Dependence of N 2 O/NO Decomposition and Formation on Temperature and Residence Time in Thermal Reactor

: Nitrogen dioxide (N 2 O) is a greenhouse gas that is harmful to the ozone layer and con-tributes to global warming. Many other nitrogen oxide emissions are controlled using the selective non-catalytic reaction (SNCR) process, but N 2 O reduction methods are few. To avoid future air pollution problems, N 2 O reduction from industrial sources is essential. In this study, a N 2 O decomposition and NO formation under an argon atmospheric N 2 O gas mixture were observed in a lab-scale SNCR system. The reaction rate and mechanism of N 2 O were calculated using a reaction path analyzer (CHEMKIN-PRO). The residence time of the gas mixture and the temperature in the reactor were set as experimental variables. The results conﬁrmed that most of the N 2 O was converted to N 2 and NO. The change in the N 2 O reduction rate increased with the residence time at 1013 and 1113 K, but decreased at 1213 K due to the inverse reaction. NO concentration increased with the residence time at 1013 and 1113 K, but decreased at 1213 K owing to the conversion of NO back to N 2 O.


Introduction
Nitrous oxide (N 2 O), carbon dioxide (CO 2 ), and methane (CH 4 ) are considered greenhouse gases. N 2 O is very stable in air, and takes an average of 135 years to decompose naturally. Upon reaching the stratosphere, N 2 O reacts with an oxygen atom (O) to form nitrogen monoxide (NO). NO reacts with ozone (O 3 ) again in the ozone layer and destroys O 3 in a chain [1,2]. Therefore, N 2 O has 310 times higher global warming potential (GWP) than CO 2 . N 2 O was designated as a greenhouse gas by the Kyoto Protocol in 1997. Afterward, N 2 O was regulated as a greenhouse gas under the Paris Agreement adopted at the UN Climate Change Conference in 2015. However, in most countries, including South Korea, specific reduction policies for N 2 O are still inadequate [3]. N 2 O reduction methods can be divided into two methods. The first is the inhibition of N 2 O production. The second is N 2 O removal at the end of the exhaust [4][5][6]. In the first method, N 2 O generation can be suppressed using a fluid medium in the fluidized bed combustion process (FBC) and by using a catalyst mounted on a generation source such as a vehicle [4][5][6]. However, this method is greatly affected by the N 2 O temperature and pressure conditions. The second method involves reducing the N 2 O generated in the incinerator and the industrial process at a subsequent stage [4][5][6]. This technology of N 2 O reduction at the end of the process is largely divided into selective catalytic reduction (SCR) and selective non-catalytic reduction (SNCR). The SCR process uses a catalyst, so the facility can be operated at a lower temperature than that required for SNCR. The SNCR process operates at a relatively high temperature (approximately 1000-1500 K) compared with the SCR process, and the initial investment cost is lower than other processes [4][5][6][7][8].
The SNCR process is a NOx reducing process, where the NOx generated in the combustion process of an incinerator is reduced using a reductant. Specifically, a reductant such as ammonia (NH 3 ) and urea aqueous solution (NH 2 CONH 2 ) is injected into the chamber at temperatures of 1173-1373 K, thereby reducing NOx to nitrogen (N 2 ) and water vapor [9]. In general, NH 3 is more efficient than the urea aqueous solution, but urea is easier to handle and results in a less costly process [10,11]. Figure 1 shows the decomposition mechanism and reaction pathway of NH 3 and urea aqueous solution in the SNCR process [12]. In the process, the Zel'dovich mechanism is used as the basic reaction pathway for oxidation and deoxidation of N 2 [13].
The SNCR process is a NOx reducing process, where the NOx generated in the combustion process of an incinerator is reduced using a reductant. Specifically, a reductant such as ammonia (NH3) and urea aqueous solution (NH2CONH2) is injected into the chamber at temperatures of 1173-1373 K, thereby reducing NOx to nitrogen (N2) and water vapor [9]. In general, NH3 is more efficient than the urea aqueous solution, but urea is easier to handle and results in a less costly process [10,11]. Figure 1 shows the decomposition mechanism and reaction pathway of NH3 and urea aqueous solution in the SNCR process [12]. In the process, the Zel'dovich mechanism is used as the basic reaction pathway for oxidation and deoxidation of N2 [13]. Basic decomposition mechanism of nitrous oxide using ammonia and urea aqueous solution reductants [13].
In a SNCR process that uses urea aqueous solution as a reductant, NOx is reduced to N2 and N2O [14]. Svoboda et al. reported that the reductant employed in the process affects the production of N2O [15]. Moreover, Kim reported that a small amount of N2O was generated when urea aqueous solution was injected for NOx removal in the operation of a waste incinerator [16].
Carbon monoxide (CO) and oxygen (O2) are typical reaction gases that affect the SNCR process. In the oxygen-free process, the NOx reducing reaction occurs at temperatures of 1400 K or higher. This results from the NO reduction proceeding at a relatively high temperature due to the low OH concentration, thereby leading to an increase in the reaction temperature. Im reported that the optimal O2 concentration in the SNCR process is 5-7% [17].
In a study on the residence time, Liang et al. reported that the optimum temperature for N2O decomposition and NOx formation decreased with increasing residence time of the mixture [18]. Various studies considering the dependence of NO reduction on the temperature and residence time have used ammonia as a reductant in the SNCR process [19][20][21][22]. Duo et al. reported that the optimum temperature for NO reduction decreased with increasing residence time [23].
Previous studies have confirmed that the reduction efficiency of N2O is determined by complex factors such as the amount of reductant, mixture composition, and residence time. Most of those studies focused on controlling the amount of N2O generated during the NOx reduction process in the SNCR. Additionally, studies of N2O decomposition have mostly used catalysts. [24][25][26]. In the SNCR system, N2O is less regulated than NOx; hence, the reduction of N2O has rarely been investigated. Therefore, in this study, we used a Figure 1. Basic decomposition mechanism of nitrous oxide using ammonia and urea aqueous solution reductants [13].
In a SNCR process that uses urea aqueous solution as a reductant, NOx is reduced to N 2 and N 2 O [14]. Svoboda et al. reported that the reductant employed in the process affects the production of N 2 O [15]. Moreover, Kim reported that a small amount of N 2 O was generated when urea aqueous solution was injected for NOx removal in the operation of a waste incinerator [16].
Carbon monoxide (CO) and oxygen (O 2 ) are typical reaction gases that affect the SNCR process. In the oxygen-free process, the NOx reducing reaction occurs at temperatures of 1400 K or higher. This results from the NO reduction proceeding at a relatively high temperature due to the low OH concentration, thereby leading to an increase in the reaction temperature. Im reported that the optimal O 2 concentration in the SNCR process is 5-7% [17].
In a study on the residence time, Liang et al. reported that the optimum temperature for N 2 O decomposition and NOx formation decreased with increasing residence time of the mixture [18]. Various studies considering the dependence of NO reduction on the temperature and residence time have used ammonia as a reductant in the SNCR process [19][20][21][22]. Duo et al. reported that the optimum temperature for NO reduction decreased with increasing residence time [23].
Previous studies have confirmed that the reduction efficiency of N 2 O is determined by complex factors such as the amount of reductant, mixture composition, and residence time. Most of those studies focused on controlling the amount of N 2 O generated during the NOx reduction process in the SNCR. Additionally, studies of N 2 O decomposition have mostly used catalysts. [24][25][26]. In the SNCR system, N 2 O is less regulated than NOx; hence, the reduction of N 2 O has rarely been investigated. Therefore, in this study, we used a laboratory-scale thermal reactor that simulated an actual SNCR system, and observed the N 2 O and NO behaviors that were generated in combustion furnace. The reaction characteristics of N 2 O and NO were studied through thermal decomposition experiments, where the temperature and residence time in an argon (Ar) atmosphere gas mixture were varied. Furthermore, the reaction equation and reaction rate (rate of progress) of N 2 O and NO were examined using a chemical reaction program (CHEMKIN-PRO) that simulated the same conditions as the experiment. Correlations between the reduction of N 2 O and NO, and the reaction rate based on the temperature and residence time, were investigated. Figure 2 shows a schematic of the experimental setup used to investigate the N 2 O thermal decomposition characteristics. The setup consisted of a gas inlet part, a thermal reactor, a measuring part, and an exhaust port. The flow of experimental gases Ar, O 2 , and N 2 O into the thermal reactor were controlled using a mass flow controller (MFC). The reactor had a cylindrical structure (inner diameter: 140 mm, and length: 1000 mm), was an electric furnace capable of heating up to 1273 K, and was heated with four internal heating coils. The temperature inside the reactor was measured using an R-type thermocouple placed in the center of each coil. After the thermal reaction, the exhaust gas passed through a heat exchanger. A non-dispersive infrared (NDIR) type gas meter (Sensonic IR-1, for the N 2 O measurement) and a gas analyzer (MK 6000+, com GmbH, for the NOx and O 2 measurement) were used to measure the gas components. The measurement error of the NDIR gas meter (equipped with an electrochemical sensor) was 3%. The measurement error of NO and N 2 O of the gas analyzer (also equipped with an electrochemical sensor) was 2% [27].

Experimental Setup
laboratory-scale thermal reactor that simulated an actual SNCR system, and observed the N2O and NO behaviors that were generated in combustion furnace.
The reaction characteristics of N2O and NO were studied through thermal decomposition experiments, where the temperature and residence time in an argon (Ar) atmosphere gas mixture were varied. Furthermore, the reaction equation and reaction rate (rate of progress) of N2O and NO were examined using a chemical reaction program (CHEM-KIN-PRO) that simulated the same conditions as the experiment. Correlations between the reduction of N2O and NO, and the reaction rate based on the temperature and residence time, were investigated. Figure 2 shows a schematic of the experimental setup used to investigate the N2O thermal decomposition characteristics. The setup consisted of a gas inlet part, a thermal reactor, a measuring part, and an exhaust port. The flow of experimental gases Ar, O2, and N2O into the thermal reactor were controlled using a mass flow controller (MFC). The reactor had a cylindrical structure (inner diameter: 140 mm, and length: 1000 mm), was an electric furnace capable of heating up to 1273 K, and was heated with four internal heating coils. The temperature inside the reactor was measured using an R-type thermocouple placed in the center of each coil. After the thermal reaction, the exhaust gas passed through a heat exchanger. A non-dispersive infrared (NDIR) type gas meter (Sensonic IR-1, for the N2O measurement) and a gas analyzer (MK 6000+, com GmbH, for the NOx and O2 measurement) were used to measure the gas components. The measurement error of the NDIR gas meter (equipped with an electrochemical sensor) was 3%. The measurement error of NO and N2O of the gas analyzer (also equipped with an electrochemical sensor) was 2% [27]. Figure 3 shows a photograph of a laboratory-scale SNCR apparatus for N2O thermal decomposition. To prevent heat loss and to maintain the temperature employed in the experimental conditions, the reactor was surrounded by an insulator.     Table 1 shows the composition of the gas mixture used in the N2O thermal decomposition experiment [27]. The gas mixture was composed of 94.96% Ar, 5% O2, and 400 ppm of N2O to simulate the exhaust gas in a selective non-catalytic reduction process. The actual combustion exhaust gas is N2 and CO2 atmosphere. However, in this experiment, an Ar atmosphere was employed to minimize the involvement of nitrogen components from sources, in addition to the N2 generated by thermal decomposition of N2O. Moreover, the molecular weights of N2O and CO2 were the same, and hence measurement errors occurred in the NDIR measurement. Therefore, no hydrocarbon-based material was added. The chemical reaction time of NOx and N2O depends on the residence time and temperature. Accordingly, the residence time of the gas mixture in the thermal reactor was selected as an experimental variable. The dependence of the N2O thermal decomposition on the residence time was investigated by fixing the composition of the gas mixture and varying the residence time (times of 10, 20, and 40 s were employed). The gas mixture flow rate corresponding to each residence time was calculated from the equation of the reactor volume and residence time (Equation (1)). In this equation, Q is the gas mixture flow rate, D is the inner diameter of the reactor, L is the length of the reactor, and t is the residence time. Table 2 shows the experimental conditions [27]. The experimental temperature was increased in 20 K intervals for temperatures ranging from 1023 to 1223 K, which is the general operating temperature of the SNCR process. In order to maintain the residence time of the gas mixture on each temperature section, the flow rate was adjusted based on Charles's law, which is given as follows:

Experimental Conditons
where Q1 is the inlet flow rate, T1 is the inlet temperature, Q2 is the flow rate for each residence time, and T2 is the reactor temperature. The gas concentration was measured in 1 min intervals after 10 min of supplying the gas mixture. For gas measurements, the concentration of exhaust gas from the rear end of the reactor was measured. The measured data are reported as the average value of five measurements performed for each experimental condition. Laboratory ambient conditions were maintained at atmospheric air (298 K, 1 atm).  Table 1 shows the composition of the gas mixture used in the N 2 O thermal decomposition experiment [27]. The gas mixture was composed of 94.96% Ar, 5% O 2 , and 400 ppm of N 2 O to simulate the exhaust gas in a selective non-catalytic reduction process. The actual combustion exhaust gas is N 2 and CO 2 atmosphere. However, in this experiment, an Ar atmosphere was employed to minimize the involvement of nitrogen components from sources, in addition to the N 2 generated by thermal decomposition of N 2 O. Moreover, the molecular weights of N 2 O and CO 2 were the same, and hence measurement errors occurred in the NDIR measurement. Therefore, no hydrocarbon-based material was added. The chemical reaction time of NOx and N 2 O depends on the residence time and temperature. Accordingly, the residence time of the gas mixture in the thermal reactor was selected as an experimental variable. The dependence of the N 2 O thermal decomposition on the residence time was investigated by fixing the composition of the gas mixture and varying the residence time (times of 10, 20, and 40 s were employed). The gas mixture flow rate corresponding to each residence time was calculated from the equation of the reactor volume and residence time (Equation (1)). In this equation, Q is the gas mixture flow rate, D is the inner diameter of the reactor, L is the length of the reactor, and t is the residence time. Table 2 shows the experimental conditions [27]. The experimental temperature was increased in 20 K intervals for temperatures ranging from 1023 to 1223 K, which is the general operating temperature of the SNCR process. In order to maintain the residence time of the gas mixture on each temperature section, the flow rate was adjusted based on Charles's law, which is given as follows:

Experimental Conditons
where Q 1 is the inlet flow rate, T 1 is the inlet temperature, Q 2 is the flow rate for each residence time, and T 2 is the reactor temperature. The gas concentration was measured in 1 min intervals after 10 min of supplying the gas mixture. For gas measurements, the con-centration of exhaust gas from the rear end of the reactor was measured. The measured data are reported as the average value of five measurements performed for each experimental condition. Laboratory ambient conditions were maintained at atmospheric air (298 K, 1 atm).

CHEMKIN Calculation Conditon
CHEMKIN-PRO [28] calculations were performed by simulating the thermal reactor used in the N 2 O thermal decomposition experiment. The 1D reaction structure analysis plug flow reactor model, a cylindrical reactor model that assumes a steady state during flow, was used for the calculations. The input gas was assumed to be perfectly mixed, and continuous chemical reactions inside the reactor were observed. The model of a reaction furnace with an 140 mm inner diameter and a 1000 mm length was built, which was the same as the actual experimental apparatus. The temperature of the thermal reactor was divided into 11 intervals, ranging from 1023 to 1223 K. The residence time of the gas mixture was calculated in the same manner as the experimental conditions at 10, 20, and 40 s. Furthermore, the N 2 O chemical reaction (see Table 3) was calculated from the chemical reaction equation included in GRI 3.0 [29], and the chemical reaction route was confirmed via reaction path analysis.  Figure 4 shows the gas concentration calculated using CHEMKIN-PRO. Calculation conditions of residence time: 1 s, gas composition: 99.96% Ar, 400 ppm N 2 O, and reactor temperature: 1020-1220 K, were employed. The results confirmed that the concentration of N 2 O in the reactor generally decreased with increasing temperature. In the initial temperature condition (1013 K), the N 2 O concentration was 397 ppm. Concentrations of 375 and 253 ppm occurred at 1113 and 1213 K, respectively. The concentration of N 2 was 2 ppm under the initial temperature conditions. The concentration of N 2 increased from 23 ppm at 1113 K to 138 ppm at 1213 K. This resulted from N 2 O thermally decomposing into two N and one O (see Reaction (1) of Table 3) with increasing reactor temperature. The decomposed two N collided with each other, thereby generating an activation energy, and stable N 2 was formed. NO was not formed under the initial temperature condition of 1013 K, but~8 ppm formed when the temperature increased to 1213 K. This resulted from  (3) and (7) of Table 3). Therefore, with the reduction of N 2 O, a small amount of NO was generated.

Formation of Nitrogen and Nitrogen Monoxide by Thermal Decomposition of Nitrous Oxide
Energies 2021, 14, x FOR PEER REVIEW 6 of 11 and N being converted into NO (see Reactions (3) and (7) of Table 3). Therefore, with the reduction of N2O, a small amount of NO was generated.  Figure 5 shows the dependence of the N2O reduction rate on the residence time in the reactor. The rate is expressed as the ratio of the inlet N2O concentration (400 ppm), and the reduced N2O concentration after the reaction, i.e.,

Dependence of Nitrous Oxide Reduction Rate on the Residence Time
The gas mixture composition was 94.96% Ar, 5% O2, and 400 ppm N2O. Residence times of 10, 20, and 40 s, and reaction temperatures of 1013, 1113, and 1213 K were employed. For each residence time, the N2O reduction rate increased with increasing reaction temperature because both variables (i.e., the time and the temperature) affect the forward reaction of the N2O thermal decomposition process. However, in an actual reactor, increasing both variables infinitely is inefficient. Therefore, determining the relationship between the residence time and the reaction temperature and finding an optimum condition between the two variables are essential.   Figure 5 shows the dependence of the N 2 O reduction rate on the residence time in the reactor. The rate is expressed as the ratio of the inlet N 2 O concentration (400 ppm), and the reduced N 2 O concentration after the reaction, i.e.,

Dependence of Nitrous Oxide Reduction Rate on the Residence Time
The gas mixture composition was 94.96% Ar, 5% O 2 , and 400 ppm N 2 O. Residence times of 10, 20, and 40 s, and reaction temperatures of 1013, 1113, and 1213 K were employed. For each residence time, the N 2 O reduction rate increased with increasing reaction temperature because both variables (i.e., the time and the temperature) affect the forward reaction of the N 2 O thermal decomposition process. However, in an actual reactor, increasing both variables infinitely is inefficient. Therefore, determining the relationship between the residence time and the reaction temperature and finding an optimum condition between the two variables are essential.
Energies 2021, 14, x FOR PEER REVIEW 6 of 11 and N being converted into NO (see Reactions (3) and (7) of Table 3). Therefore, with the reduction of N2O, a small amount of NO was generated.  Figure 5 shows the dependence of the N2O reduction rate on the residence time in the reactor. The rate is expressed as the ratio of the inlet N2O concentration (400 ppm), and the reduced N2O concentration after the reaction, i.e.,

Dependence of Nitrous Oxide Reduction Rate on the Residence Time
The gas mixture composition was 94.96% Ar, 5% O2, and 400 ppm N2O. Residence times of 10, 20, and 40 s, and reaction temperatures of 1013, 1113, and 1213 K were employed. For each residence time, the N2O reduction rate increased with increasing reaction temperature because both variables (i.e., the time and the temperature) affect the forward reaction of the N2O thermal decomposition process. However, in an actual reactor, increasing both variables infinitely is inefficient. Therefore, determining the relationship between the residence time and the reaction temperature and finding an optimum condition between the two variables are essential.    Table 4 shows the N 2 O reduction rate in each temperature section of the reactor based on the change in residence time shown in Figure 5. In residence time change Section 1, the change in percentage of the reduction rate increased linearly from 9.75% to 21.25% with increasing temperature in the reactor. However, the change in percentage decreased from 27.5% to 15.5% when the temperature increased from 1113 to 1213 K in change Section 2.  Figures 6 and 7 show the calculation results obtained from the reaction path analyzer. These results explain the trends observed in Figure 5 and Table 4. The reaction rate of conversion from N 2 O to N 2 based on the residence time was calculated with CHEMKIN-PRO. Table 5 shows the specific reaction rate (rate of progress) values plotted in Figure 6. on the change in residence time shown in Figure 5. In residence time change Section 1, the change in percentage of the reduction rate increased linearly from 9.75% to 21.25% with increasing temperature in the reactor. However, the change in percentage decreased from 27.5% to 15.5% when the temperature increased from 1113 to 1213 K in change Section 2.  Figures 6 and 7 show the calculation results obtained from the reaction path analyzer. These results explain the trends observed in Figure 5 and Table 4. The reaction rate of conversion from N2O to N2 based on the residence time was calculated with CHEMKIN-PRO. Table 5 shows the specific reaction rate (rate of progress) values plotted in Figure 6.
As shown in Figure 6, the reaction rate of the N2O decomposition decreased with increasing residence time (see Reactions (1) and (2) of Table 3). This resulted from the chemical reaction rate being related to the concentration of the reactants. The reaction time of the N2O introduced in the reactor increased with increasing residence time, and hence the concentration of exhausted N2O after the reaction decreased. When the N2O concentration decreased, the number of molecules per unit volume in the reactor and, consequently, the collisions between these molecules decreased. Therefore, the number of chemical reactions and the reaction rate were reduced, leading to an increase in the N2O reduction rate with increasing residence time and temperature, as shown in Figure 5.     As shown in Figure 6, the reaction rate of the N 2 O decomposition decreased with in-creasing residence time (see Reactions (1) and (2) of Table 3). This resulted from the chemical reaction rate being related to the concentration of the reactants. The reaction time of the N 2 O introduced in the reactor increased with increasing residence time, and hence the concentration of exhausted N 2 O after the reaction decreased. When the N 2 O concentration decreased, the number of molecules per unit volume in the reactor and, consequently, the collisions between these molecules decreased. Therefore, the number of chemical reactions and the reaction rate were reduced, leading to an increase in the N 2 O reduction rate with increasing residence time and temperature, as shown in Figure 5. Figure 7 shows that the forward reaction rate decreased with increasing residence time (see Reaction (1) of Table 3). In addition, the inverse reaction rate, corresponding to the formation of the N 2 O, appeared with increasing residence time at 1213 K. Accordingly, the N 2 O reduction rate decreased significantly since N 2 and O recombined to form N 2 O again. This observation was consistent with the previous experimental result. Figure 8 shows the dependence of the NO concentration on the residence time. The gas composition and the experimental conditions are the same as those shown in Figure 5. The results confirmed that for a given residence time, the amount of NO formed increased with increasing temperature. This resulted from NO forming via thermal decomposition of N 2 O (see Reaction (3) in Table 3). At a reactor temperature of 1013 K, only a small amount of NO formed at residence times of 10 and 20 s, but the amount increased at 40 s. Increasing the residence time at a reactor temperature of 1113 K yielded only a slight increase in the NO concentration. However, at a reactor temperature of 1213 K, the NO concentration decreased with increasing residence time.  Figures 9 and 10 show the calculated reaction rates of conversion from N2O to NO as determined by the CHEMKIN-PRO reaction path analyzer. The calculation conditions were the same as those employed in the experiment. The specific reaction rate values are shown in Table 6.   Table 6. At reaction temperatures of 1013 and 1113 K, the reaction rate decreased with increasing residence time. Times of 10 and 20 s yielded only slight differences in the reaction rate, but a time of 40 s led to a significant decrease in the reaction rate. Therefore, the concentration of NO at a residence time of 40 s was higher than that at 10 and 20 s, indicating that more N2O was converted into NO. This result corresponds to the change in NO concentration shown in Figure 8, where the experimental results are presented.  As shown in Figure 10, the overall reaction rate was further reduced when the temperature increased. This confirmed that the forward reaction rate decreased with increasing residence time, whereas the inverse reaction rate remained the same (see Reaction (3) in Table 3). Therefore, at a temperature of 1213 K, the inverse reaction rate appeared with increasing residence time and NO was converted back to N2O; hence, the NO concentration decreased at a time of 40 s.

Conclusions
In this study, N2O reduction and NO formation were observed through thermal decomposition experiments performed in a high-temperature reactor. The dependence of the chemical reaction on the residence time and temperature was confirmed by comparing the calculated reaction equation and the reaction rate with experimental results. The major findings of this work are summarized as follows: 1. The results confirmed that most of the N2O was converted to N2 and a small amount of NO via thermal decomposition in Ar atmosphere due to the high operating temperature of SNCR. Therefore, the thermal decomposition temperature must be controlled appropriately to prevent the generation of NO. 2. The change in the N2O reduction rate at temperatures of 1013 and 1113 K increased with the residence time, but decreased at 1213 K and a time of 40 s. This resulted from the inverse reaction rate and the regeneration of N2O at 1213 K. Therefore, exaggerated residence times and thermal decomposition temperatures are inefficient in terms of the energy consumption compared to the reduction of N2O.  At reaction temperatures of 1013 and 1113 K, the reaction rate decreased with increasing residence time. Times of 10 and 20 s yielded only slight differences in the reaction rate, but a time of 40 s led to a significant decrease in the reaction rate. Therefore, the con-centration of NO at a residence time of 40 s was higher than that at 10 and 20 s, indicating that more N 2 O was converted into NO. This result corresponds to the change in NO concentration shown in Figure 8, where the experimental results are presented.
As shown in Figure 10, the overall reaction rate was further reduced when the temperature increased. This confirmed that the forward reaction rate decreased with increasing residence time, whereas the inverse reaction rate remained the same (see Reaction (3) in Table 3). Therefore, at a temperature of 1213 K, the inverse reaction rate appeared with increasing residence time and NO was converted back to N 2 O; hence, the NO concentration decreased at a time of 40 s.

Conclusions
In this study, N 2 O reduction and NO formation were observed through thermal decomposition experiments performed in a high-temperature reactor. The dependence of the chemical reaction on the residence time and temperature was confirmed by comparing the calculated reaction equation and the reaction rate with experimental results. The major findings of this work are summarized as follows: 1.
The results confirmed that most of the N 2 O was converted to N 2 and a small amount of NO via thermal decomposition in Ar atmosphere due to the high operating temperature of SNCR. Therefore, the thermal decomposition temperature must be controlled appropriately to prevent the generation of NO.

2.
The change in the N 2 O reduction rate at temperatures of 1013 and 1113 K increased with the residence time, but decreased at 1213 K and a time of 40 s. This resulted from the inverse reaction rate and the regeneration of N 2 O at 1213 K. Therefore, exaggerated residence times and thermal decomposition temperatures are inefficient in terms of the energy consumption compared to the reduction of N 2 O. 3.
The NO concentration increased with residence time at temperatures of 1013 K and 1113 K, but decreased with the time at 1213 K. This resulted from the inverse reaction rate associated with the conversion of NO back to N 2 O. Therefore, the optimal operating conditions for the highest reduction efficiency must be studied.