Decomposition Mechanism for Electron Beam Irradiation of Vaporized Trichloroethylene - Air Mixtures

Article abstract of DOI:10.1021/jp991970f

Decomposition of trichloroethylene (TCE) in electron beam irradiation was examined in order to get information on treatment of industrial off-gas. Air containing vaporized TCE was sealed into a glass vessel and irradiated with an electron beam. The concentration of TCE exponentially decreased with dose (air-absorbed energy). The doses necessary to decompose 90% of input TCE molecules had a value of 3.6 kGy independent of an initial concentrations ranging from 50 to 1800 ppmv. Between G values of decomposition and initial concentrations, the following equation was obtained: G(-TCE) [μmol J^-1] = 1.5 [μmol J^-1] + 0.020 [μmol J^-1 ppmv^-1] × conc [ppmv]. The equation shows that concentrations of Cl radicals remained constant at 6.3 × 10^-4 ppmv (at 298 K and 1 bar) independent of concentrations of TCE and products under a dose rate of 2.1 kGy/s. Termination reactions in the decomposition mechanism for EB irradiation of TCE and tetrachloroethylene (PCE) were also clarified on the basis of the relation of G value of decomposition and the initial concentrations. Dichloroacetyl chloride, which was identified as a primary product, was decomposed and oxidized to give CO and CO₂. Phosgene was also identified as a primary product. From the proportion of DCAC to COCl₂ concentrations, a branching ratio was calculated to be 5.3 in the Cl radical chain reaction of TCE. Formation of HCl and/or Cl₂ was analyzed and the collection of the products from the gas phase was examined with alkaline solution. The effect of water (400-25 000 ppmv) on decomposition efficiency was examined. The decomposition efficiencies had the same value as that in dry air.

Full text of DOI:10.1021/jp991970f

J. Phys. Chem. A 2000, 104, 59-66
59
Decomposition Mechanism for Electron Beam Irradiation of Vaporized
Trichloroethylene-Air Mixtures
Teruyuki Hakoda* and Shoji Hashimoto
Department of Radiation Research for EnVironment and Resources, Takasaki Radiation Chemistry Research
Establishment, Japan Atomic Energy Research Institute, 1233 Watanuki, Takasaki, Gunma, 370-1292, Japan
Yuichi Fujiyama and Akira Mizuno
Department of Ecological Engineering, Toyohashi UniVersity of Technology, 1-1 Hibarigaoka, Tempakucho,
Toyohashi 441-8580, Japan
ReceiVed: June 16, 1999; In Final Form: October 20, 1999
Decomposition of trichloroethylene (TCE) in electron beam irradiation was examined in order to get information
on treatment of industrial off-gas. Air containing vaporized TCE was sealed into a glass vessel and irradiated
with an electron beam. The concentration of TCE exponentially decreased with dose (air-absorbed energy).
The doses necessary to decompose 90% of input TCE molecules had a value of 3.6 kGy independent of an
initial concentrations ranging from 50 to 1800 ppmv. Between G values of decomposition and initial
concentrations, the following equation was obtained: G(-TCE) [µmol J-1] ) 1.5 [µmol J-1] + 0.020 [µmol
J-1 ppmv-1] × conc [ppmv]. The equation shows that concentrations of Cl radicals remained constant at 6.3
× 10-4 ppmv (at 298 K and 1 bar) independent of concentrations of TCE and products under a dose rate of
2.1 kGy/s. Termination reactions in the decomposition mechanism for EB irradiation of TCE and
tetrachloroethylene (PCE) were also clarified on the basis of the relation of G value of decomposition and the
initial concentrations. Dichloroacetyl chloride, which was identified as a primary product, was decomposed
and oxidized to give CO and CO2. Phosgene was also identified as a primary product. From the proportion
of DCAC to COCl2 concentrations, a branching ratio was calculated to be 5.3 in the Cl radical chain reaction
of TCE. Formation of HCl and/or Cl2 was analyzed and the collection of the products from the gas phase was
examined with alkaline solution. The effect of water (400-25 000 ppmv) on decomposition efficiency was
examined. The decomposition efficiencies had the same value as that in dry air.
Introduction
concentrations of VOCs and has the advantage of being per-
formed at ambient temperature. The free radicals and ions, which
are produced from base components (water, soil, and gas) by
ionizing radiation, can directly oxidize and decompose VOC
molecules. For TCE in groundwater, the study of radiation de-
composition has been undertaken as a potential technology.
These applications with ionizing radiation demonstrated the di-
rect decomposition of TCE in water by γ ray irradiation,3,4 as
well as the decomposition of gas-phase TCE stripped from the
contaminated water by electron beam (EB) irradiation.5-7 These
experiments were performed under fixed concentrations of TCE.
Volatile organic compounds (VOCs) have caused widespread
environmental contamination such as groundwater and soil
contamination. Off-gases from factories for semiconductor and
paints also have been contaminated with VOCs. It is well-known
that several kinds of VOCs are very harmful to human health
because of carcinogenic and mutagenic effects. The Environ-
mental Agency of Japan has started to evaluate the toxicity of
various environmental pollutants, including VOCs in order to
build the emission regulations from pollution sources.1
Among these VOCs, trichloroethylene (TCE) has been widely
used as washing and de-oil solvents in various industrial fields.
The pollution from TCE has caused severe problems, especially
in the ground and drinking water or the soil at widely and highly
industrialized areas. The VOCs including TCE have been treated
with adsorption on activated carbon or catalytic combustion.
These conventional methods can effectively and economically
remove some kinds of VOCs from contaminated water and gas
when their concentrations are large.2 However, the removal
efficiencies are relatively decreased in low concentrations of
VOCs. The VOCs adsorbed on the activated carbon are not
converted into nontoxic compounds and have the possibility of
causing new pollution at places different from the original
pollution sources (secondary pollution).
As compared with γ ray irradiation, an EB accelerator can
produce large amounts of radicals and ions from base compo-
nents within a very short time because of the high dose rate. A
large amount of water, soil, and air contaminated with TCE is
expected to be effectively treated with EB. Since the penetration
range of EB in water and a solid is very small in comparison
with that in a gas, EB treatment is suitable for purification of
gas-phase TCE. Low-energy EB accelerators have recently been
used in various industrial fields. It can be operated without a
protection structure for bremsstrahlung X-ray and a skillful
operator. It is cheaper to purchase a low one than a high one.
The process will be more economical by using a low-energy
EB accelerator.
The ionizing radiation decomposition seems to be alternative
technologies for removal of chlorinated hydrocarbon within low
For developing EB treatment of gas-phase TCE, basic
information should be precisely obtained on decomposition
efficiencies under input concentrations ranging from 50 to 1800
ppmv (parts per million by volume) as well as irradiation
* To whom correspondence should be addressed. E-mail: hakoda@
taka.jaeri.go.jp.
10.1021/jp991970f CCC: $19.00 © 2000 American Chemical Society
Published on Web 12/16/1999

60 J. Phys. Chem. A, Vol. 104, No. 1, 2000
Hakoda et al.
products. In the present work, gas-phase TCE was irradiated
with EB. The decomposition efficiencies were examined under
various initial TCE concentrations and water concentrations. The
products were analyzed and the collection with an alkaline
solution was investigated. The decomposition mechanism was
discussed on the basis of product distribution.
Experimental Section
All sample gas was prepared by mixing dry air (<1 ppmv
CO, <1 ppmv CO2, and <1 ppmv CH4) and vaporized TCE
(>99.0% pure liquid TCE without further purification, Kanto
Chemical Co. Ltd.). Pyrex glass vessels were used as an
irradiation reactor.8 The reactors (50 × 50 × 200 mm) have a
volume of 500 mL and two stopcocks (without grease) at both
ends. The sample gases were introduced into the reactors and
were sealed with the stopcocks. The concentrations of TCE in
the sample gases were measured by a gas chromatograph (GC-
8A, Shimadzu Seisakusho Co. Ltd.) with FID and a packed
column (BX-20 100/120, 2 mm i.d. × 3 m, GL Science Inc.),
and the values were adjusted to 50-1800 ppmv. The water
concentrations in the dry air were measured to be 200 ppmv by
a moisture analyzer (Hygro-3M with D-2 detector, General
Eastern Co. Ltd.). Vaporized water from a liquid water bubbling
bottle was added to the sample gas, and the concentrations were
adjusted to 400-25 000 ppmv.
The sample gases in the reactors were irradiated with 2 MeV
electrons generated by an electron beam accelerator (3 MV 25
mA max., Cockcroft-Walton type, Radiation Dynamic, Inc.)
with a constant beam current of 1.0 mA. The reactors were put
on the conveyer for irradiation and passed under the scan horn.
The sample gases in the reactors were exposed at doses ranging
from 2.2 to 40 kGy. An average dose in the reactor was
measured by a CTA film dosimeter and was 2.2 kGy/pass under
these conditions. A calculated value of average dose rate was
2.1 kGy/s.
Figure 1. Concentrations of residual TCE and the products versus
dose for EB irradiation of 650 ppmv TCE: (O) TCE; (4) dichloroacetyl
chloride (DCAC); (3) CO; (0) CO2; (]) phosgene (COCl2); (∞) TC
1
(total carbon concentration) × /2.
Ozone (O3) was analyzed by an ozone analyzer (model 4560
Sensor STIK with Type 4501 Monitor, EIT).
Result
Figure 1 shows concentrations of residual TCE and products
versus dose at an initial concentration of 650 ppmv. Products
were identified as dichloroacetyl chloride (CHCl2COCl, DCAC),
carbon monoxide (CO), carbon dioxide (CO2), phosgene
(COCl2), and small amount of chloroform (CHCl3). The concen-
tration of total carbon (TC) remained constant up to 40 kGy so
that any products containing carbon did not attach on the inner
surface of the reactor. The concentration of DCAC increases
and reaches a value of 390 ppmv (62%v of initial TCE
concentration) at doses ranging from 0 to 4.4 kGy where TCE
is decomposed. Then, the concentration gradually decreases at
more than 4.4 kGy. For CO and CO2, the concentrations rapidly
increase up to 2.2 kGy and linearly increase at doses ranging
from 2.2 to 40 kGy. For COCl2, the concentration increases at
doses ranging from 0 to 4.4 kGy and remains constant (ca. 90
ppmv). The concentration of CHCl3 was measured to be 5 ppmv
(0.8% v of initial TCE concentration) at 20 kGy.
After irradiation, a rubber cap was attached to one side of
the connection tubes of the reactor and the air between the rub-
ber cap and the stopcock was evacuated by a syringe. Then,
the stopcock was opened and a syringe sampled 2.0 mL of the
inner gas.
Analyses of residual TCE and products were carried out as
follows.
Figure 2 shows the carbon concentrations for TCE and
products in the gas phase with and without NaOH solution.
Figure 2A (without NaOHaq) corresponds to Figure 1. In this
figure, the values in parentheses denote mass yield. The mass
yield stands for the proportion of summation of the carbons
contained in residual TCE and product molecules to input TC.
These values are close to 100% so that all carbon compounds
could be identified. As shown in Figure 2B (with NaOHaq),
DCAC, CO2, and COCl2 could be dissolved in NaOH solution.
These compounds can be mineralized by hydrolysis and
converted into sodium dichloroacetate (from DCAC) and NaCl
and Na2CO3 (from CO2 and COCl2), respectively. The smallest
mass yield after treatment with NaOHaq occurred at 4.4 kGy,
which is where the largest removal efficiency of total VOCs
including products was obtained. Trichloroethylene was almost
completely decomposed at this dose. At more than 4.4 kGy,
the efficiency gradually decreased with an increase in the
concentration of CO. The solutions were also analyzed by the
ion chromatograph to measure chloride ions (Cl-). The result
is shown in Table 1. The expected concentration denotes the
concentration of Cl- ions produced from DCAC, COCl2, and
HCl and/or Cl2 when these products were dissolved in 10 mL
of NaOH solution.
Trichloroethylene, dichloroacetyl chloride (DCAC), and
chloroform (CHCl3) were analyzed by the gas chromatograph.
Other organic gaseous products were barely observed. Total
carbon (TC) and carbon dioxide (CO2) were analyzed by a total
organic carbon (TOC) analyzer (TOC-5000A, Shimadzu Sei-
sakusho Co. Ltd.). The TOC analyzer can convert all carbon
compounds into CO2 in a heating tube (923 K) with a catalyst.
The resulting CO2 formed in a TOC analyzer corresponds to
TC. Carbon monoxide (CO) was analyzed by a CO analyzer
(TH-D4 detector with TA-470 Monitor, Komyo Co. Ltd.).
Phosgene (COCl2) was indirectly analyzed by a TOC analyzer.
When COCl2 is dissolved in phosphoric acid solution (25% w
H3PO4), CO2 is produced from COCl2 in the solution and
released into the gas phase. The concentration of COCl2 was
calculated from the difference in CO2 concentrations with and
without H3PO4 solution. Hydrochloric acid and/or Cl2 were
analyzed by an ion chromatograph (LC-10A, Shimadzu Sei-
sakusho Co. Ltd.) with a conductivity detector (CDD-6A,
Shimadzu Seisakusho Co. Ltd.) and a column (Shim-Pack IC-
A3, Shimadzu Seisakusho Co. Ltd.). The reactors were washed
with NaOH solution (0.1 M) of 10.0 mL to mainly scavenge
Cl2 and HCl. A part of the solution was analyzed for Cl- ions.

Electron Beam Irradiation of Trichloroethylene
J. Phys. Chem. A, Vol. 104, No. 1, 2000 61
Figure 2. Distribution of TCE and the products with and without NaOH solution (0.1 M, 10 mL). The values in parentheses stand for the relative
carbon concentration in the gas phase.
The symbol star (*) stands for an excited state.
The charge transfer reaction mainly occurs between N2+ and
O2 molecules.
+
N2+ + O2 f N2 + O2
(3)
The O2+ ions formed in reactions 2 and 3 produce H2O cluster
ions.
O2+ + H2O f O2 -H2O+
(4)
The resulting O2-H2O+ ions cause the dissociation reaction of
water cluster ions to produce OH radicals
O2-H2O+ + H2O f H3O+ + O2 + OH•
(5)
Figure 3. Concentrations of TCE versus dose in EB irradiation.
O2-H2O+ + 2H2O ff H3O+ + H2O + O2 + OH• (6)
TABLE 1: Concentration of Cl- Ion in Solution for EB
Irradiation of 650 ppmv TCE
Among O radicals produced in reaction 2, O(1D) radicals can
react with H2O molecules and produce OH radicals.
dose
(kGy)
DCAC
(ppmv)
expected Cl-
(10-3 mol/l)
measured Cl-
(10-3 mol/l)
mass yield
(%)
O(1D)• + H2O f 2OH•
(7)
2.2
4.4
13.1
26.2
35.3
298.0
412.0
345.2
284.9
222.5
2.76
2.35
2.54
2.82
3.08
2.68
2.15
2.36
2.76
3.10
97.1
91.5
92.9
97.9
100
Secondary (thermal) electrons (ethr-) formed in reactions 1 and
2 are scavenged with O2, H2O, and TCE molecules
ethr- + O2 + M f O2- + M
(8)
(9)
Figure 3 shows concentrations versus dose for initial con-
centrations ranging from 50 to 1800 ppmv. A large initial con-
centration gives a large slope of the decomposition curve within
a low dose. The doses necessary to decompose 90% of input
TCE molecules (Dose90%) were calculated from decomposition
curves. Most of the values of Dose90% were calculated to be
3.6 ( 0.1 kGy. For 408 and 1834 ppmv, the relative concentra-
tions of TCE and products showed the same distributions as
those at 650 ppmv. The reason will be discussed in the next
section.
-
O2- + H2O + M f O2 H2O + M
ethr- + TCE f Cl- + products
(10)
Here the symbol M stands for a third body (such as N2).
These ions are neutralized to produce H, HO2, and O radicals
and O3 molecules.
-
H3O+ + O2 H2O f H• + O2 + 2H2O
(11)
(12)
Discussion
Formation of Activated Species. When air containing
vaporized water is irradiated with ionizing radiation, OH, HO2,
O, N, and H radicals and O3 molecules are produced as activated
species. Electron beam irradiation of the sample gas can cause
dissociation and ionization of the base gas (N2 and O2).9-12
•
f HO2 + 2H2O
O2+ + O2- f O• + O3
f O2* + O2
(13)
(14)
e- + N2 f e- + (N2+ + ethr-), 2N•, N2*
e- + O2 f e- + (O2+ + ethr-), 2O•, O2*
(1)
(2)
f 2O• + O2
(15)
(16)
O• + O2 + M f O3 + M

62 J. Phys. Chem. A, Vol. 104, No. 1, 2000
Hakoda et al.
(23)
Cl• + C2HCl3 f C2HCl4
•
•
2C2HCl4 f chain termination
(24)
C2HCl4 + O2 f C2HCl4OO•
(25)
•
C2HCl4OO• f C2HCl4 + O2
(-25)
•
C2HCl4OO• + C2HCl4 f C2HCl4O2C2HCl4 (26)
•
2C2HCl4OO• f C2HCl4O2C2HCl4 + O2
2C2HCl4OO• f 2(C2HCl4O•) + O2
C2HCl4O• f CHCl2COCl + Cl•
(27)
(28)
(29)
(30)
Figure 4. Concentrations of O3 versus time after EB irradiation. (4,
Original) and (dash line, mean value): for 250 ppmv TCE. (O, Original)
and (solid line, mean value): for 1750 ppmv TCE.
•
f CHCl2 + COCl2
A part of the ozone is decomposed to O2 molecules through
the reaction with O radicals.
The CHCl2 radicals given in reaction 30 can react with O2
molecules to produce CHOCl molecules and ClO radicals. The
ClO radicals would react with TCE molecules to produce Cl
radicals.21
First Step of Decomposition. In the decomposition process
of TCE, OH radicals and O3 molecules are dominant chemical
oxidizing species attacking them. For decomposition of TCE
in EB irradiation, the reactions with these activated species as
well as reaction 10 are regarded as the first step of the
decomposition reactions.13-15
CHCl2 + O2 f ClO• + CHClO
(31)
(32)
•
ClO• + C2HCl3 f CHCl2COCl + Cl•
For the reaction with the OH radicals [reaction-OH],13
OH• + C2HCl3 f C2HCl3OH•
(17)
(18)
The concentration of chloroform (CHCl3), which would be
produced from the reaction of CHCl2 radicals and Cl radicals,
was measured as 5 ppmv at 20 kGy in a 650 ppmv TCE-air
mixture. Therefore, most of CHCl2 radicals formed via reaction
30 would react with O2 molecules to consequentially produce
Cl radicals and DCAC molecules via reactions 31 and 32. The
resulting Cl radicals from reactions 29 and 32 can reenter the
same oxidation of TCE molecules.
As a result, the oxidation of TCE molecules mainly proceeds
by the Cl radical chain process. From the present results (Figure
1), it can be concluded that DCAC is directly produced from
TCE as primary products and decomposed with further oxidation
to produce CO and CO2. The Cl radical chain reaction seems
to involve reactions 24, 26, and 27 as chain termination.
Effect of Water. Figure 5 shows concentrations of TCE,
DCAC (dichloroacetyl chloride), and TC (total carbon) at water
concentrations varying from 400 to 25 000 ppmv for a 650 ppmv
TCE-air mixture. Under 400 and 1000 ppmv H2O, the
decomposition efficiencies and the concentration of DCAC have
the same value as those in dry air. On the other hands, under
25 000 ppmv, the decomposition efficiency is slightly larger
than that in dry air.
The vaporized H2O in air would give a decrease in decom-
position efficiency of TCE. The clusters and fragments from
H2O molecules can scavenge Cl radicals and result in a decrease
in the concentration of Cl radicals (minus effect). These clusters
and fragments can also give an increase in the decomposition
efficiency of TCE by formation of more OH radicals (plus
effect). Because the decomposition efficiencies under 400 and
1000 ppmv H2O have the same values as in dry air, a plus effect
can counteract a minus effect in the decomposition mechanism.
Under 25 000 ppmv H2O, the value was slightly larger than
that in dry air. The concentrations of TC and DCAC were
decreased with the dose. The decrease in concentrations of these
products suggests that DCAC was adsorbed on the inner surface
of glass reactors and removed from the gas phase under wet
C2HCl3OH• + O2 f OOC2HCl3OH•
2(OOC2HCl3OH•) f 2(OC2HCl3OH•) + O2
(19)
(20)
OC2HCl3OH• f C2HCl2OH(O) + Cl•
For the reaction with O3 molecules [reaction-O3],15
O3 + C2HCl3 f CHClOO• + COCl2
(21)
(22)
CHClOO• f Cl• + products
The main decomposition of TCE proceeds by Cl radical chain
reaction. Therefore, reaction 10 producing Cl- would be a minor
reaction in the decomposition mechanism. The reaction rate
constants of TCE molecules with OH radicals and O3 molecules
are (2.0 ( 0.4) × 10-12 and <3.0 × 10-20 cm3 molecule-1 s-1
,
respectively.16,17 There is no large difference in formation of
10,11
18
OH radicals
and O3 in EB irradiation. Therefore, the
reaction with OH radicals can be assumed to be the first step
of the decomposition reaction. Figure 4 shows the changes in
the concentrations of O3 formed at 2.2 kGy for 40 min after
the irradiation. The concentrations of O3 remained constant for
40 min and had the same value as the concentration of O3 in
EB irradiation of air without TCE. The concentrations of TCE
and products also did not change for 40 min after irradiation.
These results show that OH radicals mainly cause the first step
of decomposition reaction.
According to the decomposition mechanism proposed by
Huybrechts et al.,19,20 Cl radicals can attach to the double bond
of TCE molecules to cause the oxidation of TCE (Cl radical
chain reaction).

Electron Beam Irradiation of Trichloroethylene
J. Phys. Chem. A, Vol. 104, No. 1, 2000 63
Figure 7. Concentrations of TCE and PCE versus relative dose in EB
irradiation: (line) original data; (mark) data for a run with an initial
concentration “C0” plotted where the dose is offset from zero by the
dose necessary to reduce the 1800 ppmv sample to concentration “C0”.
The value of C0 denotes the following concentrations of TCE and PCE.
TCE: (∞) 55, (3) 408, (4) 650, (]) 755, (0) 1425, (O) 1800 ppmv.
PCE: (∞) 150, (3) 322, (4) 618, (]) 880, (0) 1440, (O) 1800 ppmv.
Figure 5. Concentrations of TCE and products versus dose in EB
irradiation under different water concentrations: (O) dry air; (0) [H2O]
) 400 ppmv; (4) 1000 ppmv; (]) 25 000 ppmv; (solid marks) TCE
concentrations.
Figure 6. Concentrations versus dose for EB irradiation of TCE and
PCE. The results for TCE and PCE are shown in the upper and lower
figures, respectively. TCE: (∞) 55, (3) 408, (4) 650, (]) 755, (0)
1425, (O) 1800 ppmv. PCE: (∞) 150, (3) 322, (4) 618, (]) 880, (0)
1440, (O) 1800 ppmv.
Figure 8. G values of decomposition versus initial concentrations for
TCE and PCE in EB irradiation: (O) TCE; (0) PCE; (solid line) 1.5
+ 0.020 × conc; (dash line) 1.5 + 0.021 × conc; (dotted line) 10.0 +
0.069 × conc2/3
.
conditions. Under this condition, OH radicals can effectively
react with residual TCE molecules without reacting products.
Estimation of G Values. To quantitatively evaluate the
decomposition efficiencies of TCE under various initial con-
centrations, the G value of decomposition was introduced. The
value [number/100 eV] is defined as the number of decomposed
TCE molecules when the sample gas absorbed 100 eV. In the
present work, the G values are represented as the unit of µmol
J-1 ()(1/0.1036) × [number/100 eV], at 298 K and 1 bar)
instead of [number/100 eV]. Because the concentration of
products is much smaller than that of residual TCE, the reaction
of the activated species with products can be neglected at this
small absorbed energy. Therefore, G values of decomposition
can indicate the decomposition efficiency of TCE itself. The
decomposition curves shown in Figure 3 can be well fitted with
an exponential function (see Figure 6). The G values of
decomposition (G(-TCE)) can be calculated from this fitting
equation and are proportional to the slope of the exponential
function at 0 kGy. Using the previous results for tetrachloro-
ethylene (PCE),22 an exponential equation cannot fit the decom-
position curve at doses ranging from 0 to 12 kGy (see Figure
6). For the comparison with the values for TCE, G(-PCE)
values were also calculated from the slope of decomposition
curve at doses ranging from 0 to 2.2 kGy. As shown in Figure
7, decomposition curve under each initial concentration agrees
well with that under an initial concentration of 1800 ppmv for
TCE and PCE. Therefore, the decomposition efficiencies under
each initial concentration will be affected by only the concentra-
tion of residual TCE or PCE and little affected by formation of
the products. The value of G(-PCE) calculated from the data
at doses ranging from 0 to 2.2 kGy can also indicate the
decomposition efficiency of PCE itself at all dose ranges. The
values for TCE and PCE under each initial concentration are
also shown in Figure 8. As shown in this figure, G(-TCE)
seems to linearly increase with the initial concentration at
concentrations ranging from 50 to 1800 ppmv. The values of
G(-TCE) can be fitted with a following equation.
G(-TCE)[µmol J-1] ) C + l1 × conc [ppmv-1] (E1)
where conc is the initial concentration of TCE. C ) 1.5 [µmol
J-1] and l1 ) 0.020 [µmol J-1 ppmv-1] at concentrations ranging
from 50 to 1800 ppmv.
The chain length in the Cl radical chain reaction can be
represented as a ratio of G(-TCE) to the G value of formation
of the OH radicals (G(OH)). As G(OH) is reported as 0.44 [µmol
J-1],10,11 the chain length can be estimated as the value of 85
()37.3/0.44) at a concentration of 1800 ppmv. High initial
concentration enhances the reaction possibility of Cl radicals
with TCE molecules and results in an increase in chain length.

64 J. Phys. Chem. A, Vol. 104, No. 1, 2000
Hakoda et al.
k28kOH[OH][TCE]0
Slope and Intersect of Relationship of G(-TCE) and
Initial Concentration. During EB irradiation, the decrease in
TCE concentration can be described as the following equation
from reactions 17 and 23.
d
dt
([DCAC] + [COCl2]) )
(E9)
k27
The symbol of kOH denotes the overall reaction rate constant in
which OH radicals react with TCE molecules and consequently
produce Cl radicals via reactions 17-20.
On the other hands, if chain termination exclusively occurred
through reaction 26, the initial formation of DCAC and
phosgene molecules can be described by the following equation.
d[TCE]
-
) k17[OH][TCE] + k23[Cl][TCE] (E2)
dt
The dose rate (I) is represented as follows.
dD
d
dt
I )
(D: dose [kGy ) kJ kg-1])
(E3)
(E4)
([DCAC] + [COCl2]) )
dt
With eq E3, eq E2 can be represented as
2
1/3
k25 k28
(kOH[OH][TCE]0[O2])2/3 (E10)
2
(
)
2k26
(k17[OH] + k23[Cl])[TCE]
d[TCE]
dD
-
)
I
The initial oxidation (or decomposition) of TCE molecules has
the same value as the initial formation of DCAC and phosgene
molecules. Therefore, the G value of decomposition of TCE
can be described as next equation so long as the energy (100
eV) is going into air and none is going into the TCE molecules.
If chain termination is exclusively reaction 27, eq E11 can
be obtained.
For OH radicals the reaction rate can be described as follows.
d[OH]
dt
1
)
G(OH) × I - k17[OH][TCE] (E5)
0.0346
If the steady-state treatment can apply to OH radicals, eq E4
can be described as
2k28kOH[OH]
G(-TCE)
[TCE]0
(E11)
Ik27
k23[Cl][TCE]
d[TCE] G(OH)
dD
-
)
+
(E6)
0.0346
I
If chain termination is exclusively reaction 26, eq E12 can be
obtained.
The G value of decomposition for TCE has the following
relationship with eq E5.
2
1/3
k25 k28
(kOH[OH][O2])2/3
2
(
)
d[TCE]
∆N
∆E
2k26
2/3
G(-TCE) )
≈ -0.0346
)
t)0
|
G(-TCE)
[TCE]0
(E12)
dD
I
k23[Cl]
G(OH) + 0.0346
[TCE]0 (E7)
As shown in Figure 8, G(-TCE) is proportional to the initial
concentration of TCE. Therefore, reaction 27 seems to mainly
occur as chain termination reaction in the decomposition
mechanism of TCE.
I
where ∆N is the number of decomposed TCE molecules, ∆E
is 100 eV ) 1.6 × 10-17 J, and [TCE]0 is the initial
concentration of TCE.
The G value of formation for OH radicals should correspond
to the intercept of eq E1. The intercept of eq E1 ()1.5 µmol
J-1) had a large value in comparison with G(OH) ()0.44 µmol
J-1). The difference in two values suggests that decomposition
of TCE molecules should be also caused by the activated species
except for OH radicals, such as O radicals, ClO radicals given
in reaction 31, and thermal electrons formed via reaction 10.
Comparison with G(-PCE). As shown in Figure 8, the
values of G(-PCE) nonlinearly increase with initial concentra-
tion. At initial concentrations ranging from 150 to 700 ppmv,
the values of G(-PCE) have the same values of slope ()0.021
µmol J-1 ppmv-1) and intercept ()1.5 µmol J-1) as those of
G(-TCE). However, the values of G(-PCE) at more than 700
ppmv deviated from those of G(-TCE). The deviation can be
explained by the difference in the kinds of chain termination.
For gas-phase chlorine-photosensitized oxidation of PCE, Huy-
brechts et al.23 also proposed the same Cl radical chain reaction
as the oxidation of the gas-phase PCE and Cl2 mixture.
The slope of eq E1 corresponds to the value of 0.0346 ×
k23[Cl]/I.
k23[Cl]
l1 ) 0.0346
(E8)
I
The values of l1 () 0.020 µmol J-1 ppmv-1), I ()2.1 kGy s-1
) 2.7 × 10-3 J cm-3 s-1), and k17 ()7.8 × 10-11 cm3
molecule-1 s-1 13
) are inserted into eq E8: the concentration of
Cl radicals can be calculated as 6.3 × 10-4 ppmv (at 298 K
and 1 bar). The slope of eq E1 suggests that the concentrations
of Cl radicals remained constant at 6.3 × 10-4 ppmv within a
small dose range (100 eV) under different initial concentrations
and a constant dose rate of 2.1 kGy/s. At all dose ranges, the
Cl radicals also have the constant concentration of 6.3 × 10-4
ppmv, because the decomposition curve under each initial
concentration agrees well with that under an initial concentration
of 1800 ppmv (see Figure 7). For concentrations ranging from
50 to 1800 ppmv, the Cl radicals have the constant concentration
under decomposition process at doses ranging from 0 to 12 kGy.
The Cl radical chain oxidation of TCE molecules from
reactions 23-32 can predict the initial formation rate of DCAC,
and phosgene molecules can be described by the following
equation, if chain termination exclusively occurred through
reaction 27.
•
Cl• + C2Cl4 f C2Cl5
(23′)
(24′)
(25′)
•
2C2HCl5 f chain termination
C2Cl5 + O2 f C2Cl5OO•
•
C2Cl5OO• f C2Cl5 + O2
(-25′)
•

Electron Beam Irradiation of Trichloroethylene
J. Phys. Chem. A, Vol. 104, No. 1, 2000 65
C2Cl5OO• + C2Cl5 f C2Cl5O2C2Cl5
(26′)
(27′)
(28′)
•
2C2Cl5OO• f C2Cl5O2C2Cl5 + O2
2C2Cl5OO• f 2(C2Cl5O•) + O2
Application of the oxidation mechanism to EB irradiation of
PCE can lead to the following equations.
If chain termination is exclusively reaction 27′, eq E13 can
be obtained.
k28′kOH′[OH]
G(-PCE)
[PCE]0
(E13)
Ik27′
If chain termination is exclusively reaction 26′, eq E14 can be
obtained.
Figure 9. Concentrations of the products versus dose for 390 ppmv
DCAC, 50 ppmv TCE, and 200 ppmv Cl2 in EB irradiation: (O) TCE;
(4) dichloroacetyl chloride (DCAC); (3) CO; (0) CO2; (]) phosgene;
(∞) TC (total carbon concentration) × /2.
1/3
k25′2k28′
1
(kOH′[OH][O2])2/3
2
(
)
2k26′
produce DCAC and phosgene molecules. The increases in
concentrations of DCAC and phosgene at doses up to 2.2 kGy
were calculated as 136.4 and 25.7 ppmv/kGy, respectively. It
was elucidated that a branching ratio was estimated to be 5.3
in the Cl radical chain reaction for TCE ()reaction 29/reaction
30).
G(-PCE)
[PCE]0 (E14)
I
Accurate reaction rate constants indicated at the oxidation
mechanism for TCE and PCE are not available. If the reaction
rate constant for PCE molecules has the same value as those
for TCE ones, the relation between the value of G(-PCE) and
initial concentrations ranging from 150 to 700 ppmv would
correspond to eq E13. At these initial concentrations, reaction
Conclusion
The main results of the present work can be summarized as
follows:
•
27′ will exclusively occur as chain termination. If k26′[C2Cl5 ]
is slightly larger than k27′[C2Cl5OO•], not only reaction 27′ but
also reaction 26′ can start to occur as chain termination for a
long chain length (or large initial concentration) range. Two
kinds of chain termination (reactions 26′ and 27′) would give
different slopes of change of G(-PCE).
(1) Dichloroacetyl chloride (DCAC), CO, CO2, phosgene, and
a small amount of chloroform were identified as products.
(2) With an alkaline solution (0.1 M NaOH), the products
except for CO were completely removed from the gas phase.
The largest removal efficiency was obtained at 4.4 kGy for EB
irradiation of 650 ppmv TCE.
(3) The Cl radical chain reaction occurred as the main
decomposition mechanism. Dichloroacetyl chloride, a primary
product, was decomposed and converted into CO and CO2 with
further oxidation. Phosgene was also identified as a primary
product. The branching ratio for formation of DCAC and
phosgene was estimated as 5.3.
(4) Between G values of decomposition and initial concen-
trations for TCE, the following relationship was obtained:
G(-TCE) [µmol J-1] ) 1.5 [µmol J-1] + 0.020 [µmol J-1
ppmv-1] × conc [ppmv] for concentrations ranging from 50 to
1800 ppmv. The slope of the equation showed that concentra-
tions of Cl radicals remained constant at 6.3 × 10-4 ppmv (at
298 K and 1 bar), independent of concentrations of TCE and
products under a dose rate of 2.1 kGy/s.
(5) The values of G(-TCE) linearly increased with initial
concentrations up to 1800 ppmv. Those of G(-PCE) linearly
increased with the same slope of G(-TCE) up to 700 ppmv.
At more than 700 ppmv, the values deviated from those of
G(-TCE). The difference in a change of G values of decom-
position for TCE and PCE can be contributed to the kind of
chain termination.
Formation of Products. Previous workers2,24 reported that
dichloroacetyl chloride (DCAC) molecules, produced directly
from TCE molecules, react with Cl radicals and consequently
produce phosgene molecules. To clarify the mechanism of
DCAC decomposition and phosgene formation, the following
experiment was carried out. Air containing 50 ppmv TCE, 390
ppmv dichloroacetyl chloride (DCAC), and 200 ppmv Cl2 were
irradiated with EB to examine the formation mechanism of
carbon monoxide (CO), carbon dioxide (CO2), and phosgene
(COCl2). The components of the sample gas, except phosgene,
are reproductions of those of the irradiated gas obtained at 2.2
kGy in EB irradiation of the 650 ppmv TCE-air mixture. The
result is shown in Figure 9. The concentrations of CO and CO2
increase with a decrease in concentration of DCAC. The
decrease in DCAC concentration and the increase in CO and
CO2 were calculated as 5.2, 6.0, and 6.0 ppmv/kGy, which had
the same values as those in the TCE-air mixture. These values
suggest that DCAC molecules produced as primary products
from TCE molecules should be decomposed to produce CO and
CO2 molecules with further oxidation.
O2
CHCl2COCl f f CO + CO2 + HCl + Cl2
(3)
(6) The decomposition efficiencies under 400 and 1000 ppmv
H2O have the same values as those in dry air. Under 25 000
ppmv H2O, that value was slightly larger than that under dry
air, because the products were adsorbed on the inner surface of
the reactors and removed from the gas phase.
Phosgene molecules are supposed to be directly produced from
TCE molecules, because formation of phosgene was barely
observed in the TCE-DCAC-Cl2-air mixture. From these
results, phosgene molecules seem to be directly produced from
TCE molecules via reaction 30. The Cl radical initiated chain
reaction of TCE has two main paths in the decomposition
mechanism. The Cl radicals can react with TCE molecules to
Acknowledgment. The authors thank Dr. H. Arai and Mr.
K. Hirota of the Japan Atomic Energy Research Institute and

66 J. Phys. Chem. A, Vol. 104, No. 1, 2000
Hakoda et al.
NEDO Symposium on Non-Thermal Plasma Technology for Air Contami-
nant Control, Tokyo, Nov 1, 1996; NEDO (New Energy and Industrial
Technology Development Organization): Tokyo, 1996.
(10) Willis, C.; Boyd, A. W. Int. J. Radiat. Phys. Chem. 1976, 8, 71.
(11) Ma¨tzing, H. AdV. Chem. Phys. 1991, 80, 315.
(12) Zellner, R. Oxid. Commun. 1986, 9, 255.
Mr. H. Kim of Toyohashi University of Technology for their
valuable discussions.
References and Notes
(1) Yagishita, M. J. Resources EnViron. 1997, 32, 413 in Japanese.
(2) Prager, L.; Langguth, H.; Rummel, S.; Mehnert, R. Radiat. Phys.
Chem. 1995, 46 (4-6), 1137.
(3) Matthews, S. M.; Boegel, A. J.; Eccles, S. F.; Homann, S. G.; Rjce,
D. W.; Loftis, J. A.; Jovanovich, M. C.; Caufield, R. A.; Mincher, B. J.;
Meikrantz, D. H.; Murphy, R. J.; Gresham, G. L.; Connoly, M. J. J. Radiat.
Nucl. Chem. 1992, 161 (1), 253.
(13) Kleindienst, T. E.; Shepson, P. B.; Nero, C. M.; Bufalini, J. J. Int.
J. Chem. Kinet. 1989, 21, 863.
(14) Niki, H.; Maker, P. D.; Savage, C. M.; Breltenbach, L. P.; Martlnez,
R. I.; Herron, J. T. J. Phys. Chem. 1982, 86, 1858.
(15) Niki, H.; Maker, P. D.; Savage, C. M.; Breitenbach, L. P. EnViron.
Sci. Technol. 1983, 17 (7), 312A.
(16) Atkinson, R. Int. J. Chem. Kinet. 1987, 19, 799.
(17) Atkinson, R.; Carter, W. P. L. Chem. ReV. 1984, 84, 437.
(18) Willis, C.; Boyd, A. W.; Young, M. J.; Armstrong, D. A.Can. J.
Chem. 1970, 48, 1505.
(19) Huybrechts, G.; Meyers, L. Trans. Faraday Soc. 1966, 62, 2191.
(20) Sanhueza, E.; Hisatsune, I. C.; Heicklen, J. Chem. ReV. 1976, 76
(6), 801.
(21) Chang, W. D.; Senkan, S. M. EnViron. Sci. Technol. 1989, 23, 442.
(22) Hakoda, T.; Zhang, G.; Hashimoto, S. Radiat. Phys. Chem. 1999,
54 (5), 541.
(4) Gehringer, P.; Proksch, E.; Szinovate, W.; Eschewiler, H. Appl.
Radiat. Isot. 1988, 39 (12), 1227.
(5) Matthews, S. M.; Boegel, A. J.; Loftis, J. A.; Caufieled, R. A.;
Mincher, B. J.; Meikrantz, D. H.; Murphy, R. J. Radiat. Phys. Chem. 1993,
42 (4-6), 689.
(6) Mehnert, R.; Klenert, P.; Prager, L. Radiat. Phys. Chem. 1993, 42
(1-3), 525.
(7) Prager, L.; Mehnert, R.; Sobottka, A.; Langguth, H.; Baumann, W.;
Ma¨tzing, H.; Paur, H.-R.; Schubert, J.; Rashid, R.; Taba, K. M.; Schuch-
mann, H.-P.; Sonntag, C. V. J. AdV. Oxid. Technol. 1998, 3 (1), 87.
(8) Hakoda, T.; Yang, M.; Hirota, K.; Hashimoto, S. J. AdV. Oxid.
Technol. 1998, 3 (1), 79.
(23) Huybrechts, G.; Olbregts, J.; Thomas, K. Trans. Faraday Soc. 1967,
63, 1647.
(9) Penetrante, B. M.; Hsiao, M. C.; Bardsley, J. N.; Merritt, B. T.;
Vogtlin, G. E.; Kuthi, A.; Burkhart, C. P.; Bayless, J. R. Proceeding of the
(24) Mill, T.; Su, M.; Yao, C. C. D.; Matthews, S. M.; Wang, F. T. S.
Radiat. Phys. Chem. 1997, 50 (3), 283.