1-s2.0-s0969806x12004434-main
TRANSCRIPT
-
8/10/2019 1-s2.0-S0969806X12004434-main
1/9
E-Beam SO2 and NOx removal from flue gases in the presence
of fine 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, Romaniab National Institute for Lasers, Plasma and Radiation Physics, Electron Accelerators Laboratory, P.O.Box: MG-36, ]409 Atomistilor St., 077125 Magurele, Ilfov, Romaniac Institute of Nuclear Chemistry and Technology, ]16 Dorodna St., 03-195 Warsaw, Poland
H I G H L I G H T S
c The medium-energy EB accelerators are proposed for flue gases treatment.
cThe energy losses in the windows and in the air gap between them are reduced.
cTo increase the density of the reaction medium and to reduce the penetration depth of EB fine water droplets (FWD) are used.
cDetermining the energy efficiency the favorable effect of the method was demonstrated.
cThe 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
Article history:
Received 30 May 2012
Accepted 11 October 2012Available online 20 November 2012
Keywords:
Flue gasesNOx and SO2 removal
Electron beam
Fine water droplets
a b s t r a c t
The Electron Beam Flue Gas Treatment (EBFGT) has been proposed as an efficient method for removal of
SO2and 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 off-
gases purification. 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 theirpenetration depth through gas and this can be achieved by increasing the density of the reaction
medium by means of dispersing a sufficient amount of fine 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 flue gases with a high
concentration of SO2and NOx using high-energy EBs (9 MeV). By determining the energy efficiency of
the process the favorable effect of using FWD and high-energy EB was demonstrated.
& 2012 Elsevier Ltd. All rights reserved.
1. Introduction
New technologies and processes have promoted our economic
growth and have been remarkably changing our lifestyles. How-ever, 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 remark-
ably 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
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 theconsumption of environment-friendly energies instead of fossil
fuels, the steep demand for energy due to the recent economic
growth makes it difficult 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 sufficient
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
Contents lists available at SciVerse ScienceDirect
journal homepage: w ww.elsevier.com/locate/radphyschem
Radiation Physics and Chemistry
0969-806X/$- see front matter& 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.radphyschem.2012.10.008
n Corresponding author. Tel.: 40 214574346; fax: 40 214574243.
E-mail addresses: [email protected] (I. Calinescu),
[email protected] (A. Chmielewski),
[email protected] (D. Ighigeanu).
Radiation Physics and Chemistry 85 (2013) 130138
http://www.elsevier.com/locate/radphyschemhttp://www.elsevier.com/locate/radphyschemhttp://localhost/var/www/apps/conversion/tmp/scratch_1/dx.doi.org/10.1016/j.radphyschem.2012.10.008mailto:[email protected]:[email protected]:[email protected]://localhost/var/www/apps/conversion/tmp/scratch_1/dx.doi.org/10.1016/j.radphyschem.2012.10.008http://localhost/var/www/apps/conversion/tmp/scratch_1/dx.doi.org/10.1016/j.radphyschem.2012.10.008mailto:[email protected]:[email protected]:[email protected]://localhost/var/www/apps/conversion/tmp/scratch_1/dx.doi.org/10.1016/j.radphyschem.2012.10.008http://localhost/var/www/apps/conversion/tmp/scratch_1/dx.doi.org/10.1016/j.radphyschem.2012.10.008http://localhost/var/www/apps/conversion/tmp/scratch_1/dx.doi.org/10.1016/j.radphyschem.2012.10.008http://www.elsevier.com/locate/radphyschemhttp://www.elsevier.com/locate/radphyschem -
8/10/2019 1-s2.0-S0969806X12004434-main
2/9
have been applied to treat SO2 and NOx. However, the wet lime
scrubber method requires wastewater treatment, and the cata-
lysts 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 flue gas
purification (EBFGT) has been proposed as an efficient method
because it has the following advantages (Hatano et al., 2011):
has simultaneous denitrification and desulfurization possibilities, no wastewater treatment is required, no expensive catalyst is required, provides simple process and its operation, produces profitable products.
The application of electron beams to treat flue 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 flue
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 andAkiyama, 2000;Licki et al., 2003) and the difficulties in operation
of very high power accelerators. That is why it is mandatory to
find new solutions and to develop strategies for diminishing
radiation dose absorbed in the flue gas and to optimize the
process for better uses of EB energy.
Subjected to the electron beam irradiation process, the main
components of flue 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 , N2
n, O2n, 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 reactsvery 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 HO2in 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,H2O vapors_at_high_concn e-HO,HO2
,On,ions,excited-species
1
In the presence of these reactive species, NOx and SO2 from
flue gases are oxidized and produce nitric acid and sulfuric acid,
respectively, as intermediate products. These acids are neutra-
lized 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 thermaland 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 NOxremoval is the dose.
The presence of water vapors is mandatory for the efficient
removal of SO2and NOxfrom gases. By radiolysis of water the HO
radical is obtained and this radical will oxidize both SO2and NOx(Genuario, 2009):
NOHO
M-HNO2M (3)
NO2HOM-HNO3M (4)
SO2HOM-HSO3M (5)
HSO3O2-SO3HO2 (6)
SO3H2O-H2SO4 (7)
NOHO2-NO2HO
(8)
A gas humidity 810 vol% is necessary to obtain the optimal
removal-efficiencies 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 flue gas treatment
have sufficient 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 ioniz-
ing 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 tofast 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, radi-
cals, 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 flow rate of the treated material and the absorbed dose
needed for the process:
Dave FpP= M=T
9
Fig. 1. The main reactions used in EBFGT to convert SO2and NOX(Namba et al., 1998).
I. Calinescu et al. / Radiation Physics and Chemistry 85 (2013) 130138 131
-
8/10/2019 1-s2.0-S0969806X12004434-main
3/9
whereD(ave)is the average absorbed dose in kGy, Pis the emitted
radiation power in kW, Tis the treatment time in s and Mis the
mass of irradiated material in kg.
The factorF(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 shouldbe considered in the case of EBFGT. This quantity is difficult 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 accel-
erators that are presently used for air pollutants removal
(Chmielewski et al., 1995b). The total EB energy losses (back-
scattering, beam windows and air gap) are substantially lower
with higher EB incident energy (seeFig. 3). The useful EB energy
from input energy is around 50% for 0.5 MeV and about 95% above
3 MeV. The lower electrical efficiencies of accelerators with
higher energies are partially compensated by the lower electron
energy losses in the beam windows. In addition, accelerators with
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 (elec-trons 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. InFig. 5 are presented the calculated values of
penetration depth (on the basis of practical range R(p) for flue
gases and for flue gases with fine water droplets (FWD), the
volume fraction of water is designated as L (Vaq/Vgas), whereVaqand V
gas refer to the irradiated volumes for the two phases,
L4 103.
The range parameters can be correlated with the incident
electron energy E with sufficient accuracy for industrial applica-
tions by using the following linear equations (for polyethylene)
(Cleland, 2005):
Rp 0:510E0:145 10
Electron ranges in other materials can be estimated by multi-
plying the polyethylene range with the ratio of their CSDA ranges.
Rmaterial Rpolyethylene CSDAm=CSDApe 11
CSDAranges 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
r1.06 103 g/cm3) are treated by an electron accelerators
equipped with a Ti foil (50 mm, r4.5 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 (seeFig. 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. 2. The process of generation of reactive species by high energy electrons.
Fig. 3. Electron energy losses function of incident electron energy for flue gases
treatment by EB (Cleland, 2007).
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) 130138132
http://www.nist.gov/pml/data/star/index.cfmhttp://www.nist.gov/pml/data/star/index.cfmhttp://www.nist.gov/pml/data/star/index.cfmhttp://www.nist.gov/pml/data/star/index.cfm -
8/10/2019 1-s2.0-S0969806X12004434-main
4/9
Although the medium-energy EB (2.5 MeV) are used with high
efficiency (with lower total energy losses, seeFig. 3), the penetra-
tion 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 flue gases. The influence of volume fraction of water (L) in
flue gases on the density of the gas medium and on the penetra-
tion depth of EB with 2.5 MeV is presented inFig. 6.
Several reasons showing the potential importance of droplet-phase oxidation in the EBFGT process can be outlines (Cooper
et al., 1998):
High driving force for the most oxidation process due to lowGibbs free energy of the products obtained in liquid phase;
High solubility of the mixture components (SO2 and theintermediates NO2, N2O4) resulting in an increase in its
liquid-phase concentrations;
Rapid rates of the aqueous-phase oxidation process initiatedby ionizing radiation.
Regarding the primary partitioning of the dose rates between
the irradiated phases, denotedDRgasand DRaq, one notes (Cooper
et al., 1998) that its magnitude is governed mainly by the ratio of
the phases densities, rgasand raq.Since the ratio raq/rgasis typically
Z103, then DRaq/DRgasraq/rgasZ103, 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 duringirradiation, one gets (Cooper et al., 1998):Daq/DgasZraq*L/rgas.
WhenL is quite low (in a typical EBFGT process Lr106), it is
unlikely that the radiation-induced reactions in the liquid phase
can a play a significant role in the removal of pollutants from the
irradiated gas, but ifL 103 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 NOxand
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 SO2in water (Henrys
constant1.23 Mnatm1) 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 NO2by reactions of NO with O and O3. In the absence
of water droplets the removal efficiency of NO dropped sharply
after NO conversion to NO2(Potapkin et al., 1995). This is because
of the influence of the reactions with N and NH2(Matzing, 1989):
NO2 N-2NO 12
NO2 N-N2OO 13
NO2 NH2-N2O H2O 14
In the presence of fine water droplets the quenching of NO2take
place by dissolution and reaction in droplets of water (the Henrys
constant of NO2 is 6 times higher than the value for NO:
1.23n102 Mnatm1 respectively 0.195n102 Mnatm1 (Squadrito
and Postlethwait, 2009).
The dissolved NO2can 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 HSO3-NO
2 HSO
3 16
No2 HSO3 H
-
H2OSO24 NH
4 17
NO2 HSO3--HSO
4 N2 18
However, the experiments of Yermakov were done at quite
low value ofL (about 104), at such values the influence 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 (seeFig. 5).
Our purpose is to treat flue gases at high values of L (about
103). At this level it is possible to achieve a significant increase
in the density of the medium (45 times higher than the density
of flue 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 significant role and will increase the
overall efficiency of the process.
Fig. 5. The penetration depth of electron beam function of electron energy in flue
gases (r1 g/L) and in flue gases with FWD (L 4n103, r5 g/L).
Fig. 6. The influence of liquid water fraction (L) in flue gases on the density of flue
gases with FWD and penetration depth for EB of 2.5 MeV.
I. Calinescu et al. / Radiation Physics and Chemistry 85 (2013) 130138 133
-
8/10/2019 1-s2.0-S0969806X12004434-main
5/9
-
8/10/2019 1-s2.0-S0969806X12004434-main
6/9
that are mechanically fitted 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 mm-
thick 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 onFig. 8, are
Iexit accelerator1371 mA
Eexit accelerator9.170.23/0.1 MeV
ireactor as Faraday cup (without FWD): 1011 mA
Ireactor as Faraday cup (with FWD): 0.51 mA
Pexit accelerator13 mA 9.17 MeV119.21 W
Preactor inside power(without FWD) 10 (11) mA 9.17 MeV91.7
(100.87) W (about 27.317.34 W are losses in the two
windows of the reactor, the two air gaps, and other uncon-
trolled losses).
3. Results and discussion
In our experiments, fine water droplets, that are produced
with a high pressure pump, are injected simultaneously with a
high energy (69 MeV) EB into a proper modified 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
for air pollutants removal in order to be in accordance withCleland 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 energyelectrons 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 fine water
droplets. Research must be pursued using mid-energy accelerators
(12 MeV).
As evaluation criterion of the process were calculated pollu-
tant (P) removal efficiency (RE), reactor energy density (RIED), and
energy efficiency (EE) by using the following expressions
(Zhu et al., 2009):
RE % PinletPoutlet
Pinlet 100% 19
RIED kJ=L
EB_input_powerW
gas_f low_rate L=min 60 103 20
EE g=kWh
Pinlet RE
RIED 103 21
where [P]inletis in mg/N m3.
A number of experiments were conducted using flue gases
flow 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 installa-
tion but without FWD, are presented.
It can be noticed that the presence of FWD causes an increase
of the pollutants removal efficiency. This observed effect is bigger
in the conditions when the values of the removal efficiency in the
absence of FWD are lower, i.e. lower doses regions.
For different reaction conditions the density of reaction mix-
ture will be changed and from this reason also penetration depth
and the EB power effectively used.Table 2presents 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 thepenetration depth) and also the results obtained: reactor energy
density and energy efficiency determined for effectively used EB
power and for total EB power.
From the values presented inTable 2it 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 significantly 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 inTable 2we 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
Table 1
Treatment conditions and results obtained for flue gas treatment by EB, in the presence of ammonia and of FWD.
No. Treatment conditions Results Obs.
Flow rate (l/h) NH3 stoecha Initial conc. (ppmv) Treatment type RE SO2 RE NOx
Gases Liq. water SO2 NO
1 1000 0 0.8 2000 400 EB 0.98 0.62 Ighigeanu et al. (2012)
2 5200 0 0.9 812 44 EB 0.85 0.783 5200 50 0.9 812 44 EB FWD 0.94 0.98
4 14,000 0 0.6 1116 102 EB 0.70 0.36
5 14,000 50 0.6 1116 102 EB FWD 0.88 0.88
6 14,000 0 0.95 423 135 EB 0.97 0.32
7 14,000 50 0.95 423 135 EB FWD 0.98 0.84
8 14,000 0 0.95 520 292 EB 0.97 0.20
9 14,000 50 0.95 520 292 EB FWD 0.98 0.8
a NH3 stoech. [NH3]/(2 [SO2] [NOx]), with all concentrations in ppmv.
I. Calinescu et al. / Radiation Physics and Chemistry 85 (2013) 130138 135
-
8/10/2019 1-s2.0-S0969806X12004434-main
7/9
compensated by the low level of EB power used in the absence
of FWD.
From the analysis of the energy efficiency 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 inTable 3.
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
efficiency (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*10
6
22
Table 2
RIED and EE values for different treatment conditions.
No T reatment condi ti ons Results for effectively used EB power Results for total EB power (96.3 W)
Liquid water
fraction,L
Density (g/L) Penetration
depth, (cm)
Absorbed EB
power (W)
RIED (kJ/L) EE (g/kWh) RIED (kJ/L) EE (g/kWh)
SO2 NOx SO2 NOx
1 0 1.06 4530 9 0.032 172.8 10.3 0.347 16.2 1.0
2 0 1.06 4530 9 0.006 316.5 7.4 0.067 29.6 0.73 0.0096 10.64 426 82.6 0.057 38.1 1.0 0.067 32.7 0.9
4 0 1.06 4530 9 0.002 964.4 21.3 0.025 90.1 2.0
5 0.0036 4.6 1260 25 0.006 436.5 18.7 0.025 113.3 4.9
6 0 1.06 4530 9 0.002 506.6 25.0 0.025 47.3 2.3
7 0.0036 4.6 1260 25 0.006 184.2 23.6 0.025 47.8 6.1
8 0 1.06 4530 9 0.002 622.7 33.8 0.025 58.2 3.2
9 0.0036 4.6 1260 25 0.006 226.5 48.7 0.025 58.8 12.6
Table 3
Values for industrial EBFGT plant of Pomorzany (Chmielewski et al., 2004).
Results for total EB power Results for effectively used EB power ( 21.8% losses)
SO2 NOx SO2 NOx
Flow rate for one EB accelerator, (N m3/min) 1,125,000 1,125,000 1,125,000 1,125,000
EB power (kW) 259,700 259,700 203,085 203,085
RE (%) 0.9 0.7 0.9 0.7
Pollutant inlet (mg/N m3) 1500 391 1500 391
RIED (kJ/L) 0.014 0.014 0.011 0.011
EE (g/kWh) 97.47 19.76 123.31 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
(windowsair)
(kW or %)
Losses due to vessel diameter
(1.6 m)
Volume ratio
liquid:gas,L
Power used to obtain
FWD(P) (kW)
Proper diameter
to avoid losses, m
Energy economy
(kW or %)
Without FWD
(L1) (kW or %)
With FWD
(L2) (kW or %)
EnEc1 EnEc2
With out
FWD
With FWD (L1 (L2P))
0.5 47.4 0.4
0.6 29.2 2.9
0.7 21.8 14.8
0.7 21.8 14.8 5.2 0.0001 3.9 2.1 1.8 5.70
1 14.3 46
39 0.0001 3.89 3.81 3.47 3.00 10.5
19 0.0005 19.45 3.81 2.54 8.04 15.54
4 0.001 38.90 3.81 1.91 3.08 10.58
1.5 8.4 74
56 0.0005 19.45 6.71 4.47 1.83 11.57
39 0.001 38.90 6.71 3.35 4.21 9.19
25 0.0015 58.35 6.71 2.68 9.09 4.31
10 0.002 77.80 6.71 2.24 13.71 0.31
1 0.003 116.70 6.71 1.68 43.56 30.16
2 6.2 82
74 0.0005 19.45 9.56 6.37 11.52 4.08
65 0.001 38.90 9.56 4.97 21.83 6.23
53 0.0015 58.35 9.56 3.98 29.90 14.3
39 0.002 77.80 9.56 3.32 35.21 19.61
19 0.003 116.70 9.56 2.49 53.77 38.17
I. Calinescu et al. / Radiation Physics and Chemistry 85 (2013) 130138136
-
8/10/2019 1-s2.0-S0969806X12004434-main
8/9
where P is the output power of the pump (kW), H is the total
dynamic head (Pa) and Qis 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 aredecreasing 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 (windows air) 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
flue 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 humidification tower using
the heat of the inlet flue gases.
4. Conclusions
Electron beam accelerators most often applied for flue 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 significant 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 significantly reduced. At this
level of energy, however, the penetration depth increases sig-
nificantly (approx. up to 7.5 m), which means that the treatment
of gases becomes due to the process vessel design difficulties.
However if FWD are added to the reaction mixture, the density of
this mixture increases largely, which determines the reduction ofthe penetration depth to reasonable values (23 m).
The overall efficiency of the process can be measured using the
energy efficiency (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 volu-
metric ratios liquid/gas between 0.0035 and 0.0096.
The presence of FWD determines the increase of the EB power
used, and the energy efficiency 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
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 byirradiation with electron beam in the presence of fine 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 Scientific Publishing Company, pp. 147160.
Chmielewski, A.G., 2007. Industrial applications of electron beam flue gastreatmentfrom laboratory to the practice. Radiat. Phys. Chem. 76,14801484.
Chmielewski, A.G., Licki, J., Pawelec, A., Tyminski, B., Zimek, Z., 2004. Operationalexperience of the industrial plant for electron beam flue gas treatment. Radiat.
Phys. Chem. 71, 439442.Chmielewski, A.G., Tyminski, B., Licki, J., Iller, E., Zimek, Z., Radzio, B., 1995a. Pilot
plant for flue gas treatmentcontinuous operation tests. Radiat. Phys. Chem.46, 10671070.
Chmielewski, A.G., Zimek, Z., Panta, P., Drabnik, W., 1995b. The double window forelectron beam injection into the flue gas process vessel. Radiat. Phys. Chem.45 (6), 10291033.
Cleland, M.R., 2007. Technical aspects of flue gas irradiation. Presented at the IAEATechnical Meeting on Prospects and Challenges in Application of Radiation forTreating Exhaust Gases, Warsaw, Poland.
Cleland M.R., 2005. Industrial applications of electron accelerators. Presented atthe 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 IonizingRadiation, 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 SO2and NOXto increase ecosystem
effects? Environ. Sci. Technol. 21, 519524.Genuario, R.D., 2009. Wet discharge electron beam flue 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 flue-gas treatment for multicomponent air-pollution control.Appl. Energy 75, 145154.
Matzing, H., Namba, H., Tokunaga, O., 1993. Kinetics of SO2removal from flue gas
by electron beam technique. Radiat. Phys. Chem. 42 (46), 673677.Matzing, H., Paur, H.R., 1992. Chemical mechanisms and process parameters offlue 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 treatmentof lignite-burning flue gas with high concentrations of sulfur dioxide andwater. 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 SO2and NOxfrom 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 gaspollutants from industrial gases waste stream. Polish Patent 900427.
Potapkin, B.V., Deminski, A., Fridman, A.A., Rusanov, D., 1995. The effect of clustersand heterogeneous reactions on non-equilibrium plasma flue 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.
I. Calinescu et al. / Radiation Physics and Chemistry 85 (2013) 130138 137
-
8/10/2019 1-s2.0-S0969806X12004434-main
9/9
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 inChina: SO2, NOX, and CO. Atmos. Environ. 34, 363374.
Tokunaga, O., Suzuki, N., 1988. Radiation chemical reactions in NOX and SO2removals from flue 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 NOXremoval from gases under e-beamirradiation. Radiat. Phys. Chem. 45 (6), 10711076.
Zhu, T., Li, J., Liang, W., Jin, Y., 2009. Synergistic effect of catalyst for oxidationremoval of toluene. J. Hazard. Mater. 165 (13), 1601258.
Zimek, Z., 1995. High power electron accelerators for flue gas treatment. Radiat.Phys. Chem. 45 (6), 10131015.
I. Calinescu et al. / Radiation Physics and Chemistry 85 (2013) 130138138