolysis of liquid benzene
has been proposed because of studies based on spin trapping in
9
the radiolysis of solid benzene. Labeling experiments suggest
90
of accepting the accelerator beam exit window or a small Sr-
that the phenyl radical reacts with benzene to give a phenyl-
9
0
4,8,10
Y beta source. Therefore, a direct comparison of the fluores-
cyclohexadienyl radical, a phenyl radical-benzene adduct.
cence efficiency with fast electrons and with heavy ions could
be made.
Evidence for such a radical has been observed in the solid
phase. Pulsed ESR experiments have determined that the
addition of the carboxyphenyl radical to benzene occurs with a
2
7
Measurements of hydrocarbon product yields were performed
with an EXTREL model ELQ-400-1 gas chromatograph-mass
spectrometer operating in the selective ion monitoring mode.
Iodobenzene was detected at masses 51 and 77, whereas
biphenyl was detected at masses 51 and 154. Chromatographic
separations were made with a 30 m Chrompack CP-Sil-5-CB
apolar type capillary column. Splitless mode injection of a 0.5-
6
-1 -1
rate constant of about 5 × 10 M s , and the rate for the
2
8
phenyl radical should be similar in magnitude. Therefore, the
lifetime of the phenyl radical is only about 18 ns but sufficiently
long to be detected with selected scavenging experiments. In
the radiolysis of benzene with added iodine, the phenyl radicals
will compete in their reactions with iodine to give iodobenzene,
reaction 4, and with benzene to give the phenylcyclohexadienyl
radical, reaction 3:
1
.0 µL solution was used. The initial column temperature was
maintained at 120 °C for 10 min during which benzene and
iodobenzene were eluted. The temperature was then raised to
•
•
2
00 °C at a rate of 30 °C/min and held 11 min at this
C H + I f C H I + I
(4)
6
5
2
6
5
temperature while biphenyl eluted. A typical chromatogram took
about 25 min.
Benzene from Aldrich with a purity of about 99.9% was
percolated over activated silica. No detectable quantities of the
products formed during the radiolysis, or any unsaturated
species, were detected with the mass spectrometer.
Measurement of the iodide yield as a function of iodine
concentration gives a good estimation of the temporal depen-
dence of the phenyl radicals. Recent experiments with iodine
scavenging in the γ-radiolysis of benzene positively identified
3
phenyl radicals using this technique. The results of these

11410 J. Phys. Chem. A, Vol. 106, No. 46, 2002
LaVerne and Araos
1
29
experiments and additional data using H (LETo ) 12 eV/nm),
4
12
He (LETo ) 99 eV/nm), and C (LETo ) 686 eV/nm) ions
all at 10 MeV incident energy are shown in Figure 1. It can be
seen that iodobenzene yields increase with increasing iodine
concentration with similar results for γ-rays, protons, and helium
ions. An increase is also observed with carbon ions, but the
highest yield at 30 mM iodine is only 60% of that observed
with γ-rays.
At the maximum iodine concentration of 30 mM, the
iodobenzene yield is 0.32 molecule/100 eV with γ-rays. The
limiting phenyl radical yield at infinite iodine concentration is
expected to be about 0.37 radical/100 eV, which agrees very
4
well with the total radical yield of 0.72 radical/100 eV. Total
radical yields include both the parent radical and the H atom or
its adduct. The results suggest that decomposition of the benzene
excited state gives only phenyl radicals and H atoms, reaction
1. Formation of other radicals can be considered to be negligible.
The lower yields for iodobenzene with carbon ions can be
Figure 1. Iodobenzene yields as a function of iodine concentration
1
4
12
attributed to faster radical chemistry or lower initial radical
yields. Decreasing radical yields with increasing LET were
for 10 MeV ions: (9) H, (b) He, and (2) C ions, this work, and
[) γ-rays, ref 3.
(
13-15
observed with cyclic alkanes.
Carbon ions of 10 MeV
energy have a track average LETo of 686 eV/nm compared to
the 0.2 eV/nm for γ-rays. Radical-radical reactions are
enhanced in carbon ion tracks because of the increase in local
radical concentrations.30 The rate of iodine reaction with phenyl
radicals is not known, but it can be assumed to be similar to
10
-1 -1 31
cyclohexyl radicals in cyclohexane, 1.2 × 10
M
s . The
scavenging capacity is defined as the product of the rate
coefficient for the scavenging reaction and the scavenger
concentration. Scavenging capacities for the phenyl radical by
iodine are shown on the upper axis of Figure 1. The inverse of
the scavenging capacity is equal to the lifetime of the reactive
species. At an iodine concentration of 30 mM, the lifetime of
phenyl radicals is about 3 ns. About 40% of the total phenyl
radicals produced in 10 MeV carbon ion tracks have reacted in
this very short time, probably in combination reactions with H
atoms. It is not possible from the iodine scavenging experiments
alone to determine if the initial phenyl radical yield is dependent
on LET.
Figure 2. Biphenyl yields as a function of iodine concentration for
1
4
12
1
0 MeV ions: (9) H, (b) He, and (2) C ions, this work, and ([)
γ-rays, ref 3.
Biphenyl Formation. Biphenyl can be formed by the
recombination of phenyl radicals, reaction 5, in a manner
analogous to the dimer formation in the radiolysis of cyclic
the phenyl radical-benzene adduct is relatively long-lived.
Several studies have observed anomalous effects when iodine
is added post irradiation to benzene, which may be due to
alkanes:13-15
3
3-35
reactions of this or other long-lived species.
An increase in the iodine concentration leads to a decrease
in the biphenyl with a concurrent increase in iodobenzene. Iodine
appears to compete effectively with benzene for the scavenging
of phenyl radicals. It can be seen from the data of Figure 2 that
the yields of biphenyl and iodobenzene are equal at an iodine
concentration of about 2 mM for γ-rays, protons and helium
ions. This value suggests that the ratio of the rate coefficients
•
•
C H + C H f (C H )
(5)
(6)
6
5
6
5
6
5 2
•
•
•
C H -C H + R (R ) f (C H ) + RH (RH)
6
5
6
6
6
5 2
It can also be formed by a hydrogen atom transfer reaction from
the phenylcyclohexadienyl radical, reaction 6. Figure 2 shows
the radiation chemical yields of biphenyl as a function of iodine
concentration for γ-rays and 10 MeV protons, helium ions, and
carbon ions. It can be seen that on the addition of 0.1 mM iodine
the yield of biphenyl increases substantially. This increase is
from 0.073 to 0.31 for protons, 0.073 to 0.31 for helium ions,
and 0.066 to 0.21 molecules/100 eV for carbon ions. Iodine
may interfere with the neutralization of the benzene cation or it
may quench the excited benzene molecule, but both processes
3
k4 to k3 is about 5.6 × 10 . Assuming the rate coefficient of
iodine reaction with phenyl radicals is similar to that for
1
0
-1 -1
cyclohexyl radicals in cyclohexane, 1.2 × 10
M
s , gives
a value of k4/k3 of about 2400 when using the estimated rate
6
-1 -1
28
coefficient of 5 × 10 M
s
for reaction 3. Because the
rate coefficient for iodine addition is virtually diffusion con-
trolled, the measured value for the rate coefficient for reaction
2 must be too large by about a factor of 2. The rate coefficient
for the addition of phenyl radicals to benzene is probably about
3
2-34
are likely to lead to a decrease in product formation.
An
6
-1 -1
increase in biphenyl must indicate that iodine is causing the
phenyl radical-benzene adduct to produce more biphenyl than
would normally occur. This result may be due to a simple
hydrogen abstraction reaction leading to biphenyl and HI. The
reaction by such a small amount of iodine would indicate that
2 × 10 M s .
The complementary dependence of biphenyl and iodobenzene
yields on iodine concentration strongly suggests that the main
reaction of the phenyl radical is its addition to benzene.
However, the formation of biphenyl in neat benzene is probably
7

Heavy Ion Radiolysis of Liquid Benzene
J. Phys. Chem. A, Vol. 106, No. 46, 2002 11411
Figure 3. Production of biphenyl as a function of track average LET:
Figure 4. Relative fluorescence yields and inverse relative H
2
yields
1
4
12
4
1
4
7
(
9) H, (b) He, and (2) C ions, this work; (O) He, ref 36; and
as a function of track average LET: (0) H, (O) He, (4) Li, and (])
4
12
(+) He, ref 17. The dotted line is the γ-ray yield.
C ions, fluorescence yields, this work; solid symbols are H
ref 16.
2
yields,
due to a different mechanism than the subsequent reaction of
the phenyl radical-benzene adduct. The yields of biphenyl as
a function of track average LET are shown in Figure 3. There
are small variations for a given type of ion, but the biphenyl
yield in benzene is nearly independent of LET. The LET affects
the local concentrations of reactive species and thereby the
kinetics. The results for biphenyl suggest that it is produced by
a first-order process.
The addition of phenyl radical to benzene is a first-order
process, but this reaction is not overall independent of LET.
Phenyl radicals are expected to undergo second-order reactions
with H atoms leading to a decrease in their yields with increasing
LET. This trend can be observed for the iodobenzene formation
for carbon ions at high iodine concentrations, see Figure 2.
the scavenging capacity for this reaction at 0.2 M iodine, 2.4
9
-1
×
10 s , is not fast enough to compete significantly with H
10 -1
9
atom addition to benzene 1.2 × 10
s
(11.2 M × 1.1 × 10
-1 -1
M
s ). The decrease in molecular hydrogen yields with the
addition of m-terphenyl or iodine is probably due to a decrease
in the yield of the precursor to the H atom.
The most striking feature about benzene radiolysis is that the
molecular hydrogen yields increase by over an order of
magnitude with increasing LET from γ-rays to 10 MeV carbon
1
6
ions. The ratio of the molecular hydrogen to acetylene yields
is observed to be about 2 and independent of LET, suggesting
that these two species are due the same precursor.8
,17,18,35,39,40
Burns and co-workers have long proposed that second-order
reactions of excited states of benzene are responsible for both
1
7,18
Similar results are observed for parent radical formation in cyclic
molecular hydrogen and acetylene.
alkanes.1
3-15
It can also be observed for γ-rays, protons, and
The fluorescent yields in the radiolysis of benzene were
recently measured with a variety of heavy ions.19 These
measurements showed a dramatic decrease in fluorescence yields
with increasing LET suggesting a decrease in the singlet excited
state yield. The original published results found a decrease in
fluorescence followed by an increase at very high LET.
Subsequent experiments found that this increase is due to
scintillation of the mica window induced by passage of the high
LET ions. This extra emission is at relatively long wavelengths
and is not observed with a 300-400 nm band-pass filter, which
corresponds to the region of emission by the singlet excimer of
benzene. The fluorescence yields observed using the same
identical techniques as ref 19 using a band-pass filter or a
titanium window give the scintillation yields observed in Figure
4. Here, the fluorescence yields are given relative to that found
with fast electrons. It can be seen that the fluorescence yields
decrease by an order of magnitude with increasing LET from
γ-rays to 10 MeV carbon ions.
helium ions that at the highest iodine concentrations essentially
all phenyl radicals have been scavenged, but the biphenyl yield
is not zero. For all of the radiations used here, the biphenyl
radical yield never decreases below a value of about 0.075
molecule/100 eV, as shown by the dotted line in Figure 3. This
limiting yield must be due to some other process dominated by
a unimolecular reaction. As discussed below, the singlet excited
state of benzene is very dependent on LET so it is probably
not the source of biphenyl. Biphenyl may be formed in the decay
of a triplet excited state, because it is not known if this state is
dependent on LET. Previous studies on alkanes show that there
is an overall singlet to triplet transition due to random
recombination of spin states in high LET tracks.3 It is not
known if this phenomenon is significant in benzene. Another
alternative for the production of biphenyl is an ion-molecule
reaction of the parent cation with benzene. This reaction is
expected to be slow, but the yield of biphenyl is small and
probably due to a minor process like this.37
6
The excited singlet state of liquid benzene couples with the
4
1
Molecular Hydrogen Production. Molecular hydrogen is
one of the most examined products in the radiolysis of benzene,
especially at high LET.16 The production of molecular hydrogen
in hydrocarbons is usually due to H atom-H atom combination
or H atom abstraction reactions. The latter are not expected to
be significant in benzene.38 Scavenger studies with m-terphenyl
and iodine show that very high solute concentrations are required
to lower molecular hydrogen yields in the radiolysis with low
ground state to give an excimer. The pseudo-first-order rate
1
1
-1 42
for the singlet excimer formation is about 1.2 × 10 s ,
43-45
whereas its lifetime is about 27 ns.
Within a few hundred
picoseconds following the absorption of energy, the singlet
excimer-monomer excited states of benzene are in equilibri-
43
um. Benzene strongly self-absorbs light at wavelengths below
about 280, so it is actually the excimer decay that is monitored
in these experiments. Because the excimer and monomer states
are in equilibrium, it is not important which state is being
observed and they will be treated as one unit. Cutoff filters also
eliminate any emission from higher lying excited states.
3
7,39
LET radiation.
At 0.2 M iodine, the molecular hydrogen
39
yield is decreased 23% with γ-rays. If a rate constant of 1.2
10
-1 -1
×
10
M
s
is assumed for H atom reaction with iodine,

11412 J. Phys. Chem. A, Vol. 106, No. 46, 2002
LaVerne and Araos
ionic pairs formed from the single initial energy loss event. The
tracks of higher LET particles lead to an enhancement of triplet
formation because of the near homogeneous distribution of
electrons and cations within them. At an LET above 100 eV/
nm, cross combination reactions dominate the neutralization
3
6
reactions in alkanes. Specific experiments have not been
performed in benzene, but similar results are expected. For fast
electrons, the initial yield of benzene triplet state due to
neutralization reactions is about 4.2 molecules/100 eV.4 Like
the singlet excited state, the triplet also forms an excimer by
9,50
5
0,51
coupling with a ground-state benzene molecule.
Both the
triplet and singlet excimers are probably in equilibrium with
their respective monomers, and both equilibriums are expected
4
3
to be predominantly in the excimer state. The triplet excimer
5
1
state decays with a lifetime of about 162 ns. Because radical
yields decrease only slightly or remain constant with increasing
LET, they are probably due to the decomposition of the triplet
excimer. The quantum yield for intersystem crossing from the
Figure 5. Schematic representation for the radiolytic decomposition
of benzene.
The inverse of the molecular hydrogen yields relative to that
found with γ-rays, i.e., G(H2)γ-ray/G(H2)heavy ion, is shown in
Figure 4 with the observed decrease in relative fluorescence
yields. The agreement between the relative increase in molecular
hydrogen yields and the decrease in fluorescence yields is
excellent. These results strongly suggest that molecular hydrogen
production is due to the decay of a single singlet excited state
of benzene. It is not possible in these experiments to determine
if it is the singlet state or its precursor that decreases with
increasing LET. Because the acetylene yields track those of
hydrogen as a function of LET, it must also come from the
same singlet excited benzene. Isotopic studies suggest that
molecular hydrogen production involves significant scrambling,
whereas acetylene essentially has a single benzene molecule as
5
1
singlet excimer to the triplet state is 39%, so the total triplet
excited-state yield with γ-rays is expected to be about 4.8. It is
interesting that the measured total radical yield is about 0.36
radical pairs, so most of the triplet states are decaying to ground
without product formation or there are substantial yields of
products not yet determined.
Polymer formation in benzene generally refers to the forma-
tion of trimers and higher order compounds. The yield of
polymer in the γ-radiolysis of benzene is 0.76 molecule/100
eV, which agrees well with the total radical yield and the
4
52
scheme in Figure 5. Polymer yields increase with increasing
LET. This increase is similar to that observed for molecular
53
hydrogen, but polymer formation has not been extensively
studied at high LET. However, it is possible that the triplet
excited state makes a significant contribution to product
formation in the radiolysis of benzene at high LET. Polymer
formation is usually determined by the loss of benzene at
relatively high doses. Direct measurements of specific products,
such as trimers or tetramers, would help clarify the role of the
triplet excited state of benzene at high LET.
3
2
a precursor. Migration of H atoms can be expected from the
excimer-monomer excited-state equilibrium. There is an inverse
correlation between fluorescence yields and molecular hydrogen
yields as a function of LET. Because the molecular hydrogen
yield with γ-rays is very small, the normal mode of singlet decay
is by fluorescence or intersystem crossing to the triplet state. A
second order process must be involved in order to exhibit an
LET dependence, so two excited singlet states must be reacting
together or the singlet state is reacting with a radical. Fluores-
cence is thereby decreased and molecular hydrogen yield
increased. Only one singlet state decomposes to molecular
Conclusions
Phenyl radicals and H atoms make up almost all of the
radicals produced in the radiolysis of benzene with a total yield
of 0.7 radical/100 eV. These radicals add to benzene and
ultimately lead to polymer formation. The phenyl radical-
benzene adduct is relatively long-lived and can react with
scavengers such as iodine to give enhanced yields of biphenyl.
Biphenyl is the predominant single hydrocarbon product in the
radiolysis of neat benzene with a yield of 0.075 molecule/100
eV. Its yield is nearly independent of radiation type and energy
suggesting that its formation in neat benzene is due to a fast
ion-molecule process and not due to phenyl radicals. The
similarity of the LET dependence of the fluorescence yields
with that of the molecular hydrogen yields suggests that the
excited singlet state of benzene is the precursor of molecular
hydrogen. The triplet excited state decays to phenyl radicals
and H atoms. Most of the excited states formed in the
γ-radiolysis of benzene seem to decay to ground without
formation of any product. However, extensive product formation
may occur through triplet and singlet excited states of benzene
at high LET.
hydrogen. The benzene singlet excite state has a lifetime of about
4,45
2
7 ns.4 In order for significant reaction between excited states
to occur, they must diffuse together within this time period.
Benzene Decomposition Scheme. The absorption of energy
by benzene leads to ionization giving an electron and the parent
cation with a lifetime of 10-20 ps.46 Neutralization reactions
then produce the singlet excimer-monomer equilibrium state
discussed above and the triplet excited state. The yield of the
4
4,47
singlet state with γ-rays is about 1.6 molecules/100 eV.
Measurements of molecular hydrogen yields with a variety of
heavy ions extrapolate to a high LET limit of about 0.85
molecules/100 eV.16 Burns et al. report a molecular hydrogen
18
4
8
yield of 1.12 for 21 MeV neon ions, and Boyd and Connor
report a yield of 2.14 with fission fragments, which is probably
an overestimate. Clearly, most of the singlet excited states can
lead to molecular hydrogen production at high LET. There are
also obviously other products formed in conjunction with the
molecular hydrogen and acetylene. These products have not been
identified, and future studies should focus in this area. A simple
overview of the formation of the excited states and the products
from them are shown in Figure 5.
Acknowledgment. The authors thank Prof. J. J. Kolata for
making the facilities of the Notre Dame Nuclear Structure
Laboratory available to us. The Nuclear Structure Laboratory
Neutralization reactions can also lead to the triplet because
of the relaxation of spin or cross combination from the multiple

Heavy Ion Radiolysis of Liquid Benzene
J. Phys. Chem. A, Vol. 106, No. 46, 2002 11413
(24) Sauer, M. C., Jr.; Ward, B. J. Phys. Chem. 1967, 71, 3971.
is supported by the U.S. National Science Foundation. The
research described herein was supported by the Office of Basic
Energy Sciences of the U.S. Department of Energy. This is
contribution NDRL-4387 from the Notre Dame Radiation
Laboratory.
(25) Roduner, E.; Bartels, D. M. Ber. Bunsen-Ges. Phys. Chem. 1992,
9
6, 1037.
(
26) Fessenden, R. F.; Schuler, R. H. J. Chem. Phys. 1962, 38, 773.
(27) Zubarev, V. E. Dokl. Phys. Chem. 1980, 253, 667.
(28) Madhavan, V.; Schuler, R. H.; Fessenden R. W. J. Am. Chem. Soc.
1
978, 100, 888.
Eo
(29) Track average LET (LETo ) 1/Eo ∫0 LET dE) is used in this
References and Notes
manuscript because the measured yields are averaged from the incident
particle energy to zero.
(
1) Roder, M. Aromatic hydrocarbons. In Radiation Chemistry of
(
(
(
(
(
(
(
(
(
(
(
(
30) LaVerne, J. A. Radiat. Res. 2000, 153, 487.
31) LaVerne, J. A.; Wojnarovits, L. J. Phys. Chem. 1994, 98, 12635.
32) Schuler, R. H. Trans. Faraday Soc. 1965, 61, 100.
33) Fellows, A. T.; Schuler, R. H. J. Phys. Chem. 1961, 65, 1451.
34) G a¨ umann, T. HelV. Chim. Acta 1963, 46, 2873.
35) G a¨ umann, T.; Schuler, R. H.; J. Phys. Chem. 1961, 65, 703.
36) LaVerne, J. A.; Brocklehurst, B. J. Phys. Chem. 1996, 100, 1682.
37) Burton, M.; Patrick, W. N. J. Chem. Phys. 1954, 22, 1150.
38) Hardwick, T. J. J. Phys. Chem. 1962, 66, 117.
Hydrocarbons; Foldiak, G., Ed.; Elsevier: Amsterdam, 1981; p 393.
(
(
(
2) Burton, M.; Lipsky, S. J. Phys. Chem. 1957, 61, 1461.
3) LaVerne, J. A.; Araos, M. S. Radiat. Phys. Chem. 1999, 55, 525.
4) Cherniak, E. A.; Collinson, E.; Dainton, F. S. Trans. Faraday Soc.
1
964, 60, 1408.
(
5) G a¨ umann, T. HelV. Chim. Acta 1961, 44, 1337.
(6) Gordon, S.; van Dyken, A. R.; Doumani, T. F. J. Phys. Chem. 1958,
6
39) Schuler, R. H. J. Phys. Chem. 1956, 60, 381.
40) Burns, W. G.; Marsh, W. R. Trans. Faraday Soc. 1969, 65, 1827.
41) Birks, J. B.; Braga, C. L.; Lumb, M. D. Proc. R. Soc. London 1965,
3
A283, 83.
(
(
10) Barson, C. A.; Bevington, J. C. Trans. Faraday Soc. 1959, 55, 1266.
11) Weber, E. N.; Forsyth, P. F.; Schuler, R. H. Radiat. Res. 1955, 3,
(
(
42) Hirayama, F.; Lipsky, S. J. Chem. Phys. 1969, 51, 1939.
43) Burns, W. G.; Cundall, R. B.; Griffiths, P. A.; Marsh, W. R. Trans.
6
8.
(
Faraday Soc. 1968, 64, 129.
12) Hughes, G. J. Phys. Chem. 1961, 65, 2160.
(
13) LaVerne, J. A.; Schuler, R. H.; Foldiak, G. J. Phys. Chem. 1992,
(44) Cooper, R.; Thomas, J. K. J. Chem. Phys. 1968, 48, 5097.
(45) Thomas, J. K.; Mani, I. J. Chem. Phys. 1969, 51, 1834.
(46) Miyasaka, H.; Masuhara, H.; Mataga, N. J. Phys. Chem. 1985, 89,
1631.
(47) Skarstad, P.; Ma, R.; Lipsky, S. Mol. Cryst. 1968, 4, 3.
(48) Boyd, A. W.; Connor, H. W. J. Can. J. Chem. 1964, 42, 1418.
(49) Hentz, R. R.; Perkey, L. M. J. Phys. Chem. 1970, 74, 3047.
(50) Baxendale, J. H.; Fiti, M. J. Chem. Soc., Faraday Trans. 1972, 68,
218.
9
6, 2588.
(
(
(
(
(
14) LaVerne, J. A.; Wojnarovits, L. J. Phys. Chem. 1995, 99, 9862.
15) Wojnarovits, L.; LaVerne, J. A. J. Phys. Chem. 1995, 99, 11292.
16) LaVerne, J. A.; Schuler, R. H. J. Phys. Chem. 1984, 88, 1200.
17) Burns, W. G.; Marsh, W. R. Trans. Faraday Soc. 1968, 64, 2375.
18) Burns, W. G.; Marsh, W. R.; Reed, C. R. V. Nature 1968, 218,
8
67.
(
(
(
19) LaVerne, J. A. J. Phys. Chem. 1996, 100, 18757.
20) LaVerne, J. A.; Schuler, R. H. J. Phys. Chem. 1987, 91, 6560.
21) Ziegler, J. F.; Biersack, J. P.; Littmark, U. The Stopping and Range
(51) Bensasson, R. V.; Richards, J. T.; Thomas, J. K. Chem. Phys. Lett.
1971, 9, 13.
of Ions in Solids; Pergamon: New York, 1985.
(52) Manion, J. P.; Burton, M. J. Phys. Chem. 1952, 56, 560.
(53) Yang, J. Y.; Strong, J. D.; Burr, J. G. J. Phys. Chem. 1965, 69,
1157.
(22) Schuler, R. H.; Allen, A. O. J. Chem. Phys. 1956, 24, 56.
23) Meshitsuka, G.; Burton, M. Radiat. Res. 1959, 10, 499.
(