NO Separation Characteristics in Integrated Electromigration Membrane Reactor

: An integrated electromigration membrane absorption method has been proposed for the separation of NO from simulated mixed gas. The experiments were conducted to investigate the e ﬀ ect of discharge voltage, gas ﬂow rate, inlet concentrations, and absorbents on the NO separation e ﬃ ciency and total mass transfer coe ﬃ cient in the integrated electromigration membrane reactor. The experimental results demonstrated that the NO separation e ﬃ ciency and total mass transfer coe ﬃ cient increased with the increase in the applied discharge voltage of the integrated electromigration membrane reactor. Regardless of discharge or not, the separation e ﬃ ciency of NO continuously decreased with the increase in the gas ﬂow rate and inlet concentration of NO in the experimental process. The total mass transfer coe ﬃ cient of NO increased ﬁrst and then decreased with an increase in the gas ﬂow rate, while it decreased with an increase in NO inlet concentration. Compared with the membrane absorption without discharge voltage under the condition tested, at a discharge voltage of 18kV, the NO separation e ﬃ ciency and the total mass transfer coe ﬃ cient increased by 48.7% and 9.7 times, respectively.


Introduction
With the frequent occurrence of pollution and the aggravation of resource shortage, pollutant control technology has been investigated from pollution removal to the resource utilization of pollutants [1]. The NOx released from fossil fuel combustion causes bad environmental questions, such as acid rain, haze, and photochemical smog, which are harmful to human health and the ecological environment [2,3]. At present, the technology of selective non-catalytic reduction denitrification (SNCR) and selective catalytic reduction denitrification (SCR), which reduces NOx emission, shows a high removal efficiency. However, they also are accompanied by some technical difficulties, such as large equipment, air preheater clog, trashy catalyst disposal, and so on [4][5][6]. Thus, it has been important to develop new denitrification technologies or improve the existing technology for stricter emission standards [3,7].
The membrane gas separation method, which takes the membrane fiber medium as the separation interface, is a new type of gas absorption process with good application prospects. It has the advantages of low-energy consumption, compact structure, simple operation, and strong selectivity, and effectively overcomes the disadvantages of flooding, channeling, and foaming in traditional wet absorption equipment by non-contact between gas and liquid [8]. In recent years, a series of experiments of membrane absorption have been carried out to separate the acid gases of hydrogen sulfide (H 2 S), SO 2 , NOx. For example, Qi [9,10] studied the mass transfer process from the gas phase to liquid phase for H 2 S, SO 2 , NH 3 , and CO 2 in a membrane absorption reactor and measured the mass transfer 2 of 10 coefficients of some gases. Rami et al. [11] studied the removal efficiency of hydrogen sulfide in natural gas by the membrane absorption method under high pressure. Sun et al. [12] researched the feasibility of SO 2 removal by a hollow fiber membrane using seawater as an absorbent, and studied the effect of related parameters on its mass transfer coefficient. Zhang et al. [13] analyzed the effect of some parameters on the SO 2 removal efficiency in a membrane contactor. Kartohardjono et al. [14] and Wang et al. [15] carried out an investigation on the characteristics of NOx removal by membrane absorption. Among these studies, the transfer of gas molecules through a membrane mainly depends on the concentration difference and the pressure difference of gas in the two sides of the membrane [16], which is accompanied by high gas driving force and large unit gas membrane area for achieving a high gas separation efficiency [8].
Electromigration, which uses electric field force as the driving force, is an effective separation method to concentrate or separate specific components from the mixture. The method of using electric field force as the driving force has been widely used in industrial production and has shown an excellent technical and economic performance in applications such as electrodialysis, electro-ultrafiltration [17], and electrostatic precipitators [18]. However, few studies have reported enhancing the gas separation efficiency by electromigration in membrane absorption. In view of the above, this paper proposed an integrated electromigration membrane separation process which organically combined plasma technology, electromigration, membrane separation, and chemical absorption. It analyzed the separation mechanism of NO in an integrated electromigration membrane separation reactor. The effect of discharge voltage, gas flow rate, inlet concentration, and absorbents on the separation efficiency and total mass transfer coefficient of NO were discussed. It is hoped that this research can provide new ideas and references for the development of gas separation technology.

Experimental Methods
The schematic diagram of the experimental process is shown in Figure 1. The experimental apparatus mainly consisted of a gas distribution unit, an integrated electromigration membrane separation reactor, and a gas measurement system. The 230 × 50 × 60 mm rectangular reactor was made of polymethyl methaerylate (PMMA).
Appl. Sci. 2020, 10, x FOR PEER REVIEW 3 of 11 Figure 1. The schematic diagram of the integrated electromigration membrane reactor for NO separation.

Process Description and Calculation
The separation process of NO in the integrated electromigration membrane reactor can be divided into three steps: (1) Formation of NO negative ions:  Discharge needle electrodes were placed evenly in the upper wall of the reactor and connected with an external negative high-voltage DC power supply. The grounding electrode made of a stainless steel plate was placed on the internal bottom of the reactor, which was filled with aqueous absorbent. A hydrophobicmicroporous membrane consisted of PVDF (polyvinylidene fluoride) and PTFE (Poly tetra fluoroethylene) with a pore size of 0.22 µm and the thickness of 150 µm was installed between the needle electrodes and the grounding electrode, parallel to the grounding electrode and dividing the reactor into two parts. A mixture gas of N 2 -NO provided by compressed steel cylinders went through the space between the needle electrodes and the membrane, while the aqueous absorbent continuously flowed through the space between the membrane and the grounding electrode at a constant flux of 65 mL·min −1 . The treated gas was exhausted through a gas absorbing device. The inlet and outlet concentrations of NO were measured by a flue gas analyzer.

Process Description and Calculation
The separation process of NO in the integrated electromigration membrane reactor can be divided into three steps: (1) Formation of NO negative ions: The N and O atoms in the NO molecule are hybridized with sp orbitals to form a бbond, a π bond, and a large two-center three-electron π 3 2 bond which contains an unpaired electron. In the reactor, NO gas molecule collides with low-energy electrons are generated by gas discharge, capturing electrons to form NO negative ions, which can be briefly described as follows [19,20]: (2) Separation of NO from simulated mixture gas: Under the synergistic action of the concentration gradient and electric field force, the NO negative ions drift towards the interface of the gas membrane from the simulated mixture gas.
Because the membrane is set up in the electric field, the NO negative ions continue to migrate through the membrane pores into the absorbent liquid side of the membrane due to the grounding electrode, which is immersed in the absorption solution. The total mass transfer flux (N) in the process can be expressed as follows [21]: where C is the concentration of NO negative ions in the surface membrane; v is the migration velocity of negative ions in an electric field, v = µE, m·s −1 ; µ is the ion migration velocity at the unit electric-field intensity, m 2 V −1 s −1 ; D is the gas diffusion coefficient, m 2 ·s −1 . In the coexistence process of gas diffusion and electromigration, according to Einstein's equation, the relationship of the gas diffusion coefficient and its electromigration rate can be expressed as follows [22]: where k is the Boltzmann constant, 1.38 × 10 −23 J/K; T is experimental temperature, K; e is the charge amount of an electron, 1.6 × 10 −19 C. In T < 25 • C, it can be concluded that the electromigration rate of an NO negative ion (µ) is about 39 times higher than its own diffusion coefficient (D). Thus, the effect of molecule diffusion on the NO separation efficiency is negligible. Therefore, the separation of NO in the reactor is mainly depended on the electromigration movement.
(3) Absorption of NO in the absorption solution: NO negative ions through the membrane are absorbed by aqueous solutions, such as NaClO 2 or KMnO 4 /NaOH aqueous solution. The absorption reactions for NO removal can be briefly expressed as follows in the experimental process [23,24]: In the study, the separation efficiency and the total mass transfer coefficient are used to evaluate the performance of the integrated electromigration membrane separation reactor on the NO separation from mixture gas, and are calculated by the following formulas [25,26] where η is the separation efficiency of NO, %; K is the total mass transfer coefficient of NO, m·s −1 ; Q is the gas flow rate, L·min −1 ; C in and C out are the inlet and outlet concentration of NO in the reactor, mg·m −3 ; A is the membrane surface area, m 2 .

Effect of Discharge Voltage on Separation Efficiency and Total Mass Transfer Coefficient
The effect of discharge voltage on the NO separation efficiency and total mass transfer coefficient is shown in Figures 2 and 3. It can be seen from Figure 2 that the NO separation efficiency was related to the discharge voltage. When the discharge voltage increased from 8 to 18 kV, the NO separation efficiency first slightly rose and then rapidly increased. At an applied voltage of 18 kV, the separation efficiency was about 57%, which improved by 48.7% compared with the one without discharge. The formation of NO negative ions is related to the electron concentration in the reaction area [27] (Hajime Tamon 1996). When the discharge voltage was lower than 8 kV, the efficiency increment was lower than 1.05% compared with the efficiency without discharge, where few NO negative ions were formed due to the low electron concentration in the reaction space. While the discharge voltage exceeded the corona onset voltage, the electron amounts increased rapidly due to electron avalanche and prompted the formation of NO negative ions. Therefore, the NO separation efficiency increases rapidly with the increase in the discharge voltage [28]. When the discharge voltage was higher than 18 kV, the electric field was broken down. Therefore, the maximum discharge voltage was set at 18 kV. Similarly, the total mass transfer coefficient of NO increases with the increase in the discharge voltage ( Figure 3). When the discharge voltage increases from 0 to 18 kV, the total mass transfer coefficient of NO increases from 0.91 × 10 −4 to 8.81 × 10 −4 m·s −1 ; the total mass transfer coefficient at discharge voltage of 18 kV is about 9.7 times than one without applied voltage. This may be because a high discharge voltage caused an increase in the amount and electromigration velocity of the NO negative ions, which not only prompted NO absorption but also reduced the mass transfer resistance in the gas phase. coefficient of NO increases from 0.91 × 10 −4 to 8.81 × 10 −4 m·s −1 ; the total mass transfer coefficient at discharge voltage of 18 kV is about 9.7 times than one without applied voltage. This may be because a high discharge voltage caused an increase in the amount and electromigration velocity of the NO negative ions, which not only prompted NO absorption but also reduced the mass transfer resistance in the gas phase.   coefficient of NO increases from 0.91 × 10 −4 to 8.81 × 10 −4 m·s −1 ; the total mass transfer coefficient at discharge voltage of 18 kV is about 9.7 times than one without applied voltage. This may be because a high discharge voltage caused an increase in the amount and electromigration velocity of the NO negative ions, which not only prompted NO absorption but also reduced the mass transfer resistance in the gas phase.

Effect of Gas Flow Rate on Separation Efficiency and Total Mass Transfer Coefficient
The effect of the gas flow rate on the NO separation efficiency and total mass transfer coefficient with discharge and non-discharge is demonstrated in Figures 4 and 5. It can be seen that the NO separation efficiency decreased regardless of whether there was discharge or not when the gas flow rate increased from 0.5 to 0.9 L·min −1 . This tendency was consistent with the Wang et al. [15] research result of absorbing NO in a membrane reactor. It also can be seen from Figure 4 that the decreasing trend in separation efficiency during discharge was more pronounced than the trend without discharge, when the gas flow rate increased. When the gas flow rate increased from 0.5 to 0.9 L·min −1 , the NO separation efficiency at 0kV discharge voltage decreased from 8.37% to 6.7%, and its reduction value was about 1.67%, while the one at 18 kV discharge voltage decreased from 57.1% to 44.0%, and its reduction value was about 13%. This may be because that NO separation mainly owed to the NO electromigration movement, and the increase in the gas flow rate not only reduced the residence time of NO gas in the reactor but also reduced the formation probability and proportion of the NO negative ions under the discharge condition, so that the NO separation efficiency reduces significantly when the gas flow rate increases. Unlike separation efficiency, the total mass transfer coefficient of NO increased first and then decreased as the gas flow rate increased at an 18 kV discharge voltage, and its maximum value was 11.6 × 10 −4 m·s −1 at the given gas flow rate of 0.8 L·min −1 ( Figure 5). Meanwhile, it had no significantly change with the increase in the gas flow rate under the non-discharge condition. This may be because the increase in the gas flow rate reduces the mass transfer resistance of the gas-phase, as well as reducing the residence time of NO gas in the reactor. The former improved the total mass transfer coefficient based on the resistance-in-serials equation of the membrane process, while the latter reduced the NO separation efficiency and mass transfer coefficient, as can be seen from Equation (4). the NO separation efficiency at 0kV discharge voltage decreased from 8.37% to 6.7%, and its reduction value was about 1.67%, while the one at 18 kV discharge voltage decreased from 57.1% to 44.0%, and its reduction value was about 13%. This may be because that NO separation mainly owed to the NO electromigration movement, and the increase in the gas flow rate not only reduced the residence time of NO gas in the reactor but also reduced the formation probability and proportion of the NO negative ions under the discharge condition, so that the NO separation efficiency reduces significantly when the gas flow rate increases. Unlike separation efficiency, the total mass transfer coefficient of NO increased first and then decreased as the gas flow rate increased at an 18 kV discharge voltage, and its maximum value was 11.6 × 10 −4 m·s −1 at the given gas flow rate of 0.8 L·min −1 ( Figure 5). Meanwhile, it had no significantly change with the increase in the gas flow rate under the non-discharge condition. This may be because the increase in the gas flow rate reduces the mass transfer resistance of the gas-phase, as well as reducing the residence time of NO gas in the reactor. The former improved the total mass transfer coefficient based on the resistance-in-serials equation of the membrane process, while the latter reduced the NO separation efficiency and mass transfer coefficient, as can be seen from Equation (4).

Effect of Inlet Concentration on Separation Efficiency and Total Mass Transfer Coefficient
The effect of the NO inlet concentration on the NO separation efficiency and total mass transfer coefficient was shown in Figures 6 and 7, respectively. It can be observed in Figure 6 that the

Effect of Inlet Concentration on Separation Efficiency and Total Mass Transfer Coefficient
The effect of the NO inlet concentration on the NO separation efficiency and total mass transfer coefficient was shown in Figures 6 and 7, respectively. It can be observed in Figure 6 that the separation efficiency decreased with the increase in the NO inlet concentration under the experimental conditions. When the NO inlet concentration increased from 202 to 612 mg·m −3 , the NO separation efficiency decreased from 56.4% to 41.2% at an 18 kV discharge voltage. This is the reason why the increased amount of NO negative ions is less than the increased amount of NO molecules in the reactor with the increase in the NO inlet concentration. Some studies [29,30] have reported a similar changing rule in removing NO using other methods. It can be seen from Figure 7 that the total mass transfer coefficient of NO decreased with the increase in the inlet concentration. It also can be seen in Figure 7 with the increasing of inlet concentration that the total mass transfer coefficient decreased more significantly at a discharge voltage of 18 kV than at a discharge voltage of 12 kV. It may be the reason that the NO separation efficiency decreased as the NO inlet concentration increased (Equations (3) and (4)) [26].

Effect of Inlet Concentration on Separation Efficiency and Total Mass Transfer Coefficient
The effect of the NO inlet concentration on the NO separation efficiency and total mass transfer coefficient was shown in Figures 6 and 7, respectively. It can be observed in Figure 6 that the separation efficiency decreased with the increase in the NO inlet concentration under the experimental conditions. When the NO inlet concentration increased from 202 to 612 mg·m −3 , the NO separation efficiency decreased from 56.4% to 41.2% at an 18 kV discharge voltage. This is the reason why the increased amount of NO negative ions is less than the increased amount of NO molecules in the reactor with the increase in the NO inlet concentration. Some studies [29][30] have reported a similar changing rule in removing NO using other methods. It can be seen from Figure 7 that the total mass transfer coefficient of NO decreased with the increase in the inlet concentration. It also can be seen in Figure 7 with the increasing of inlet concentration that the total mass transfer coefficient decreased more significantly at a discharge voltage of 18 kV than at a discharge voltage of 12 kV. It may be the reason that the NO separation efficiency decreased as the NO inlet concentration increased (Equations (3) and (4)) [26].

Effect of Absorbent on no Separation Efficiency and Total Mass Transfer Coefficient
The effect of two different absorbents, NaClO2 and KMnO4/NaOH, on the separation efficiency

Effect of Absorbent on no Separation Efficiency and Total Mass Transfer Coefficient
The effect of two different absorbents, NaClO 2 and KMnO 4 /NaOH, on the separation efficiency and total mass transfer coefficient is shown in Figures 8 and 9. It can be seen from Figures 8 and 9 that the corona discharge could improve the NO separation efficiency and total transfer coefficient, and the separation efficiency and total transfer coefficient with NaClO 2 as the absorbent were higher than those with KMnO 4 /NaOH as the absorbent in the condition of discharge or no discharge. This may be due to three reasons: (i) the migration of NO negative ions improved the absorption efficiency in the electric field; (ii) the reaction rate of NaClO 2 with NO was higher than that of KMnO 4 /NaOH with NO [23,29,[31][32][33]

Effect of Absorbent on no Separation Efficiency and Total Mass Transfer Coefficient
The effect of two different absorbents, NaClO2 and KMnO4/NaOH, on the separation efficiency and total mass transfer coefficient is shown in Figures 8 and 9. It can be seen from Figures 8 and 9 that the corona discharge could improve the NO separation efficiency and total transfer coefficient, and the separation efficiency and total transfer coefficient with NaClO2 as the absorbent were higher than those with KMnO4/NaOH as the absorbent in the condition of discharge or no discharge. This may be due to three reasons: (i) the migration of NO negative ions improved the absorption efficiency in the electric field; (ii) the reaction rate of NaClO2 with NO was higher than that of KMnO4/NaOH with NO [23,29,[31][32][33]

Conclusions
The following conclusions can be drawn from the experimental research: (1) The integrated electromigration membrane separation method can effectively separate NO from NO-N2 mixture gas. At the discharge voltage of 18kV, the separation efficiency of NO in the reactor was about 57%. It increased by 48.7%, and the total mass transfer coefficient of NO was raised 9.7 times compared with the one in the membrane absorption process without discharge.
(2) The electromigration of NO negative ions can enhance the NO separation and mass transfer. The higher discharge voltage has a greater separation efficiency and total mass transfer

Conclusions
The following conclusions can be drawn from the experimental research: (1) The integrated electromigration membrane separation method can effectively separate NO from NO-N 2 mixture gas. At the discharge voltage of 18kV, the separation efficiency of NO in the reactor