Radiation Physics and Chemistry 85 (2013) 130138

Contents lists available at SciVerse ScienceDirect

Radiation Physics and Chemistry

journal homepage: www.elsevier.com/locate/radphyschem

E-Beam SO2 and NOx removal from ue gases in the presence

of ne water droplets
Ioan Calinescu a, Diana Martin b, Andrezj Chmielewski c, Daniel Ighigeanu b,n
a

University POLITEHNICA of Bucharest, ]149 Calea Victoriei St., 010072 Bucharest, Romania
National Institute for Lasers, Plasma and Radiation Physics, Electron Accelerators Laboratory, P.O.Box: MG-36, ]409 Atomistilor St., 077125 Magurele, Ilfov, Romania
c
Institute of Nuclear Chemistry and Technology, ]16 Dorodna St., 03-195 Warsaw, Poland
b

H I G H L I G H T S
c
c
c
c
c

The medium-energy EB accelerators are proposed for ue gases treatment.

The energy losses in the windows and in the air gap between them are reduced.
To increase the density of the reaction medium and to reduce the penetration depth of EB ne water droplets (FWD) are used.
Determining the energy efciency the favorable effect of the method was demonstrated.
The maximum amount of FWD was determined from the total energy balance of the process.

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 30 May 2012
Accepted 11 October 2012
Available online 20 November 2012

The Electron Beam Flue Gas Treatment (EBFGT) has been proposed as an efcient method for removal of
SO2 and NOx many years ago. However, the industrial application of this procedure is limited to just a
few installations. This article analyses the possibility of using medium-power EB accelerators for offgases purication. By increasing electron energy from 0.7 MeV to 12 MeV it is possible to reduce the
energy losses in the windows and in the air gap between them (transformer accelerators can be applied
as well in the process). In order to use these mid-energy accelerators it is necessary to reduce their
penetration depth through gas and this can be achieved by increasing the density of the reaction
medium by means of dispersing a sufcient amount of ne water droplets (FWD). The presence of FWD
has a favorable effect on the overall process by increasing the level of liquid phase reactions. A special
reactor was designed and built to test the effect of FWD on the treatment of ue gases with a high
concentration of SO2 and NOx using high-energy EBs (9 MeV). By determining the energy efciency of
the process the favorable effect of using FWD and high-energy EB was demonstrated.
& 2012 Elsevier Ltd. All rights reserved.

Keywords:
Flue gases
NOx and SO2 removal
Electron beam
Fine water droplets

1. Introduction
New technologies and processes have promoted our economic
growth and have been remarkably changing our lifestyles. However, technological innovation is not always welcome. It has
impact on the environment and may have effects related to the
peoples health. The emission of inorganic pollutants such as
nitrogen oxides (NOx) and sulfur dioxide (SO2) has been remarkably reduced by application of state-of-the-art technologies
during the past 60 years. However, industrialized countries have
been using the big amounts of oil and coal for energy production
n

Corresponding author. Tel.: 40 214574346; fax: 40 214574243.

E-mail addresses: calin@tsocm.pub.ro (I. Calinescu),
chmielewski@ichtj.waw.pl (A. Chmielewski),
daniel.ighigeanu@inpr.ro (D. Ighigeanu).
0969-806X/$ - see front matter & 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.radphyschem.2012.10.008

via combustion of fossil fuels to convert chemical energy of the

substrate into electricity, vehicles fuels and process heat to be
used in industries.
Although the developing countries also try to increase the
consumption of environment-friendly energies instead of fossil
fuels, the steep demand for energy due to the recent economic
growth makes it difcult or even impossible to reduce the usage
of fossil fuels. This causes a large amount of emission of NOx, SO2,
as well as carbon dioxide to the atmosphere (Gaffney et al., 1987;
Kato and Akimoto, 2007; Ramanathan and Feng, 2009; Streets and
Waldhoff, 2000).
While the industrial countries have achieved a sufcient
reduction of NOx and SO2 emissions, it is still an urgent issue
for the developing countries to follow this trend. To meet the
strict regulations established by local governments, the wet lime
scrubber method and the selective catalytic reduction method

I. Calinescu et al. / Radiation Physics and Chemistry 85 (2013) 130138

131

have been applied to treat SO2 and NOx. However, the wet lime
scrubber method requires wastewater treatment, and the catalysts have to be replaced periodically. New technology is expected
for simple and simultaneous treatment processing of the both
pollutants. The electron-beam irradiation process for ue gas
purication (EBFGT) has been proposed as an efcient method
because it has the following advantages (Hatano et al., 2011):







has simultaneous denitrication and desulfurization possibilities,

no wastewater treatment is required,
no expensive catalyst is required,
provides simple process and its operation,
produces protable products.

The application of electron beams to treat ue gases, such as

removing sulfur dioxide, was started by Ebara Corporation in
Japan and in the US (Chmielewski, 2011).
There are only few pilot and industrial installations for ue
gases treatment by electron beam (EB) irradiation (Chmielewski
et al., 2004) operated at this moment. Its drawback, like in other
conventional technologies, is high energy consumption (the
necessary power for the electron beam is around 24% from the
total electrical energy produced by the plant) (Hackman and
Akiyama, 2000; Licki et al., 2003) and the difculties in operation
of very high power accelerators. That is why it is mandatory to
nd new solutions and to develop strategies for diminishing
radiation dose absorbed in the ue gas and to optimize the
process for better uses of EB energy.
Subjected to the electron beam irradiation process, the main
components of ue gases (N2, O2 and H2O), are transformed into
divergent ions and radicals. The primary species include: e  , N2 ,
N , O2 , O , H2O , OH , CO2 , CO , Nn2, On2, N, O, H, OH and CO

(Matzing
and Paur, 1992; Tokunaga and Suzuki, 1988). In the case
of high water concentration, oxidizing radicals HO and HO2 and
excited species such as O(3P) are the most important product
formed (Person and Ham, 1988). In the presence of water
droplets, the radiolytically produced hydrated electron reacts
very fast with the dissolved oxygen to produce the superoxide
(O2 ) radical. Since the O2 has a pKa value of 4.7, it will be
converted to HO2 in acidic medium. HO2 is also produced by the
reaction of H-atom with oxygen. However the yield from this
reaction is only about 0.06 mmol per joule.
N2 ,O2 ,H2 Ovapors_at_high_concn e -HO ,HO2  ,On ,ions,excited-species

1
In the presence of these reactive species, NOx and SO2 from
ue gases are oxidized and produce nitric acid and sulfuric acid,
respectively, as intermediate products. These acids are neutralized with ammonia, giving powders of ammonium nitrate and
ammonium sulfate, respectively (Namba et al., 1998) (Fig. 1).
The total yield of SO2 removal consists of the yield of thermal
and radiation induced reactions that can be written (Chmielewski

et al., 1995a; Matzing

et al., 1993):


ZSO2 Z1 F,T Z2 D, aNH3 ,T
2
The yield of the thermal reaction depends on the temperature
and humidity. The yield of the radiation induced reaction depends
on the dose, temperature and ammonia stoichiometry. The main
parameter in NOx removal is the dose.
The presence of water vapors is mandatory for the efcient
removal of SO2 and NOx from gases. By radiolysis of water the HO
radical is obtained and this radical will oxidize both SO2 and NOx
(Genuario, 2009):
NOHO M-HNO2 M

(3)

Fig. 1. The main reactions used in EBFGT to convert SO2 and NOX (Namba et al., 1998).

NO2 HO M-HNO3 M

(4)

SO2 HO M-HSO3 M

(5)

HSO3 O2-SO3 HO2

(6)

SO3 H2O-H2SO4

(7)

NOHO2 -NO2 HO

(8)

A gas humidity 810 vol% is necessary to obtain the optimal

removal-efciencies of both pollutants (Chmielewski, 2007).
The selection of an adequate accelerator is an important issue
in the process engineering. The primary effect of any ionizing
radiation is based on its ability to excite and ionize molecules, and
this leads to the formation of free radicals, which then initiate
chemical reactions.
Accelerated electron beams suitable for ue gas treatment
have sufcient energy (up to 5 MeV) to affect the electrons in the
atom shell, but not its nucleus, and can therefore only initiate
chemical reactions (Zimek, 1995). Typically, the reactions
initiated by electron beam are extremely fast and are completed
in fractions of a second.
Electrons that are capable of electronically exciting and ionizing molecules, such as N2, O2, CO2, H2O etc., must have energies in
the range from 12 to 16 eV. Such electrons can be produced from
fast electrons by the energy degradation process in solids, liquids,
and gases. These secondary electrons show energy distribution
with the maximum in the range from 50 to 100 eV. In contrast to
fast electrons, exhibiting energies in the keV and MeV range,
secondary electrons are capable of penetrating solids and liquids
only a few nanometers. Consequently, they generate ions, radicals, and excited molecules in droplets along the paths of the
fast electrons (Drobny, 2010). Fig. 2 illustrates schematically the
process of generation of reactive species.
The basic fundamental properties for the EB machine; beam
current and beam energy should be derived from the process
requirements to ensure satisfactory results with minimum capital
and operating costs. The power of EB is determined function of
the ow rate of the treated material and the absorbed dose
needed for the process:


Dave F p P= M=T
9

132

I. Calinescu et al. / Radiation Physics and Chemistry 85 (2013) 130138

Fig. 2. The process of generation of reactive species by high energy electrons.

higher electron energies can provide higher beam powers with

lower beam currents (Cleland, 2007).
Much attention should be paid to the range of the electron beam
penetration. The penetration of high-energy (relativistic) electron
beams in irradiated materials increases linearly with the incident
energy. The electron pathway range also depends on the atomic
composition of the irradiated material. The energy deposition is
caused mainly by collisions of the incident electrons with atomic
electrons. Therefore, materials with higher electron contents (electrons per unit mass) will absorb higher doses near the entrance
surface, but shorter electron penetration ranges (Cleland, 2005).
Fig. 4 shows the depthdose distribution curves for beam
energies between 1.0 and 5.0 MeV in centimeters of water as
derived from Monte Carlo calculations using the ITS3 (Integrated
Tiger Series code) (Cleland, 2004).
For mid-energy (500 keV to 5 MeV) and high-energy (5 MeV to
10 MeV) electron accelerators, it is common to express beam
penetration on the basis of equal-entrance and exit doses in unit
density material. In Fig. 5 are presented the calculated values of
penetration depth (on the basis of practical range R(p) for ue
gases and for ue gases with ne water droplets (FWD), the
volume fraction of water is designated as L(Vaq/Vgas), where Vaq
and Vgas refer to the irradiated volumes for the two phases,
L4  10  3.
The range parameters can be correlated with the incident
electron energy E with sufcient accuracy for industrial applications by using the following linear equations (for polyethylene)
(Cleland, 2005):
Rp 0:510E0:145

10

Electron ranges in other materials can be estimated by multiplying the polyethylene range with the ratio of their CSDA ranges.
Rmaterial Rpolyethylene  CSDAm=CSDApe

Fig. 3. Electron energy losses function of incident electron energy for ue gases
treatment by EB (Cleland, 2007).

where D(ave) is the average absorbed dose in kGy, P is the emitted

radiation power in kW, T is the treatment time in s and M is the
mass of irradiated material in kg.
The factor F(p) is the fraction of emitted power absorbed by the
material, which depends on the size, shape, thickness and density
of the object and the penetrating quality of the radiation. The
energy absorption in the primary and secondary window should
be considered in the case of EBFGT. This quantity is difcult to
measure (by using dosimeters placed inside the reaction vessel),
but it can be calculated by Monte Carlo simulation. It can range
from 0.25 to 0.75, depending on the particular application.
EB energy losses in the beam windows and in the air space
between them are high for the low energyhigh power accelerators that are presently used for air pollutants removal
(Chmielewski et al., 1995b). The total EB energy losses (backscattering, beam windows and air gap) are substantially lower
with higher EB incident energy (see Fig. 3). The useful EB energy
from input energy is around 50% for 0.5 MeV and about 95% above
3 MeV. The lower electrical efciencies of accelerators with
higher energies are partially compensated by the lower electron
energy losses in the beam windows. In addition, accelerators with

11

CSDA ranges for many materials with a wide range of electron

energies can be obtained from ICRU Report 37 or from (http://
www.nist.gov/pml/data/star/index.cfm).
Supposing exhaust gases at a temperature of 60 1C (density
r 1.06  10  3 g/cm3) are treated by an electron accelerators
equipped with a Ti foil (50 mm, r 4.5 g/cm3) window, air gap
15 cm and another Ti foil for the reactor (50 mm), the range of
electrons generated from 750 and 2500 kV are 235 and 1240 cm,
respectively (see Fig. 5). The depth of the reactor in the irradiation
direction must be less than these values to avoid non-irradiation
space in the reactor.

Fig. 4. Electron energy deposition vs. thickness  density in water at 1.0, 1.5, 2.0,
2.5, 3.0, 3.5 4.0, 4.5 and 5.0 MeV incident electron energy (IAEA, 2010).

I. Calinescu et al. / Radiation Physics and Chemistry 85 (2013) 130138

Fig. 5. The penetration depth of electron beam function of electron energy in ue

gases (r 1 g/L) and in ue gases with FWD (L 4n10  3, r 5 g/L).

Fig. 6. The inuence of liquid water fraction (L) in ue gases on the density of ue
gases with FWD and penetration depth for EB of 2.5 MeV.

Although the medium-energy EB (2.5 MeV) are used with high

efciency (with lower total energy losses, see Fig. 3), the penetration depth of these EB is too high (approx. 12.4 m, see Fig. 5). In
order to reduce the penetration depth, it is necessary to increase
the density of the medium. This can be achieved by dispersing
FWD in ue gases. The inuence of volume fraction of water (L) in
ue gases on the density of the gas medium and on the penetration depth of EB with 2.5 MeV is presented in Fig. 6.
Several reasons showing the potential importance of dropletphase oxidation in the EBFGT process can be outlines (Cooper
et al., 1998):

 High driving force for the most oxidation process due to low
Gibbs free energy of the products obtained in liquid phase;

 High solubility of the mixture components (SO2 and the



intermediates NO2, N2O4) resulting in an increase in its

liquid-phase concentrations;
Rapid rates of the aqueous-phase oxidation process initiated
by ionizing radiation.

Regarding the primary partitioning of the dose rates between

the irradiated phases, denoted DRgas and DRaq, one notes (Cooper
et al., 1998) that its magnitude is governed mainly by the ratio of

133

the phases densities, rgas and raq. Since the ratio raq/rgas is typically
Z103, then DRaq/DRgas  raq/rgas Z103, assuming that the stopping
powers of the gas and liquid phase are quite the same. Thus the dose
rate in the droplet is higher than that of the gas by a factor of at least
three orders of magnitude. This means that the ratio of the chemical
reactions rates of the pollutants oxidation in the two phase system
is controlled by the phases densities.
With regard to the portioning of absorbed doses between the two
phases, which determines the extent of pollutant removal during
irradiation, one gets (Cooper et al., 1998): Daq/Dgas Z raq*L/rgas.
When L is quite low (in a typical EBFGT process Lr10  6), it is
unlikely that the radiation-induced reactions in the liquid phase
can a play a signicant role in the removal of pollutants from the
irradiated gas, but if L 10  3 the reactions in liquid phase become
important. For such conditions, the chain oxidation takes place in
liquid phase, but the chain is initiated by active particles formed
in the gas and liquid phase (Potapkin et al., 1995).
It was found (Yermakov et al., 1995) that the extents of NOx and
SO2 depletion on simultaneous irradiation and spraying of water
(or water solution) are sensitive not only to absorbed dose but also
to the amount of water existing in the gas as liquid droplets.
The conversion of SO2 can be obtained by reactions in gas
phase (thermal and radio-induced) and by reactions in liquid
phase. Because of high value of solubility of SO2 in water (Henrys
constant1.23 Mnatm  1) SO2 is easily absorbed in water and
here the SO2 is oxidized by a chain reaction in the presence of
dissolved oxygen (Potapkin et al., 1995). The conversion of NO
takes place only by radio-induced reactions. In gas phase NO is
oxidized to NO2 by reactions of NO with O and O3. In the absence
of water droplets the removal efciency of NO dropped sharply
after NO conversion to NO2 (Potapkin et al., 1995). This is because

of the inuence of the reactions with N and NH2 (Matzing,

1989):
NO2 N-2NO

12

NO2 N-N2 O O

13

NO2 NH2 -N2 O H2 O

14

In the presence of ne water droplets the quenching of NO2 take

place by dissolution and reaction in droplets of water (the Henrys
constant of NO2 is 6 times higher than the value for NO:
1.23n10  2 Mnatm  1 respectively 0.195n10  2 Mnatm  1 (Squadrito
and Postlethwait, 2009).
The dissolved NO2 can be oxidized by OH radicals obtained by
water radiolysis or reduced by reaction with S(IV) compounds
(Potapkin et al., 1995):
NO2 OH-HNO3

15



No2 HSO
3 -NO2 HSO3

16

H2 O


2

No
2 HSO3 H - SO4 NH4

17



NO
2 HSO3 --HSO4 N2

18

However, the experiments of Yermakov were done at quite

low value of L (about 10  4), at such values the inuence of FWD
on density of the gas-liquid system is quite low and is not
possible to use the accelerators with mid-energy (about 2 MeV)
due to high penetration depth of electrons (see Fig. 5).
Our purpose is to treat ue gases at high values of L (about
10  3). At this level it is possible to achieve a signicant increase
in the density of the medium (45 times higher than the density
of ue gases without FWD) and also it is possible to use electron
beam with mid-energy in irradiation reactors with a small
diameter. Moreover, the radiation-induced reactions developed
in liquid phase will play a signicant role and will increase the
overall efciency of the process.

134

I. Calinescu et al. / Radiation Physics and Chemistry 85 (2013) 130138

2. Experimental
The experimental part took place in the irradiation laboratories of the National Institute for Lasers, Plasma and Radiation
Physics Bucharest using an ALIN-10 EB accelerator.
The ALIN-10 accelerator is a laboratory installation designed
for fundamental research. It is located in a horizontal position
inside the irradiation room and is equipped with a postacceleration beam focusing and bending to project accelerated

Fig. 7. The schematic drawing of ring-shaped EB collection monitor.

EB loading characteristic
EB power characteristic (100 Hz)
182
12.46 MeV
12

156

10

130

PEB max

104

9.17 MeV

78
6.23MeV

13 A

52

26.25 A

26

0
0

10
15
20
25
EB average current, IEB (A)

30

35

EB average power, PEB (W)

EB average energy, EEB (MeV)

14

Fig. 8. EB average energy (EEB) and electron average beam power (PEB) versus EB
average current (IEB).

EB either on the accelerating structure axis (through a horizontal

vacuum window exit port) or at right angles to the accelerator
structure (through a vertical vacuum window exit port) ALIN-10
is current used to develop new technologies in the eld of
environment pollution control, material processing, biomedicine
and health-care products synthesis studies. The ALIN-10 accelerator is of traveling-wave type, driven by 2 MW peak power
tunable EEV M5125 type magnetrons (English Electric Valve
Company Ltd.) operating in S-band (29923001 MHz). In the
ALIN-10 output arrangement with horizontal and vertical exit
ports, the electron beam passes a ring-shaped EB collection
monitor (RS-EBCM) shown in Fig. 7.
The RS-EBCM collection monitor that intercept only a fraction
of exit accelerator EB give, together with its associated electronics
instrumentation, a relative value of the EB average current
(EBAC). The EBAC relative level has been calibrated with a Faraday
cup placed at the EB exit end of the ALIN-10. Fig. 8 shows the
ALIN-10 beam loading characteristic and beam power characteristic. As seen in Fig. 8, the optimum values of the EB average
current IEB and EB average energy EEB to produce maximum
average output power PEB for a xed pulse duration tEB and
repetition frequency fEB are as follows: EEB 6.23 MeV;
IEB 26.25 mA; PEB 164 W (fEB 100 Hz, tEB 3.5 ms).
The ue gases are obtained in a burner and had a composition
similar to gases from power plants that burn coal or oil with high
sulfur content. The sulfur and nitrogen content from the dieselfuel used was assured by adding carbon disulde and aniline,
respectively (Calinescu et al., 2012).
The laboratory installation for ue gases treatment with EB,
generated by ALIN-10 accelerator, consists of the following
technological units: (i) the fuel burning unit; (ii) the ue gas
conditioning system; (iii) the reactor for ue gas irradiation with
EB; (iv) the liquid product separation unit; (v) the continuous
gases monitoring system; and (vi) the treated gas elimination unit
and is described in our previous paper (Calinescu et al., 2012).
In Fig. 9 is presented the EB reactor with devices for FWD
formation. The EB reactor is designed like a Faraday cage in order
to permit the EB average current monitoring during gaseous
mixture irradiation. It consists of two concentric stainless steel
cylindrical vessels: a gas and electrically insulated inner vessel of
0.2 m inner diameter and 2.967 m long and a grounded outer
vessel of 0.28 m inner diameter and 3.145 m long.
On the inner cylinder were xed 15 nozzles of 0.2 mm by
which is introduced water as fog curtains with an Aero Mist pump
of 0.5 gallons per minute (GPM) and 700 W, that have working
pressure of 70 bar. The pump contains a 5 mm particulate lter.
Water is pushed through a pressure-resistant nylon hose in nozzles

Fig. 9. The schematic drawing of EB reactor for treatment of ue gases by EB in the presence of FWD.

I. Calinescu et al. / Radiation Physics and Chemistry 85 (2013) 130138

135

Table 1
Treatment conditions and results obtained for ue gas treatment by EB, in the presence of ammonia and of FWD.
No.

Treatment conditions
Flow rate (l/h)

1
2
3
4
5
6
7
8
9
a

Gases

Liq. water

1000
5200
5200
14,000
14,000
14,000
14,000
14,000
14,000

0
0
50
0
50
0
50
0
50

Results
NH3 stoecha

0.8
0.9
0.9
0.6
0.6
0.95
0.95
0.95
0.95

Initial conc. (ppmv)

SO2

NO

2000
812
812
1116
1116
423
423
520
520

400
44
44
102
102
135
135
292
292

Treatment type

RE SO2

RE NOx

EB
EB
EB FWD
EB
EB FWD
EB
EB FWD
EB
EB FWD

0.98
0.85
0.94
0.70
0.88
0.97
0.98
0.97
0.98

0.62
0.78
0.98
0.36
0.88
0.32
0.84
0.20
0.8

Ighigeanu et al. (2012)

NH3 stoech. [NH3]/(2  [SO2] [NOx]), with all concentrations in ppmv.

that are mechanically tted with screw clamps and helically and
equidistant distributed to 1201. The high pressure causes water to be
sprayed through nozzles as small droplets of the diameter close to
10 mm.
The electrons are introduced in the EB reactor through two exit
windows, one of each in the form of cylinder, made of 100 mmthick aluminum foils. The EB current is collected on the electrical
insulated inner vessel of the EB reactor, integrated and displayed
on the control desk.
The characteristics of ALIN-10 EB accelerator used in our
experiments (Ibeam current, Eelectrons energy and PEB
power) highlight on Fig. 8, are
Iexit accelerator 13 71 mA
Eexit accelerator 9.17 0.23/ 0.1 MeV
ireactor as Faraday cup (without FWD): 1011 mA
Ireactor as Faraday cup (with FWD): 0.51 mA
Pexit accelerator 13 mA  9.17 MeV119.21 W
Preactor inside power (without FWD) 10 (11) mA  9.17 MeV 91.7
(100.87) W (about 27.317.34 W are losses in the two
windows of the reactor, the two air gaps, and other uncontrolled losses).

3. Results and discussion

In our experiments, ne water droplets, that are produced
with a high pressure pump, are injected simultaneously with a
high energy (69 MeV) EB into a proper modied EB reactor
reported in our previous paper (Ighigeanu et al., 2012) for the
following purposes:

 To promote the use of medium to high-energy EB accelerators

Obs.

for air pollutants removal in order to be in accordance with

Cleland recommendations: the electron energy losses in the
dual beam windows and the air space will be substantially
lower with higher electron energies (Cleland, 2007).
To reduce the length of useful penetration of high energy
electrons that is in air much bigger than the length of the usual
EB reactors (EBRs) that are presently used for air pollutants
removal. Therefore, the presence of FWD simultaneously with
a high energy electron beam makes possible the use of EB
reactors with suitable size.

In view of above conditions, the injection of FWD in conjunction

with high energy EB in the gaseous mixture is able to shorten EB
ranges and to improve high energy EB absorption and consumption
in useful chemical reactions. A laboratory accelerator of high

energy (69 MeV) was used to reveal the effect of ne water

droplets. Research must be pursued using mid-energy accelerators
(12 MeV).
As evaluation criterion of the process were calculated pollutant (P) removal efciency (RE), reactor energy density (RIED), and
energy efciency (EE) by using the following expressions
(Zhu et al., 2009):
RE %

Pinlet Poutlet
 100%
Pinlet

19

EB_input_power W

 60  103
RIED kJ=L
gas_f low_rate L=min

20

Pinlet  RE
 103
EE g=kWh
RIED

21

where [P]inlet is in mg/N m3.

A number of experiments were conducted using ue gases
ow rates ranging between 5.2 and 14.0 m3/h. Two operation
modes were tested EB only or EB combined with FWD. The
treatment conditions and the results are shown in Table 1. For
comparison, the results obtained previously on the same installation but without FWD, are presented.
It can be noticed that the presence of FWD causes an increase
of the pollutants removal efciency. This observed effect is bigger
in the conditions when the values of the removal efciency in the
absence of FWD are lower, i.e. lower doses regions.
For different reaction conditions the density of reaction mixture will be changed and from this reason also penetration depth
and the EB power effectively used. Table 2 presents the treatment
conditions, the effectively used EB power (determined according
to the accelerator power, the losses in windows and the air gap,
and function of the ratio between the reactor length and the
penetration depth) and also the results obtained: reactor energy
density and energy efciency determined for effectively used EB
power and for total EB power.
From the values presented in Table 2 it can be noticed that the
presence of FWD leads to an increase in the density of the
medium and thus, the penetration depth of EB in the reaction
medium is signicantly changed. Because the reactor length is
limited to 300 cm, we can consider that the ratio between the
reactor length and the penetration depth provides information
about the effectively used power. From this value of power used
we can determine the real values of RIED and EE.
From the analysis of values listed in Table 2 we can notice that
in the presence of FWD the real EE values are lower than those
obtained in the absence of water droplets. However, this is

136

I. Calinescu et al. / Radiation Physics and Chemistry 85 (2013) 130138

Table 2
RIED and EE values for different treatment conditions.
No

Treatment conditions
Liquid water
fraction, L

1
2
3
4
5
6
7
8
9

0
0
0.0096
0
0.0036
0
0.0036
0
0.0036

Density (g/L)

1.06
1.06
10.64
1.06
4.6
1.06
4.6
1.06
4.6

Penetration
depth, (cm)

4530
4530
426
4530
1260
4530
1260
4530
1260

Absorbed EB
power (W)

9
9
82.6
9
25
9
25
9
25

Results for effectively used EB power

Results for total EB power (96.3 W)

RIED (kJ/L)

RIED (kJ/L)

0.032
0.006
0.057
0.002
0.006
0.002
0.006
0.002
0.006

EE (g/kWh)
SO2

NOx

172.8
316.5
38.1
964.4
436.5
506.6
184.2
622.7
226.5

10.3
7.4
1.0
21.3
18.7
25.0
23.6
33.8
48.7

0.347
0.067
0.067
0.025
0.025
0.025
0.025
0.025
0.025

EE (g/kWh)
SO2

NOx

16.2
29.6
32.7
90.1
113.3
47.3
47.8
58.2
58.8

1.0
0.7
0.9
2.0
4.9
2.3
6.1
3.2
12.6

Table 3
Values for industrial EBFGT plant of Pomorzany (Chmielewski et al., 2004).
Results for total EB power

Flow rate for one EB accelerator, (N m3/min)

EB power (kW)
RE (%)
Pollutant inlet (mg/N m3)
RIED (kJ/L)
EE (g/kWh)

Results for effectively used EB power (  21.8% losses)

SO2

NOx

SO2

NOx

1,125,000
259,700
0.9
1500
0.014
97.47

1,125,000
259,700
0.7
391
0.014
19.76

1,125,000
203,085
0.9
1500
0.011
123.31

1,125,000
203,085
0.7
391
0.011
25.00

Table 4
Values of beam power losses function of EB energy obtained at pilot plant operated at EPS Kaweczyn (for 0.5, 0.6 and 0.7 MeV) and calculated for different energies of
electron beam in the presence of FWD (for 12 MeV).
EB energy
(MeV)

Beam power losses

(windows air)
(kW or %)

Losses due to vessel diameter Volume ratio

(1.6 m)
liquid:gas, L

Power used to obtain Proper diameter

FWD (P) (kW)
to avoid losses, m

Without FWD With FWD

(L1) (kW or %) (L2) (kW or %)

0.5
0.6
0.7
0.7

47.4
29.2
21.8
21.8

0.4
2.9
14.8
14.8

14.3

46

1.5

8.4

74

6.2

82

5.2
39
19
4
56
39
25
10
1
74
65
53
39
19

Energy economy
(kW or %)
EnEc1

0.0001
0.0001
0.0005
0.001
0.0005
0.001
0.0015
0.002
0.003
0.0005
0.001
0.0015
0.002
0.003

compensated by the low level of EB power used in the absence

of FWD.
From the analysis of the energy efciency values function of
the EB power provided by the accelerator the positive effect of
FWD presence in the reaction mixture can be noticed. This effect
is more important for NOx. In order to compare our experimental
data with those obtained in an industrial installation, we made
some calculations for the operational conditions of EBFGT plant in
EPS Pomorzany, Poland (Chmielewski et al., 2004). The values are
presented in Table 3.

3.9
3.89
19.45
38.90
19.45
38.90
58.35
77.80
116.70
19.45
38.90
58.35
77.80
116.70

With out
FWD

With FWD

(L1  (L2 P))

2.1
3.81
3.81
3.81
6.71
6.71
6.71
6.71
6.71
9.56
9.56
9.56
9.56
9.56

1.8
3.47
2.54
1.91
4.47
3.35
2.68
2.24
1.68
6.37
4.97
3.98
3.32
2.49

5.70
3.00
8.04
3.08
 1.83
 4.21
 9.09
 13.71
 43.56
 11.52
 21.83
 29.90
 35.21
 53.77

EnEc2

10.5
15.54
10.58
11.57
9.19
4.31
 0.31
 30.16
4.08
 6.23
 14.3
 19.61
 38.17

Our experiments 8 and 9 are carried out with pollutants

concentrations similar with those existing in the Pomorzany power
plant. It can be noticed that the values obtained for the energy
efciency (EE) are higher when FWD are used as compared to those
obtained in the industrial plant (for effectively used EB power).
It is also important to determine the power of the pump used
to obtain FWD. The output power of this pump can be determined
by the equation (Perry, 1999):
P H*Q =3:599*106

22

I. Calinescu et al. / Radiation Physics and Chemistry 85 (2013) 130138

where P is the output power of the pump (kW), H is the total

dynamic head (Pa) and Q is the capacity (m3/h).
Table 4 contains some data reported from the EBFGT pilot
plant experiments at EPS Kaweczyn, Poland (20,000 N m3/h,
100 kW beam power and initial electrons energy of 0.7 MeV)
(Chmielewski et al., 1995a) and calculated values of EB losses for
different energies of electron beam in the presence of FWD (for
1 to 2 MeV).
It can be noticed how the losses in windows and air gap are
decreasing while the electrons energy raises but the reactor
diameter has to be very large in order to trap the full energy of
electrons. The losses due to the reactor diameter are lower in the
presence of FWD. In determining the power necessary to obtain
FWD it can be seen that there is a limitation in the amount of
liquid water used.
Thus using EB of 1 MeV in a reactor with 1.6 m diameter in the
absence of FWD the power losses are 46% which means 46 kW for
the pilot plant that has 2 EB accelerators of 50 kW. In the presence
of FWD the losses are reduced at 39%, 19% or 4% (function of
Liquid:gas ratio used) for obtaining FWD a power of 3.938.9 kW
is used. The energy economy due to the use of FWD (EnEc1) can
be determined as the difference between the losses due to vessel
diameter in the absence of FWD and the sum of losses due to
vessel diameter in the presence of FWD and the power used to
obtain FWD.
The energy economy due to the use of mid-energy electron
beams and the use of FWD (EnEc2) also considers the reduction of
beam power losses (windowsair) as compared to the situation
in which EBs of 0.7 MeV are used.
We should remember, however, that this application leads to
application of so called semi-wet process and the byproduct is in
the form of solution. The mist can be eliminated from the outlet
ue gases through condensation ESP application (Porowski et al.,
1995). To obtain solid state fertilizer (crystals) the water has to be
evaporated. This can be done in the humidication tower using
the heat of the inlet ue gases.

4. Conclusions
Electron beam accelerators most often applied for ue gas
treatment are those with medium electrons energy ( o1 MeV) in
order to obtain low penetration depths (  3 m). At this energy
level of electrons a signicant level of electron beam power is lost
in the metallic foil windows and in the air gap between them. If
higher electron beam accelerators can be used (minimum
1.5 MeV), these losses would be signicantly reduced. At this
level of energy, however, the penetration depth increases signicantly (approx. up to 7.5 m), which means that the treatment
of gases becomes due to the process vessel design difculties.
However if FWD are added to the reaction mixture, the density of
this mixture increases largely, which determines the reduction of
the penetration depth to reasonable values (23 m).
The overall efciency of the process can be measured using the
energy efciency (g pollutant/kWh of effectively used EB). The use
of this parameter allows us to compare the classical experiments
(in which only EB treatment was used) with the experiment in
which EB treatment in the presence of FWD was used at volumetric ratios liquid/gas between 0.0035 and 0.0096.
The presence of FWD determines the increase of the EB power
used, and the energy efciency values obtained are higher than
the ones obtained in an industrial plant in which only EB
treatment is used.
When the power necessary to pump water in order to obtain
FWD is included in the total energy balance it can be noticed that
there is a limit value of the volumetric ratio liquid:gas (L). For a

137

higher amount of water it can be noticed that the energy needed

for obtaining FWD is higher than the economy of energy obtained
by using higher energy electrons. The optimal value of L is
function of the EB energy and of the maximum diameter of the
irradiation reactor.
The proposed solution requires some changes in the process
engineering since the mist has to be removed from the outlet
gases and then water evaporated to obtain solid product which is
fertilizer component.

References
Calinescu, I., Martin, D., Ighigeanu, D., Bulearca, A., 2012. Flue gases treatment by
irradiation with electron beam in the presence of ne water droplets. Rev.
Chim. Bucharest. 63 (6), 576579.
Chmielewski, A.G., 2011. Electron accelerators for environmental protection. In:
Chao, A.W., Chou, W. (Eds.), Reviews of Accelerator Science and Technology,
vol. 4. World Scientic Publishing Company, pp. 147160.
Chmielewski, A.G., 2007. Industrial applications of electron beam ue gas
treatmentfrom laboratory to the practice. Radiat. Phys. Chem. 76,
14801484.
Chmielewski, A.G., Licki, J., Pawelec, A., Tyminski, B., Zimek, Z., 2004. Operational
experience of the industrial plant for electron beam ue gas treatment. Radiat.
Phys. Chem. 71, 439442.
Chmielewski, A.G., Tyminski, B., Licki, J., Iller, E., Zimek, Z., Radzio, B., 1995a. Pilot
plant for ue gas treatmentcontinuous operation tests. Radiat. Phys. Chem.
46, 10671070.
Chmielewski, A.G., Zimek, Z., Panta, P., Drabnik, W., 1995b. The double window for
electron beam injection into the ue gas process vessel. Radiat. Phys. Chem.
45 (6), 10291033.
Cleland, M.R., 2007. Technical aspects of ue gas irradiation. Presented at the IAEA
Technical Meeting on Prospects and Challenges in Application of Radiation for
Treating Exhaust Gases, Warsaw, Poland.
Cleland M.R., 2005. Industrial applications of electron accelerators. Presented at
the CERN Accelerator School Small Accelerator Course Zeegse, Netherlands.
Cleland, M.R., 2004. Comparisons of Monte Carlo and ICRU electron energy vs.
range equations. Radiat. Phys. Chem. 71 (12), 585589.
Cooper, W.J., Curry, R.D., OShea, K.E., 1998. Environmental Applications of Ionizing
Radiation, 1st ed. John Wiley &Sons, New York.
Drobny, J.G., 2010. Radiation Technology for Polymers, second ed. CRC Press.
Gaffney, J.S., Streit, G.E., Spall, W.D., Hall, J.H., 1987. Beyond acid rain. Do soluble
oxidations and organic toxins interact with SO2 and NOX to increase ecosystem
effects? Environ. Sci. Technol. 21, 519524.
Genuario, R.D., 2009. Wet discharge electron beam ue gas scrubbing treatment.
US Patent 2009/0188782.
Hackman, R., Akiyama, H., 2000. Air pollution control by electrical discharges. IEEE
Trans. Dielectr. Electr. Insul. 7 (5), 654682.
Hatano, Y., Katsumura, Y., Mozumder, A., 2011. Charged Particle and Photon
Interactions with Matter. CRC Press.
IAEA (International Atomic Energy Agency), 2010. Industrial Electron Beam
Processing. Working materialRevision 4.
Ighigeanu, D., Martin, D., Calinescu, I., Bulearca, A., Manaila, E., Craciun, C., 2012.
Gaseous pollutants removal by electron beam based hybrid systems. Rev.
Chim. Bucharest 63 (2), 183188.
Kato, N., Akimoto, H., 2007. Anthropogenic emissions of SO2 and NOX in Asia:
emission inventory. Atmos. Environ. 41, S171S191.
Licki, J., Chmielewski, A.G., Iller, E., Zimek, Z., Mazurek, J., Sobolewski, L., 2003.
Electron-beam ue-gas treatment for multicomponent air-pollution control.
Appl. Energy 75, 145154.

Matzing,
H., Namba, H., Tokunaga, O., 1993. Kinetics of SO2 removal from ue gas
by electron beam technique. Radiat. Phys. Chem. 42 (46), 673677.

Matzing,
H., Paur, H.R., 1992. Chemical mechanisms and process parameters of
ue gas cleaning by electron beam. In: Nriagu, J.O. (Ed.), Gaseous Pollutants:
Characterization and Cycling. Wiley, New York, pp. 307331.

Matzing,
H., 1989. Chemical Kinetics of Flue Gas Cleaning by Electron Beam.
Keruforschunfzektrum Karslruhe. KFK494.
Namba, H., Hashimoto, S., Tokunaga, O., Suzuki, R., 1998. Electron beam treatment
of lignite-burning ue gas with high concentrations of sulfur dioxide and
water. Radiat. Phys. 53, 673681Chem 53, 673681.
Perry, H.R., 1999. Perrys Chemical Engineeres Handbook, 7th edition McGraw-Hill.
Person, J.C., Ham, D.O., 1988. Removal of SO2 and NOx from stack gases by electron
beam irradiation. Radiat. Phys. Chem. 31 (13), 18.
Porowski L., Chmielewski A.G., Iller E., Luty, L., 1995. Method for removing acid gas
impurities from industrial gases stream and equipment for removal of acid gas
pollutants from industrial gases waste stream. Polish Patent 900427.
Potapkin, B.V., Deminski, A., Fridman, A.A., Rusanov, D., 1995. The effect of clusters
and heterogeneous reactions on non-equilibrium plasma ue gas cleaning.
Radiat. Phys. Chem. 45 (6), 10811088.
Ramanathan, V., Feng, Y., 2009. Air pollution, greenhouse gases and climate
change: global and regional perspectives. Atmos. Environ. 43, 3750.

138

I. Calinescu et al. / Radiation Physics and Chemistry 85 (2013) 130138

Squadrito, G., Postlethwait, E., 2009. On the hydrophobicity of nitrogen dioxide:

could there be a lens effect for NO2 reaction kinetics? Nitric Oxide 21 (2),
104109.
Streets, D.G., Waldhoff, S.T., 2000. Present and future emissions of air pollutions in
China: SO2, NOX, and CO. Atmos. Environ. 34, 363374.
Tokunaga, O., Suzuki, N., 1988. Radiation chemical reactions in NOX and SO2
removals from ue gas. Radiat. Phys. Chem. 24 (1), 145165.

Yermakov, A.N., Zhitomirsky, B.M., Sozurakov, D.M., Poskrebyshev, G.A., 1995.

Water aerosols spraying for SO2 and NOX removal from gases under e-beam
irradiation. Radiat. Phys. Chem. 45 (6), 10711076.
Zhu, T., Li, J., Liang, W., Jin, Y., 2009. Synergistic effect of catalyst for oxidation
removal of toluene. J. Hazard. Mater. 165 (13), 1601258.
Zimek, Z., 1995. High power electron accelerators for ue gas treatment. Radiat.
Phys. Chem. 45 (6), 10131015.