J O U RN A L OF E N V I RO N ME N TA L SC IE N CE S 4 1 (2 0 1 6) 5 1–5 8

Available online at www.sciencedirect.com

ScienceDirect
www.elsevier.com/locate/jes

Numerical evaluation of the effectiveness of NO2 and N2O5

generation during the NO ozonation process

Haiqiang Wang⁎, Zhuokai Zhuang, Chenglang Sun, Nan Zhao, Yue Liu, Zhongbiao Wu⁎
Department of Environment Engineering, Zhejiang University, Hangzhou 310058, China. E-mail: haiqiangwang@zju.edu.cn
Zhejiang Provincial Engineering Research Center of Industrial Boiler, Furnace Flue Gas Pollution Control, Hangzhou 310058, China

AR TIC LE I N FO ABS TR ACT

Article history: Wet scrubbing combined with ozone oxidation has become a promising technology for
Received 12 February 2015 simultaneous removal of SO2 and NOx in exhaust gas. In this paper, a new 20-species,
Revised 15 April 2015 76-step detailed kinetic mechanism was proposed between O3 and NOx. The concentration
Accepted 4 May 2015 of N2O5 was measured using an in-situ IR spectrometer. The numerical evaluation results
Available online 21 July 2015 kept good pace with both the public experiment results and our experiment results. Key
reaction parameters for the generation of NO2 and N2O5 during the NO ozonation process
Keywords: were investigated by a numerical simulation method. The effect of temperature on
Numerical simulation producing NO2 was found to be negligible. To produce NO2, the optimal residence time was
O3 1.25 sec and the molar ratio of O3/NO about 1. For the generation of N2O5, the residence time
NO should be about 8 sec while the temperature of the exhaust gas should be strictly controlled
N2O5 and the molar ratio of O3/NO about 1.75. This study provided detailed investigations on the
In-situ IR spectra reaction parameters of ozonation of NOx by a numerical simulation method, and the results
obtained should be helpful for the design and optimization of ozone oxidation combined
with the wet flue gas desulfurization methods (WFGD) method for the removal of NOx.
© 2015 The Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences.
Published by Elsevier B.V.

Introduction
The most abundant gaseous pollutants emitted from power
plants and industrial boilers are sulfur dioxide (SO2) and
nitrogen oxides (NOx), which make a dramatic contribution to
acid rain and smog formation (Price et al., 1997). For the recent
severe hazy weather that invaded most areas of China more
than once (Hou et al., 2011; Jiang et al., 2014; Liu et al., 2014),
China has established more rigorous legislation to control the
emission of SO2 and NOx from stationary sources (Huang et
al., 2014). Generally, SO2 is controlled effectively by wet flue
gas desulfurization methods (WFGD), and low NOx burners
(Nishimura et al., 1997), selective catalytic reduction (SCR)
(Topsøe, 1994) and selective non-catalytic reduction (SNCR)
(Tayyeb Javed et al., 2007) are typically used for the removal of
NOx. Since exhaust gas from power plants and industrial boilers
contains multiple pollutants, combined technologies such as
WFGD + SCR or WFGD + SNCR are commonly applied. How-
ever, these combined technologies require expensive invest-
ments and operating costs, resulting in an increasing amount of
attention paid to simultaneous removal technologies with high
efficiency, low investment and reasonable operating cost.
Up to now, numerous simultaneous removal technologies
have been reported, such as activated carbon adsorption tech-
nology (Sumathi et al., 2010), the dielectric barrier discharge (DBD)
plasma flue gas treatment (Obradović et al., 2011), and various
types of wet scrubbing (Hu et al., 2010; Liu and Zhang, 2011; Raju
et al., 2008; Wang et al., 2012; Zhao et al., 2011). For wet scrubbing

⁎ Corresponding authors. E-mail: zbwu@zju.edu.cn (Zhongbiao Wu).

http://dx.doi.org/10.1016/j.jes.2015.05.015
1001-0742/© 2015 The Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences. Published by Elsevier B.V.
52 J O U RN A L OF E N V I RO N ME N TA L SC IE N CE S 4 1 (2 0 1 6) 51–5 8
technology, ozone has been a typical oxidant used for the
oxidation of NO.
In 1997, Nelo et al. (1997) already began to study the use of
ozone for the oxidation of nitrogen oxides. Mok and Lee (2006)
reported a two-step process which combined rapid oxidation
by ozone and aqueous absorption by sodium sulfide. The
process achieved high removal efficiencies for both SO2 and
NOx. Wang et al. (2007) also studied a similar two-step process
that could remove NOx, SO2 and Hg effectively. As the higher
oxidation states of NO (N2O3, N2O5) are much more soluble in
water (500.0 g/dm3) than NO (0.032 g/dm3) and NO2 (213.0 g/dm3)
(Dora et al., 2009), the wet-scrubbing removal efficiency of
NOx relies highly on the resultant composition of NOx during
the NO oxidation process; whereas, the oxidation products of
NO ozonation vary as the reaction environment and running
parameters change. Compared with NO2, N2O5 is easier to remove
in scrubbing systems, but needs more O3 for its generation.
Hence, detailed investigations on the reaction environment
and running parameters for ozonation of NOx are required
for the design and optimization of this ozone oxidation combined
with WFGD method for the removal of NO.
For this complex reaction system, numerical simulation
is needed because it can provide more detailed information
about the reaction system. CHEMKIN is one of the most
widely used and validated kinetic software programs (Zajemska
et al., 2014). It is accurate for industrial burners, gas turbines,
chemical processing and so on (Abián et al., 2011; Gui et al., 2014;
Li et al., 2010; Yang et al., 2013). Therefore, CHEMKIN is suitable to
study the effectiveness of NO2 and N2O5 generation during the NO
ozonation process. Wang et al. (2006) studied an ozone-NOx
reaction co-flow jet by direct numerical simulation. However, in
that research, only the reactions between O3 and NOx in the
turbulence time scale were observed. Because of the low reaction
rate of O3–NO reactions, NO2 was the main product, with minor
amounts of NO3 and N2O5. Furthermore, the concentration of
N2O5 of the simulation was not verified by laboratory testing due
to the difficulty of measuring of N2O5.
Recently, Skalska et al. (2011) have investigated the kinetic
model of NOx ozonation and determined the rate constants of
NOx (NO2, NO3, and N2O5) based on Fourier transform infrared
(FT-IR) spectroscopy, although only at 25°C. Also, we studied
the O3 oxidation processes of NO and SO2 on a qualitative
scale, as well as their coexistence, with the help of an in-situ IR
spectrometer (Sun et al., 2014). These studies suggested that
in-situ FT-IR spectroscopy should be a suitable technology for
the measurement of N2O5. However, these reported studies
did not focus on the effect of running parameters, especially
the residence time, on the production of NO2 or N2O5 for
industrial application. Moreover, as far as we know, quanti-
tative studies on the production of N2O5 have rarely been
reported.
This article focused on simulating the reactions between
O3 and NO under different reaction conditions and running
parameters using a numerical simulation method. Mean-
while, the simulated concentration of N2O5 was verified by
experimental studies with the help of in-situ FT-IR spectros-
copy. More importantly, the suitable running parameters for
the generation of NO2 and N2O5 including reaction temper-
ature, residence time, and the molar ratio of O3/NO, were
studied.
1. Experimental set-up
Fig. 1 illustrates the schematic diagram of the experimental
system for NO oxidation by O3, consisting of a gas supply
system, gas-phase mixer, reactor and flue gas analyzing
system. The details of each apparatus and the experimental
gas conditions were described in our previous research (Sun
et al., 2014).
In Fig. 2, different IR spectra are shown. The bands at 743,
1246 and 1720 cm− 1 were assigned to N2O5 (Wängberg, 1993).
The concentration of N2O5 was calculated according to the
formula:
A1000
c¼ ð1Þ
bσN A
where c (mol/dm3) is the concentration, A is the absorbance,
NA (mol− 1) is the Avogadro constant, b (cm) is the path
length, and σ (cm2/mol) is the absorption cross section. The
absorption cross section could be calculated by the equation
given by Wängberg (1993).
2. Kinetic mechanism of the reactions between O3
and NOx
The reactions between O3 and NOx are complicated, which
include not only the decomposition of ozone and N2O5 but
also numerous intermediate transient reactions. Mok and
Lee (2006) came to the conclusion that 12 reactions were
involved in the mechanism. However, several important
species such as OH in the reactions between O3 and NOx were
not considered. Wang et al. (2006) provided a more detailed
kinetic mechanism. However, reactions related to NO3 were
missed. NO3 is unstable due to its strong oxidizing proper-
ties, but it is one of the higher oxidation states of NOx and
should not be neglected in the mechanism. The reactions
between NO3 and other species are R66–R76 shown in
Table 1, where a new 20-species, 76-step detailed kinetic
mechanism was proposed and listed (R1–R76). The rate
constants for the all the reactions listed in Table 1 were
acquired from the National Institute for Standards and
Technology (NIST) (Datebase).
3. Modeling
Computations of the NO ozonation process were all made
using the CHEMKIN 4.1 software package (Reaction Design
Co., Ltd, San Diego, USA) (Kee et al., 2006). The cylindrical
shear-flow reactor model was used in the computations. The
model took radial diffusion into account, neglecting axial
diffusion. Moreover, the model only included gas chemistry.
The reaction mechanism of the NO ozonation process has
been described above. NO concentration in the inlet ranged
from 200 to 400 ppm. NO2 concentration in the inlet was from
0 to 20 ppm. The ozone concentration in the inlet depended
on both NO concentration and O3/NO molar ratio. The other
component of the inlet atmosphere was N2.
 J O U RN A L OF E N V I RO N ME N TA L SC IE N CE S 4 1 (2 0 1 6) 5 1–5 8 53

Mixing tank

Exhaust
MFC MFC Reactor

NO N2 FTIR Ozone analyzer

Ozone
MFC Computer
generator
Infrared gas
analyzer

O2

Fig. 1 – Schematic diagram of the experimental apparatus for the oxidation of NOx. FTIR: Fourier transform infrared
spectroscopy; MFC: mass flow controller.

of O3/NO was so large, the subsequent transformation of NO2

4. Results and discussion
4.1. Verification of the model
In Fig. 3 the simulation results and Mok's experiment results
at 443 K and 2.94 sec, with 280 ppm NO and 20 ppm NO2, are
compared (Mok and Lee, 2006). As shown in Fig. 3, the
production of NO2 was determined by the total amount of
added ozone, and the calculation results fitted quite well
with those in Mok's experiment.
Both the results of the simulation and Mok's experiment
showed that NO mainly reacted to form NO2 with the added
O3 under this reaction condition. The mechanism indicated
that reactions like R1 (Table 1), R29, R31, R32, R33, R64 and R69
took part in the conversion of NO. However, reaction R1 is the
predominant step because of the high conversion rate of R1.
Fig. 4 shows the comparison between the results of simula-
tion and the results of our experiments at 298–318 K. Experiments
were conducted with 300 ± 10 ppm NO. Because the molar ratio
1.2
1.0
N2O5
Absorbance (a.u.)
should be considered. Many reactions occurred in the process of
NO oxidation. However, reactions R1, R2 and R3 dominated
at first due to their high reaction rates. Reactions R2 and R3
indicated the deep transformation of NO2 into NO3 or N2O5.
Therefore, the concentration of N2O5 could not be ignored when
the molar ratio O3/NO was >1. Based on the comparison of the
reaction rate of R2 and R3 (Table 1), the existence of NO3 was
neglected because of the lower reaction rate of R2.
From Fig. 4, the results of the simulation agreed well with
the results of the experiments. The above comparisons of
Mok's and our experiments demonstrated that the mecha-
nism was suitable and accurate for the O3–NOx reactions.
4.2. Reaction parameter studies for the generation of NO2
As we discussed above, the generation of NO2 could be easily
achieved when the molar ratio of O3/NO was below 1. Under
these molar ratios of O3/NO, the major oxidation product was
NO2. However, the exhaust gas conditions of the power plants
and industrial boilers vary due to the complicated operating
conditions. Typically, as Saidur et al. (2010) reported, the
temperature of the exhaust gas leaving industrial boilers
ranges from 423 to 523 K. Due to the various types of industrial
boilers, working conditions of boilers and other reasons, the
NOx concentration of industrial boilers could range from 72

0.8 to 455 ppm (Costa and Azevedo, 2007; Costa et al., 2003;
Yamamoto et al., 2010). Thus, the effect of flue gas conditions
0.6
on the generation of NO2 should be considered and the
experimental results are shown in Fig. 5.
0.4

0.2 4.2.1. Effect of temperature on the production of NO2

Fig. 5a shows the effect of temperature on the production of
0 NO2. The simulation was carried out by keeping the residence
600 800 1000 1200 1400 1600 1800 2000 time at 5 sec and the molar ratio of O3/NO at 1. In Fig. 5, the
Wavenumber (cm -1 ) concentration of NO2 was 294.65 ppm at 333 K when the
initial NO concentration was 300 ppm. It could be concluded
Fig. 2 – FTIR spectra gathered from the experiment for that the NO was almost completely converted to NO2. Com-
NO initial concentration of 300 ppm and molar ratio of paring the concentrations of NO2 at 333 and 423 K, it was
O3/NO = 1.75. FTIR: Fourier transform indicated that the higher temperature was beneficial to the
infrared spectroscopy. generation of NO2, but the effect was not remarkable. Due to
54 J O U RN A L OF E N V I RO N ME N TA L SC IE N CE S 4 1 (2 0 1 6) 51–5 8

Table 1 – Detailed O3–NOx kinetic reaction mechanism.

Reaction no. Reaction A Ea β Reaction No. Reaction A Ea β
(J/mol) (J/mol)

1 O3 + NO = NO2 + O2 8.43E+11 2600 0 39 HNO3 + O = NO3 + OH 1.81E+07 0 0

2 O3 + NO2 = O2 + NO3 8.43E+10 4870 0 40 HNO3 + H = NO3 + H2 3.40E+12 16332 1.5
3 NO2 + NO3 = N2O5 7.98E+17 0 −3.9 41 HNO3 + H = NO2 + H2O 8.39E+09 6255 3.3
4 O3 + H = O2 + OH 1.64E+13 750 0.8 42 HNO3 + NO = NO2 + HNO2 4.48E+03 0 0
5 O3 + H = O + HO2 4.52E+11 0 0 43 HNO3 + OH = NO2 + H2O2 4.82E+08 0 0
6 O3 + OH = O2 + HO2 1.15E+12 1990 0 44 HNO3 + OH = NO3 + H2O 5.36E+10 0 0
7 O3 + H2O = O2 + H2O2 6.62E+01 0 0 45 HNO3 = NO2 + OH 6.90E+17 45730 0
8 O3 + HO2 = 2O2 + OH 1.19E+08 1380 4.6 46 HNO + O = NO + OH 2.29E+13 0 0
9 O2 + O = O3 4.86E+13 0 −1 47 HNO + HNO = N2O + H2O 2.55E+07 1188 4
10 O3 + O = 2O2 4.82E+12 4090 0 48 HNO + H = NO + H2 2.70E+13 651 0.3
11 H + O2 + M = M + HO2 3.61E+17 0 −0.7 49 HNO + NO2 = NO + HNO2 6.03E+11 1980 0
12 H + H + M = H2 + M 1.00E+18 0 −1 50 HNO + OH = NO + H2O 4.82E+13 990 0
13 H + H + H2 = 2H2 9.20E+16 0 −0.6 51 H + HO2 = OH + OH 8.22E+12 0 0.8
14 H + H + H2O = H2 + H2O 6.00E+19 0 −1.3 52 NO2 + HO2 = HNO2 + O2 2.20E-01 0 0
15 H + OH + M = H2O + M 1.60E+22 0 −2 53 NO + HO2 = HNO + O2 5.84E+05 5600 0
16 H + O + M = OH + M 6.20E+16 0 −0.6 54 NO + HO2 = NO2 + OH 6.32E+11 1430 0.6
17 O + O + M = O2 + M 1.89E+13 -1788 0 55 NO + HO2 = HNO3 3.47E+12 −5720 0
18 H2O2 + M = OH + OH + M 1.30E+17 45500 0 56 H2O + HO2 = H2O2 + OH 2.80E+13 32790 0
19 H2 + O2 = 2OH 1.70E+13 47780 0 57 H2O2 + HO2 = O2 + H2O + OH 6.03E+10 0 0
20 OH + H2 = H2O + H 1.17E+09 3626 1.3 58 OH + HO2 = O2 + H2O 4.28E+13 110 −0.2
21 O + OH = O2 + H 3.61E+14 0 −0.5 59 HO2 + HO2 = H2O2 + O2 1.87E+12 1540 0
22 O + H2 = OH + H 5.06E+04 6290 2.7 60 HO2 = H + O2 1.45E+16 48490 −1.2
23 O + HO2 = O2 + OH 1.40E+13 1073 0 61 O3 + N = O2 + NO 6.00E+07 0 0
24 2OH = O + H2O 6.00E+08 0 1.3 62 N + NO2 = O + O + N2 1.30E-01 0 0
25 H + HO2 = H2 + O2 1.25E+13 0 0 63 NO2 + H = NO + OH 2.41E+14 680 0
26 H2O2 + H = HO2 + H2 1.60E+12 3800 0 64 NO + NO = O2 + N2 3.10E+13 63190 0
27 N + O2 = NO + O 6.40E+09 6280 1 65 NO + N2O = NO2 + N2 1.73E+11 46300 2.2
28 N + OH = NO + H 3.80E+13 0 0 66 NO3 + O = NO2 + O2 1.02E+13 0 0
29 NO + M = N + O + M 9.64E+14 620910 0 67 NO3 + H = OH + NO2 5.66E+13 0 0
30 N + HO2 = NO + OH 1.00E+13 8390 0 68 NO3 + NO3 = O2 + NO2 + NO2 2.71E+10 2504 0
31 NO + N = N2 + O 3.27E+12 0 0.3 69 NO + NO3 = NO2 + NO2 1.08E+13 −219 0
32 NO + O = NO2 1.30E+15 0 −0.8 70 HO2 + NO3 = HNO3 + O2 1.15E+12 0 0
33 NO + OH = HNO2 5.45E+17 0 0 71 HO2 + NO3 = OH + O2 + NO2 1.51E+12 0 0
34 O + HNO2 = NO2 + OH 1.21E+13 5938 0 72 OH + NO3 = HO2 + NO2 1.20E+13 0 0
35 H + HNO2 = HNO + OH 7.57E+12 4914 0.9 73 H2O2 + NO3 = HO2 + HNO3 1.21E+09 0 0
36 H + HNO2 = NO2 + H2 1.37E+12 6592 1.6 74 HNO2 + NO3 = HNO3 + NO2 1.21E+09 0 0
37 H + HNO2 = NO + H2O 3.85E+11 3840 1.9 75 O3 + NO3 = O2 + O2 + NO2 6.03E+06 0 0
38 O3 + HNO2 = O2 + HNO 3.01E+05 0 0 76 H2O + N2O5 = HNO3 + HNO3 1.51E+02 0 0

Reaction rate constants (k) expressed as: k = ATβ exp(−Ea/RT). A: exponential factor; T: temperature, unit K; Ea: activation energy; R: gas constant
8.3145 J/(mol · K); β: temperature exponent.

the kinetic reaction mechanism, the increase of temperature

300 was beneficial to the reaction between NO and O3. However
the reaction rate of R1 was very high, and the effect of reaction
250
temperature could be neglected.
Concentration (ppm)

200
NO sim 4.2.2. Effect of the residence time on the production of NO2
150 NO2 sim The effect of the residence time on the production of NO2 is
NO Mok presented in Fig. 5b. The simulation was conducted at 373 K,
100 NO2 Mok
when the molar ratio of O3/NO was 1. In Fig. 5b, when the
50 initial NO concentration was 300 ppm and the residence time
was 0.417 sec, the concentration of NO2 was 220.68 ppm,
0 which meant that the production rate of NO2 was only 73.56%.
0 50 100 150 200 250 300 When the residence time was extended to 1.25 and 2.08 sec,
Ozone added (ppm) the concentration of NO2 was 286.83 and 294.76 ppm, respec-
tively. Hence, the production rate of NO2 only gained 3%
Fig. 3 – Comparison between the simulation results (sim) increment, when the residence time changed from 1.25 to
and Mok's experiment (Mok) results. Initial conditions: 2.08 sec. For industrial application, 95.61% oxidation percent-
[NO] = 280 ppm, [NO2] = 20 ppm, residence time t = 2.94 sec, age in 1.25 sec would be sufficient. Moreover, the oxidation
and reaction temperature T = 443 K. rates of NO for two other initial NO concentrations were also
 J O U RN A L OF E N V I RO N ME N TA L SC IE N CE S 4 1 (2 0 1 6) 5 1–5 8 55

90 450
400
a
80
N2O5 concentration (ppm)

Concentration of NO2 (ppm)

70 350

60 300

50 250

40 200
298 K.sim
30 308 K.sim 150
318 K.sim
20 298 K.exp 100 200 ppm
308 K.exp 300 ppm
10 50 400 ppm
318 K.exp
0 0
1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1 320 340 360 380 400 420 440
Molar ratio of O3 to NO Temperature (K)

400
Fig. 4 – Comparison between the simulation results (sim) b
and our experiment results (exp). Initial conditions: 350

Concentration of NO2 (ppm)

[NO] = 300 ppm, t = 5 sec, and reaction temperature
300
T = 298–318 K.
250

200

150
high when the residence time was 1.25 sec. Thus the optimal
residence time can be set as 1.25 sec in order to decrease the 100 200 ppm
volume of the reactor. 300 ppm
50 400 ppm
0
4.2.3. Effect of the molar ratio of O3/NO on producing NO2 0 1 2 3 4 5
Fig. 5c displays the effect of the molar ratio of O3/NO on the Residence time (sec)
production of NO2. The simulation was executed at 373 K and
400
5 sec. With the increase of the molar ratio of O3/NO, the c
concentration of NO2 increased until the molar ratio of O3/NO 350
Concentration of NO2 (ppm)

reached 1.0 for all three initial NO concentrations. The 300

reaction of R1 was very fast, thus R1 was the dominant
reaction step when the molar ratio of O3/NO was less than 1. 250
We also observed the same phenomenon in our previous 200
research (Sun et al., 2014). We found that the concentration of
150
NO decreased and the generation of NO2 increased in
proportion to the increase of the injected ozone concentra- 100
200 ppm
tion. We also discovered that when the molar ratio of O3/NO 50 300 ppm
was less than 1, NO was gradually oxidized to NO2 and no 400 ppm
higher NOx chemical state products were formed. When the 0
0.2 0.4 0.6 0.8 1.0 1.2
molar ratio of O3/NO was more than 1, N2O5 was produced. Molar ratio of O3/NO
The generation of N2O5 was mainly ascribed to reactions such
as R2 and R3. In order to obtain NO2, the molar ratio of O3/NO Fig. 5 – Effects of (a) temperature, (b) residence time, and
can simply be selected as 1. (c) molar ratio of O3/NO on the production of NO2 at initial NO
concentration 200–400 ppm. Conditions: (a) residence time
4.3. Reaction parameters studies for the generation of N2O5 5 sec, molar ratio of O3/NO = 1; (b) reaction temperature
373 K, molar ratio of O3/NO = 1; (c) residence time 5 sec,
As analyzed above, N2O5 is the easiest to absorb among all reaction temperature 373 K.
oxidation states of nitrogen oxides. However, the generation
conditions of N2O5 are relatively strict. In order to get the
optimal conditions for the generation of N2O5 to reduce the
operating cost, research on the reaction parameters for
producing N2O5 is needed, and the experimental results are with the molar ratio of O3/NO at 1.75. As shown in Fig. 6a, with the
shown in Fig. 6. increase of the temperature of the exhaust gas, the concentration
of N2O5 increased until the temperature of the exhaust gas
4.3.1. Effect of temperature on the production of N2O5 reached 383 K for all three initial NO concentrations. When
Fig. 6a shows the effect of temperature on the production of the temperature of gas was higher than 383 K, the con-
N2O5. The investigated temperature range in this study varied centration of the produced N2O5 decreased very rapidly. The
from 333 to 423 K. The simulation was carried out at 5 sec decrease of N2O5 at this temperature should be ascribed
56 J O U RN A L OF E N V I RO N ME N TA L SC IE N CE S 4 1 (2 0 1 6) 51–5 8

350
the exhaust should be controlled strictly for the production
a 200 ppm
300 ppm of N2O5. If the temperature of the exhaust is higher than
300
Concentration of 2N2O5 (ppm)

400 ppm 373 K, facilities such as a sprinkling system should be used in

250 the system.

200 4.3.2. Effect of the residence time on the production of N2O5

Fig. 6b presents the effect of the residence time on the
150
production of N2O5. The simulation was conducted at 353 K
100 with the molar ratio of O3/NO at 1.75. As shown in Fig. 6b, when
the initial NO concentration was 300 ppm and the residence
50 time was 7, 8 and 9 sec, the concentration of 2N2O5 was 222.25,
0
231.05 and 239.15 ppm, respectively. This meant that the
320 340 360 380 400 420 440 production of N2O5 varied from 74.08% to 79.71% when the
Temperature (K) residence time was changed from 7 to 9 sec. Clearly, the
400
concentration of N2O5 increased with increasing residence
350 b time. These results indicated that longer residence time was
Concentration of 2N2O5 (ppm)

favorable for the formation of N2O5, so that less NO2 was left in
300
the reaction zone. Dora et al. (2009) also observed that residence
250 time lower than 8 sec reduced the effectiveness of NO
ozonation when the applied molar ratio of O3/NO was equal to
200
1.14. Thus, the residence time should be chosen at about 8 sec.
150
4.3.3. Effect of the molar ratio of O3/NO on the production of
100
200 ppm N2O5
50 300 ppm Fig. 6c shows the effect of the molar ratio of O3/NO on the
400 ppm
production of N2O5. The simulation was executed at 353 K and
0
0 2 4 6 8 10 the residence time was selected as 5 sec. With the increase of
Residence time (sec) molar ratio of O3/NO, the concentration of N2O5 increased.
350
When the initial NO concentration was 300 ppm and the
c
300 molar ratio of O3/NO varied from 1.45 to 2, the concentration
Concentration of 2N2O5 (ppm)

200 ppm
300 ppm of N2O5 increased from 187.16 to 275.78 ppm. This showed
250 400 ppm that the production rate of N2O5 grew from 62.38% to 91.93%
when the O3/NO molar ratio varied from 1.45 to 2.
200
If the reaction rates of R1, R2 and R3 are high enough, the
150 molar ratio O3/NO should be selected as 1.5 to ensure N2O5 as
the main product. However, the reaction rates of R3 and
100 especially R2 were not very high. Thus, N2O5 was not easy to
generate. In order to produce more N2O5, more O3 is needed.
50
We also found that the amount of NO2 was small when the
0 molar ratio of O3/NO was 2 (Sun et al., 2014). Considering the
1.0 1.2 1.4 1.6 1.8 2.0 usage of O3 and the production rate of N2O5, the molar ratio
Molar ratio of O3/NO
O3/NO can be chosen as about 1.75.

Fig. 6 – Effect of (a) temperature, (b) residence time, and

(c) molar ratio of O3/NO on producing N2O5 at initial NO
5. Conclusions
concentration 200–400 ppm. Conditions: (a) residence time
5 sec, molar ratio of O3/NO = 1.75; (b) reaction temperature In this study, suitable process conditions of NO2 and N2O5
353 K, molar ratio of O3/NO = 1.75; (c) residence time 5 sec, were investigated by numerical simulation. A new 20-species,
reaction temperature 353 K. 76-step detailed kinetic mechanism was proposed between O3
and NOx. The concentration of N2O5 was measured by an IR
spectrometer. The calculation results showed good agree-
ment with both the published experimental results and our
to the decomposition of O3, NO3 and N2O5. High reaction experimental results. Due to the high oxidation rate of NO by
temperature accelerated the reaction rate of the reverse ozone, the effect of temperature on the generation of NO2
reactions such as R3 and R9, and the forward reactions such can be neglected. The production rate of NO2 for 300 ppm NO
as R61 and R69. We also observed a similar phenomenon in varied from 95.61% to 98.28% when the residence time changed
our previous study (Sun et al., 2014). We found that the N2O5 from 1.25 to 2.08 sec. The optimal residence time can be set as
decreased with the increasing reaction temperature in the 1.25 sec in order to decrease the volume of the reactor. With the
high temperature region (373–453 K). Even at 453 K, we found increase of the molar ratio of O3/NO, the concentration of NO2
that N2O5 and O3 disappeared. Therefore, the temperature of increased until the molar ratio of O3/NO reached 1.0. In order to
 J O U RN A L OF E N V I RO N ME N TA L SC IE N CE S 4 1 (2 0 1 6) 5 1–5 8
obtain sufficient NO2, the molar ratio of O3/NO can be selected as
1.0 to reduce the usage of O3.
N2O5 is the easiest to absorb among all the states of
nitrogen oxides, but the reaction conditions for the generation
of N2O5 are rigorous. With the increase of the temperature, the
concentration of N2O5 increased until the reaction temperature
reached 383 K. The temperature of the exhaust gas should be
controlled strictly for a high production rate of N2O5. The
production rate of N2O5 for 300 ppm NO varied from 74.08%
to 79.71% when the residence time was changed from 7 to
9 sec. The residence time can be chosen as 8 sec in order to
produce enough N2O5. The production rate of N2O5 for 300 ppm
NO concentration varied from 62.38% to 91.93% when the molar
ratio of O3/NO was from 1.45 to 2. Considering both the usage of
O3 and the production rate of N2O5, the molar ratio O3/NO can be
selected as about 1.75.
Overall, the study provided suitable reaction parameters
for the generation of NO2 or N2O5, depending on specific
emission requirements and the changing exhaust conditions.
Meanwhile, the model is a powerful tool to develop an effective
and economical reactor for the ozonation of NO.
Acknowledgments
This work was financially supported by the National High
Tech Research and Development Program (863) of China (No.
2011AA060801) and the Program for Zhejiang Leading Team of
S&T Innovation (No. 2013TD07).
REFERENCES
Abián, M., Giménez-López, J., Bilbao, R., Alzueta, M.U., 2011. Effect
of different concentration levels of CO2 and H2O on the
oxidation of CO: experiments and modeling. Proc. Combust.
Inst. 33 (1), 317–323.
Costa, M., Azevedo, J.L.T., 2007. Experimental characterization of
an industrial pulverized coal-fired furnace under deep staging
conditions. Combust. Sci. Technol. 179 (9), 1923–1935.
Costa, M., Silva, P., Azevedo, J.L.T., 2003. Measurements of gas
species, temperature, and char burnout in a low-NOx
pulverized-coal-fired utility boiler. Combust. Sci. Technol. 175
(2), 271–289.
Datebase, d. Available at: http://kinetics.nist.gov.
Dora, J., Gostomczyk, M.A., Jakubiak, M., Kordylewski, W., Mista,
W., Tkacuzuk, M., 2009. Parametric studies of the effectiveness
of NO oxidation process by ozone. Chem. Process. Eng. 32 (1),
621–634.
Gui, Y., Deng, K.Y., Xu, M., Shi, L., Sun, Y.C., 2014. An integrated
model for negative valve overlap early injection HCCI
combustion. J. Energy Inst. 87 (4), 341–353.
Hou, B., Zhuang, G.S., Zhang, R., Liu, T.N., Guo, Z.G., Chen, Y., 2011.
The implication of carbonaceous aerosol to the formation of
haze: revealed from the characteristics and sources of OC/EC
over a mega-city in China. J. Hazard. Mater. 190 (1-3), 529–536.
Hu, G.X., Sun, Z.G., Gao, H.Y., 2010. Novel process of simultaneous
removal of SO2 and NO2 by sodium humate solution. Environ.
Sci. Technol. 44 (17), 6712–6717.
Huang, R.J., Zhang, Y.L., Bozzetti, C., Ho, K.F., Cao, J.J., Han, Y.M., et al.,
2014. High secondary aerosol contribution to particulate pollution
during haze events in China. Nature 514 (7521), 218–222.
57
Jiang, J.K., Zhou, W., Cheng, Z., Wang, S.X., He, K.B., Hao, J.M., 2014.
Particulate matter distributions in China during a winter
period with frequent pollution episodes (January 2013).
Aerosol Air Qual. Res. 15, 494–503.
Kee, R.J., Rupley, F.M., Miller, J.A., 2006. CHEMKIN 4.1. Theory
manual.
Li, Q.H., Meng, A.H., Jia, J.Y., Zhang, Y.G., 2010. Investigation of
heavy metal partitioning influenced by flue gas moisture and
chlorine content during waste incineration. J. Environ. Sci. 22
(5), 760–768.
Liu, Y.X., Zhang, J., 2011. Photochemical oxidation removal of NO
and SO2 from simulated flue gas of coal-fired power plants by
wet scrubbing using UV/H2O2 advanced oxidation process. Ind.
Eng. Chem. Res. 50 (7), 3836–3841.
Liu, J., Li, J., Zhang, Y.L., Liu, D., Ding, P., Shen, C.D., et al., 2014.
Source apportionment using radiocarbon and organic tracers
for PM2.5 carbonaceous aerosols in Guangzhou, South China:
contrasting local- and regional-scale haze events. Environ. Sci.
Technol. 48 (40), 12002–12011.
Mok, Y.S., Lee, H.J., 2006. Removal of sulfur dioxide and nitrogen
oxides by using ozone injection and absorption–reduction
technique. Fuel Process. Technol. 87 (7), 591–597.
Nelo, S.K., Leskela, K.M., Sohlo, J.J.K., 1997. Simultaneous
oxidation of nitrogen oxides and sulfur dioxide with ozone and
hydrogen peroxide. Chem. Eng. Technol. 20 (1), 40–42.
Nishimura, M., Suzuki, T., Nakanishi, R., Kitamura, R., 1997.
Low-NOx combustion under high preheated air temperature
condition in an industrial furnace. Energy Convers. Manag. 38
(10-13), 1353–1363.
Obradović, B.M., Sretenović, G.B., Kuraica, M.M., 2011. A dual-use
of DBD plasma for simultaneous NOx and SO2 removal from
coal-combustion flue gas. J. Hazard. Mater. 185 (2-3),
1280–1286.
Price, D., Birnbaum, R., Batiuk, R., McCullough, R., Smith, R., 1997.
Nitrogen Oxides Impacts on Public Health and the
Environment. Environmental Protection Agency. Office of Air
and Radiation, Washington, DC (United States).
Raju, T., Chung, S.J., Moon, I.S., 2008. Novel process for
simultaneous removal of NOx and SO2 from simulated flue gas
by using a sustainable Ag (I)/Ag (II) redox mediator. Environ.
Sci. Technol. 42 (19), 7464–7469.
Saidur, R., Ahamed, J.U., Masjuki, H.H., 2010. Energy, exergy and
economic analysis of industrial boilers. Energ Policy 38 (5),
2188–2197.
Skalska, K., Miller, J.S., Ledakowicz, S., 2011. Kinetic model of NOx
ozonation and its experimental verification. Chem. Eng. Sci. 66
(14), 3386–3391.
Sumathi, S., Bhatia, S., Lee, K.T., Mohamed, A.R., 2010. Selection of
best impregnated palm shell activated carbon (PSAC) for
simultaneous removal of SO2 and NOx. J. Hazard. Mater. 176 (1-3),
1093–1096.
Sun, C.L., Zhao, N., Zhuang, Z.K., Wang, H.Q., Liu, Y., Weng, X.L., et
al., 2014. Mechanisms and reaction pathways for simultaneous
oxidation of NOx and SO2 by ozone determined by in situ IR
measurements. J. Hazard. Mater. 274, 376–383.
Tayyeb Javed, M., Irfan, N., Gibbs, B.M., 2007. Control of
combustion-generated nitrogen oxides by selective
non-catalytic reduction. J. Environ. Manag. 83 (3), 251–289.
Topsøe, N.Y., 1994. Mechanism of the selective catalytic reduction
of nitric oxide by ammonia elucidated by in situ on-line
Fourier transform infrared spectroscopy. Science 265 (5176),
1217–1219.
Wang, Z.H., Zhou, J.H., Fan, J.R., Cen, K.F., 2006. Direct numerical
simulation of ozone injection technology for NOx control in
flue gas. Energy Fuel 20 (6), 2432–2438.
Wang, Z.H., Zhou, J.H., Zhu, Y.Q., Wen, Z.C., Liu, J.Z., Cen, K.F.,
2007. Simultaneous removal of NOx, SO2 and Hg in nitrogen
flow in a narrow reactor by ozone injection: experimental
results. Fuel Process. Technol. 88 (8), 817–823.
58 J O U RN A L OF E N V I RO N ME N TA L SC IE N CE S 4 1 (2 0 1 6) 51–5 8
Wang, Z.H., Zhang, X., Zhou, Z.J., Chen, W.Y., Zhou, J.H., Cen, K.F.,
2012. Effect of additive agents on the simultaneous absorption
of NO2 and SO2 in the calcium sulfite slurry. Energy Fuel 26 (9),
5583–5589.
Wängberg, I., 1993. An Experimental Study of Some Nitrate
Radical Reactions of Importance to Tropospheric Chemistry
(PhD thesis) Inorg. Chem. Gothenburg University, Gothenburg.
Yamamoto, T., Fujishima, H., Okubo, M., Kuroki, T., 2010.
Pilot-scale and removal from boiler emission using
indirect-plasma and chemical hybrid process. IEEE Trans. Ind.
Appl. 46 (1), 29–37.
Yang, H., Hou, W., Zhang, H., Zhou, L., 2013. Kinetic interpretation
on mercury oxidation and transformation in simulated flue
gases. Int. J. Environ. Sci. Technol. 10 (4), 689–696.
Zajemska, M., Musiał, D., Poskart, A., 2014. Application of
CHEMKIN and COMSOL programs in the calculations of
chemical composition of natural gas combustion products.
Combust. Sci. Technol. 186 (2), 153–172.
Zhao, Y., Han, Y.H., Ma, T.Z., Guo, T.X., 2011. Simultaneous
desulfurization and denitrification from flue gas by ferrate(VI).
Environ. Sci. Technol. 45 (9), 4060–4065.