Pulsed high-current discharge in water: adiabatic model of expanding plasma channel and acoustic wave

1.   Introduction

Ammonia is one of the main raw materials for the manufacture of nitrogen fertilizers for plant growth, and also an important chemical raw material [1–3] widely used in agriculture, chemical industry, pharmaceuticals and many other fields. At present, the most common and relatively well-established method for the industrial ammonia synthesis is the Haber-Bosch method [1, 2, 4]. Usually an iron- or molybdenum-based catalyst is used to catalyze the conversion reactions of the feed gases, N₂ and H₂ to ammonia under high temperatures (typically 500 °C–600 °C [5]) and high pressures (typically 200–500 atm). However, the synthesis of ammonia by this method is an energy intensive process. The high temperature conditions mean that a large amount of energy is required to maintain the reaction conditions. At the same time, the high-pressure conditions mean that the reactor and associated equipment need to be able to withstand the high pressure, which adds to the cost and complexity of the equipment.

There are also technologies such as photocatalytic ammonia synthesis [6–8], electrolytic ammonia synthesis [9–12], and plasma ammonia synthesis [13–17]. Photocatalytic ammonia synthesis is a technology that uses the input of light energy to promote the formation of ammonia from nitrogen and hydrogen in the gas phase through the action of a catalyst. In this process, developing materials with efficient photocatalytic properties, such as photocatalysts and photoabsorbers, is particularly important. However, at present, such methods require sufficient light limitation and stable photocatalysts, and therefore are still in the laboratory research stage. Electrocatalytic ammonia synthesis is usually a method of synthesizing ammonia gas from N₂ and water in the liquid phase by inducing the synthesis of ammonia gas through electrical energy, mainly through two pathways: the nitrogen reduction pathway and the nitrogen reduction pathway (also called the nitrogen oxidation pathway). In the nitrogen reduction pathway, nitrogen molecules are adsorbed on the surface of the catalyst under the action of a catalyst, and by applying an electric current or potential, the nitrogen accepts electrons provided by the catalyst to form negative ions (N⁻), the nitrogen negative ions accept more electrons to further reduce to nitrogen hydroxide ions (${\mathrm{NH}}^-_ 2$), which are eventually converted into ammonia molecules. The nitrogen molecule in nitrogen gas accepts electrons to undergo a reduction reaction to form ammonia. In this pathway, the solubility of nitrogen in the electrolyte and the catalyst’s ability to adsorb nitrogen are both relatively low leading to inefficiency. In addition, the hydrogen reduction reaction coexisting in the solution competes with the nitrogen reduction reaction for electrons. It has been reported in the reference [10] that breaking the dominance of the competition of hydrogen precipitation reactions is the key to achieve high selectivity and efficiency in ammonia synthesis. In the nitrate reduction pathway, nitrogen is first oxidized to nitrogen oxides and then ammonia is formed through further reactions. This pathway also requires a higher potential due to the fact that the nitrogen oxidation reaction itself requires more energy. The advantage of this pathway is that it solves the problem of low solubility and adsorption capacity. The reference [18] reported the synthesis of ammonia through the electroreduction of nitric acid, catalyzed by strained ruthenium nanoclusters, at a maximum rate of 5.56 ${\mathrm{mol}}\cdot {\mathrm{g}}_{\mathrm{cat}}^{-1}\cdot {\mathrm{h}}^{-1}$. However, the nitrate reduction reaction is a cumbersome process involving 8 electrons and 10 protons [19], and the consumption of electricity in this pathway is a problem when it comes from non-renewable energy sources, as well as the challenge of maintaining the activity and stability of the catalyst over a long period of time.

Therefore, the direct synthesis of ammonia from N₂ and H₂ in a single reactor is a simpler and more economical method for the process described above. The development of a one-step ammonia synthesis is necessary to simplify the process, improve energy efficency (EE), reduce environmental impact, and increase flexibility in ammonia production and supply. Plasma synthesis of ammonia is a method of synthesizing ammonia using a high-temperature plasma reaction system. During plasma synthesis of ammonia, electrons in hydrogen molecules can be excited by high-energy electrons through collisions. Similarly, it can cause the electrons in the nitrogen molecule to be excited to a higher electronic state, so that the activated nitrogen molecule reacts with the hydrogen molecule to form an ammonia molecule. Compared to the conventional Haber-Bosch ammonia synthesis process, plasma ammonia synthesis can be carried out at lower ambient temperatures and pressures, which significantly reduces energy consumption and operating costs, and plasma ammonia synthesis technology has on-and-off controllability. Therefore, the plasma technology has many potential advantages in developing ammonia synthesis.

Currently, many researchers aim to find the most effective method to realize ammonia synthesis by exploring a number of different power drive forms, electrode structures, and discharge forms. Meanwhile, the synergistic effect between catalyst and plasma which provides new possibilities for plasma ammonia synthesis is also studied. In terms of the form of power supply, different types of power supplies, such as AC power [2] supply and radio frequency (RF) power [20] supply, are mostly used to realize the generation and maintenance of plasma in ammonia synthesis. The design of the electrode structure directly affects the plasma formation and discharge characteristics, and is one of the keys to improve the efficiency of ammonia synthesis. Up to now, different forms of discharges, such as glow discharges [21], arc discharges [22, 23], and dielectric barrier discharges (DBD) [1, 3, 6, 24], are explored to achieve efficient ammonia synthesis. These discharges can provide energy and stimulate the activity of gas molecules to promote the ammonia synthesis reaction. In addition, Zhou et al [3] increased the synthesis of ammonia by adding gas to the reactor. However, in order to industrially realize plasma ammonia synthesis at high flow rates and operate at ambient temperature and pressure, lower cost and faster processing discharge methods need to be developed. Therefore, microwave discharge plasma synthesis of ammonia has become the focus of researchers’ attention. As illustrated in table 1, a review of the extant literature on the subject of ammonia synthesis by atmospheric pressure plasma reveals a clear challenge in this field. It is evident that, in the absence of a catalyst, ammonia synthesis at ambient temperature and pressure is a difficult process to achieve.

Table  1.  Studies of ammonia synthesis by atmospheric pressure plasma.

Plasma source	Plasma parameters	Catalyst	V(N2):V(H2)	Temperature (K)	Pressure (Pa)	Highest NH3 yield
DBD	Pulse power, 5 kHz	MgO	0.8	Room temperature	1.01×105	3.7 g·m−3
DBD	AC, 3 kV, 500 Hz	Fe	1/3	323	1.01×105	0.9 g·(kW·h)−1
DBD	Pulse power, 10 kHz	/	N2/CH4 = 1/3	295	1.01×105	1.88 mmol·min−1
DBD	AC, 13 kHz	Ru/Al2O3	0.14%H2O+40%H2+N2	Room temperature	1.01×105	0.68 g·(kW·h)−1
Glow	AC, 200 W, 60 Hz	Pt	1/3	Room temperature	1.01×105	1.46 mmol·h−1
Microwave	180 W, 2.45 GHz	Fe	/	573	400	1.5 mmol·g−1
Pulse modulated-microwave (this work)	84.42 W, 2.45 GHz	/	7/1	Room temperature	1.01×105	2.93 μmol·min−1

 | Show Table

DownLoad: CSV

Microwave discharge shows certain advantages in several aspects. First of all, the gas flow rate is significantly increased, with the characteristics of instant on and instant off, which is more suitable for industrialized applications. Moreover, the high electron density in the discharge process will produce a large number of active species in the vibrational excited state, thus promoting the reaction. It is worth noting that the use of pulse modulation of microwave power can effectively slow down the gas heating phenomenon, and further enhance the energy efficiency. In summary, plasma by pulse-modulated microwave has the significant advantages as a potential green alternative because of its mild reaction conditions (no need for high temperature and high pressure), direct one-step synthesis without catalysts, flexible operation with on-and-off operation, and no emission of carbon dioxide greenhouse gases from the traditional industries, and so on. Parameters such as pulse modulation frequency and duty cycle will have a significant effect on the discharge characteristics, thus becoming an important factor affecting the effect of ammonia synthesis. Therefore, this form of discharge has significant applicability in ammonia synthesis research, and it is necessary to further study these parameters in depth in order to fully understand their effects on the ammonia synthesis process.

Therefore, in this work, we explored a one-step direct method for NH₃ synthesis employing pulse modulated microwave plasma technology at atmospheric pressure. The effects of different discharge parameters including gas flow rate, microwave absorbed power, pulse modulation frequency and pulse duty cycle on ammonia synthesis were investigated, and the mechanism of synthesis was studied.

2.   Experimental setup and measurements

2.1   Experimental setup

This experimental setup consists of an input section, a reactor, an optical diagnostic section, and a product detection section, as shown in figure 1. The input section mainly consists of instruments such as a home-made pulse-modulated microwave power supply, a raw gas supply, and mass flow controllers. A pulse modulated microwave power supply provides the energy input for the experiments with a microwave frequency between 2.4 GHz and 2.5 GHz and a microwave output power between 0 and 200 W. Meanwhile, the pulse frequency and duty cycle of power supply can be adjusted in the ranges of 10–200 kHz and 0.1%–100%, respectively. During the experiment, N₂ (99.999% purity) and H₂ (99.999% purity) are controlled by the mass flow controllers before flowing into the reactor. The reactor is a λ/4 type coaxial resonance chamber, whose basic components include an aluminum alloy shell, a brass stub port, an adjustable-length brass needle, a chamber formed by two brass connectors, a sub-miniature A (SMA) RF connector, a quartz tube, and a coaxial transmission line. The schematic cross-section of the reactor used in this study is shown in figure 2, which utilizes a brass chamber with a diameter of 26 mm and the length of 31 mm (about 1/4 of the wavelength). The resonance chamber structure is designed with an inclined spinning gas design to ensure that the vortex can be formed inside the brass chamber. The brass parts are connected to an SMA RF connector with five 2-mm-diameter air inlets oriented at an angle of 45° to the axis. N₂ enters the aluminum alloy housing through the brass stub ports on the aluminum alloy housing and enters the resonance chamber through the inlet ports on the brass piece. A brass pin with a diameter of 2 mm and a length of 31 mm is threaded into the brass connector and can be adjusted back and forth to control the distance of the tip from the bottom of the connector. The gas forms a vortex around the brass needle inside the resonance chamber and is ejected through the tip of the needle into the quartz tube. At the same time, microwave power is fed to the needle electrode through the SMA RF interface and coupled into the plasma. The reaction is carried out in a quartz tube and the product is collected at the end of the quartz tube. The inner diameter, the outer diameter, and the length of the quartz tube were 26 mm, 28 mm, and 12 cm, respectively, and the inner diameter, the outer diameter, and the length of the stub port were 4 mm, 6 mm, and 20 mm, respectively. H₂ is passed through the stub port below the quartz tube to react with a jet of nitrogen injected from the tip of the needle for the synthesis of ammonia. The optical diagnostic section includes a spectrometer (Andor SR-750i), an ICCD (Andor New iStar DH334T), and a fiber optic probe. The fiber optic probe is placed at 30° from the reactor axis for emission spectrum acquisition, and the exposure time of ICCD is set to 0.5 s.

Figure  1.  Schematic diagram of the experimental setup.

DownLoad: Full-Size Img PowerPoint

Figure  2.  λ/4 type coaxial resonant cavity reactor cross-section (a) and physical picture (b).

DownLoad: Full-Size Img PowerPoint

2.2   Product inspection

Product detection relies heavily on a Ultraviolet (UV) visible spectrophotometer (Cary 300, Agilent) for analysis. In the experiment, the gas product is introduced into two identical gas absorption bottles which contain 20 mL H₂SO₄ solution with the concentration of 0.005 mol·L⁻¹. The first absorption bottle is used to absorb the synthesized ammonia while the second absorption bottle is used to test whether the ammonia in the tail gas is completely absorbed or not. To ensure the accuracy of the experimental measurements, the following procedure was employed. After setting the experimental conditions, the gas was pretreated as follows. During the first three minutes before the discharge, the gas in the system was completely evacuated. From the fourth minute of discharge, the reaction gases were continuously introduced into two absorption bottles until the sixth minute, with a total collection time of three minutes. Immediately after the collection, 5 mL of the absorbent solution was sampled using a disposable pipette, followed by the addition of Nessler’s reagent to initiate a colorimetric reaction. The concentration of ${{\mathrm{N}\mathrm{H}}_{4}^{+}}$ in the solution was measured using a UV-visible spectrophotometer, and the ammonia content was calculated accordingly. To ensure the reliability of the data, each experimental condition was repeated three times.

2.3   Calculation method

The microwave absorbed power is calculated by the difference between the microwave incident power and the microwave reflected power.

Here, ${P}_{\mathrm{a}\mathrm{b}}$, ${P}_{\mathrm{i}\mathrm{n}}$ and ${P}_{\mathrm{r}\mathrm{e}}$ are microwave absorbed power, microwave incident power and microwave reflected power, respectively.

Ammonia synthesis rate is an important indicator to assess the effectiveness of ammonia synthesis, which can be calculated by equation (2) [25], in addition, nitrogen conversion [25] and EE [6] can also assess the extent of the reaction to a certain extent, which are calculated by equations (3) and (4), respectively:

in which, ${r}_{\left({\mathrm{N}\mathrm{H}}_{3}\right)}$, ${c}_{\left({\mathrm{N}\mathrm{H}}_{4}^{+}\right)}$, V and t are the rate of NH₃ production, the molar concentration of ${{\mathrm{N}\mathrm{H}}_{4}^{+}}$, the volume of the absorbing solution and the discharge time, respectively.

where ${X}_{\left({\text{N}}_{\text{2}}\right)}$ and ${\nu }_{\left({\text{N}}_{\text{2}}\right)}$ are the conversion rate of nitrogen and the flow rate of nitrogen, respectively.

3.   Results and discussion

3.1   Mechanism of ammonia synthesis

The formula for the synthesis of ammonia from nitrogen and hydrogen is:

Hong et al [26] showed that NH radicals are an important trigger in the ammonia synthesis pathway as well as illustrated that their generation is mainly controlled by electronic reactions and that electron dynamics has a strong influence on the ammonia synthesis process. Li et al [27] similarly found that a large number of NH radicals are generated in the plasma torch during the preparation of N₂-H₂ microwave plasma torch. In addition, Nakajima and Sekiguchi [25] specifically emphasized that N(²D) is the key substance for the generation of NH radicals. Therefore, a large number of researches have shown that NH radicals are important intermediates during the ammonia synthesis pathway [25, 26, 28–32].

There are four main possible pathways (L1‒L4) for the generation of NH radicals in plasma, as shown in figure 3. In addition, the excited H atoms and N₂ also collide to generate NH radicals [25]. Since the dissociation energy of nitrogen molecular N₂ (9.8 eV) is much larger than hydrogen molecule H₂ (4.5 eV) (shown in table 2), this reaction is negligible, so this pathway is generally not mentioned. Among the four pathways for the generation of NH radicals, the excited nitrogen molecular ${\mathrm{N}}^*_2$ is necessary, so the generation of a large amount of excited nitrogen molecular ${\mathrm{N}}^*_2$ is the key to the synthesis of the intermediate NH radicals. A large number of high-energy electrons exist in the microwave plasma, and these high-energy electrons collide with nitrogen molecular N₂ to generate a large number of excited nitrogen molecular ${\mathrm{N}}^*_2$ after obtaining energy from the electromagnetic field (R9). Excited nitrogen molecular ${\mathrm{N}}^*_2$ will collide with H atoms to produce NH radicals (R19‒R20) [26]. In addition, the collision of high-energy electrons with H₂ also produces a large number of excited hydrogen molecule ${\mathrm{H}}^*_2$ (R10), and the chemical reaction between nitrogen atom N and excited hydrogen molecule ${\mathrm{H}}^*_2$ is also an important pathway for the production of NH radicals, which consumes about 1.4 eV of energy (R7) [28].

Figure  3.  Pathway diagram of the reaction between N₂ and H₂ to form NH₃.

DownLoad: Full-Size Img PowerPoint

Table  2.  Important reaction processes and corresponding activation energies in the ammonia synthesis process [28].

No.	Process	Reaction	Activation energy (eV)
R1	Ionization	$\mathrm{e}+{\text{N}}_{\text{2}}\to {\text{N}}_{\text{2}}^{\text{+}}+\text{2}\mathrm{e}$	15.6
R2	Ionization	$\mathrm{e}+{\text{H}}_{\text{2}}\to {\text{H}}_{\text{2}}^{\text{+}}+\text{2}\mathrm{e}$	15.4
R3	Electron impact dissociation	$\mathrm{e}+{\text{N}}_{\text{2}}\to \text{2N}+\mathrm{e}$	13.0
R4	Bond dissociation	${\text{N}}_{\text{2}}\to \text{2N}$	9.8
R5	Electronic excitation	${\text{N}}_{\text{2}}\left(\text{X}\right)\to {\text{N}}_{\text{2}}\left(\text{A3}\right)$	6.2
R6	Electron impact dissociation	$\mathrm{e}+{\text{H}}_{\text{2}}\to \text{2H}+\mathrm{e}$	4.5
R7	Chemical reaction	$\text{N}+{\text{H}}_{\text{2}}^{\text{*}}\to \text{H}+\text{NH}$	1.4
R8	Vibrational excitation	$\text{N}_{\text{2}}\left(\text{X}\right)\to\text{N}_{\text{2}}\left(\text{X}\text{,}\ v\ \text{=}\mathrm{\ 1}\right)$	0.29

 | Show Table

DownLoad: CSV

Similarly, there are four possible pathways (L5‒L8) for the generation of NH₃ in the plasma. NH₃ can be synthesized by NH₂ hydrogenation or obtained by direct hydrogenation of NH radicals, and the reaction processes involved are listed in table 3. In this case, the reaction rate of N(²D) with NH₃ (1.1×10⁻¹⁰ cm³·s⁻¹) (R21) is much higher than that of N(²D) with H₂ (2.3×10⁻¹² cm³·s⁻¹) (R19), which causes the synthesized ammonia to decompose again. The decomposition of NH₃ is not expected during the synthesis of ammonia, and the reaction rates of processes such as R22 and R24 are related to the temperature at the time of reaction [26], so we expect to suppress the occurrence of the reverse reaction by regulating the plasma temperature. Therefore, regulating the vibrational temperature and rotational temperature in the reaction by changing the plasma parameters is very important in the ammonia synthesis process.

Table  3.  Other reaction processes in the ammonia synthesis process.

No.	Process	Rate coefficient (cm3·s−1)	Reference
R9	$\mathrm{e}+{\text{N} }_{\text{2} }\to {\mathrm{e}+\text{N} }_{\text{2} }^{\text{*} } ,$ ${\text{N}}_{\text{2}}^{\text{*}}={\text{N}}_{\text{2}}\left(\text{A3}\right),{\text{N}}_{\text{2}}\left(\text{B3}\right),{\text{N}}_{\text{2}}\left(\text{a'1}\right),{\text{N}}_{\text{2}}\left(\text{C3}\right)$	BOLSIG+ [33]	[26]
R10	$\mathrm{e}+{\text{H} }_{\text{2} }\to \mathrm{e}+{\text{H} }_{\text{2} }^{\text{*} } ,$ ${\text{H}}_{\text{2}}^{\text{*}}={\text{H}}_{\text{2}}\left(\text{B3}\right),{\text{H}}_{\text{2}}\left(\text{B1}\right),{\text{H}}_{\text{2}}\left(\text{C3}\right),{\text{H}}_{\text{2}}\left(\text{A3}\right)$	BOLSIG+ [33]	[26]
R11	$\mathrm{e}+{\text{H}}_{\text{2}}^{\text{+}}\to \text{H}+\text{H}$	$7.51\times {10}^{-9}-1.12\times {10}^{-9}\times {T}_{\mathrm{e}}+1.03\times {10}^{-10}\times {T}_{\mathrm{e}}^{2}-4.15\times {10}^{-12}\times {T}_{\mathrm{e}}^{3}+5.86\times {10}^{-14}\times {T}_{\mathrm{e}}^{4}$	[26]
R12	$\mathrm{e}+{\text{N}}_{\text{2}}^{\text{+}}\to \text{N}+\text{N}$	$9.0\times {10}^{-8}\times {(0.026/{T}_{\mathrm{e}})}^{0.39}$	[26]
R13	${\text{N}}_{\text{2}}^{\text{+}}+{\text{H}}_{\text{2}}\to \text{NH}+\text{product}$	-	[3]
R14	${\text{N}}_{\text{2}}^{\text{+}}+{\text{H}}_{\text{2}}\to {\text{N}}_{\text{2}}{\text{H}}^{\text{+}}+\text{H}$	$2.0\times {10}^{-9}$	[26]
R15	$\text{e}+{\text{N}}_{\text{2}}{\text{H}}^{\text{+}}\to \text{NH}+\text{H}$	-	[25]
R16	${\text{H}}_{\text{2}}^{\text{+}}+{\text{N}}_{\text{2}}\to {\text{N}}_{\text{2}}{\text{H}}^{\text{+}}+\text{H}$	$2.0\times {10}^{-9}$	[26]
R17	${\text{N}}_{\text{2}}^{\text{+}}+\text{2H(s)}\to \text{2NH(s)}+\text{product}$	-	[3]
R18	${\text{N}}_{\text{2}}\left(\text{a'1}\right)+{\text{N}}_{\text{2}}\left(\text{a'1}\right)\to {\text{N}}_{\text{2}}^{\text{+}}+{\text{N}}_{\text{2}}+\mathrm{e}$	$1.0\times {10}^{-11}$	[26]
R19	$\text{N}\left({}_{}{}^{2}\text{D}\right)+{\text{H}}_{\text{2}}\to \text{H}+\text{NH}$	$2.3\times {10}^{-12}$	[26]
R20	$\text{N}\left({}_{}{}^{\text{2}}\text{P}\right)+{\text{H}}_{\text{2}}\to \text{H}+\text{NH}$	$2.5\times {10}^{-14}$	[26]
R21	$\text{N}\left({}_{}{}^{\text{2}}\text{D}\right)+{\text{NH}}_{\text{3}}\to \text{NH}+{\text{NH}}_{\text{2}}$	$1.1\times {10}^{-10}$	[26]
R22	${\text{N}}_{\text{2}}\left(\text{A3}\right)+{\text{H}}_{\text{2}}\to {\text{N}}_{\text{2}}+2\text{H}$	$2.0\times {10}^{-10}\times \mathrm{e}\mathrm{x}\mathrm{p}(-3500/{T}_{\mathrm{g}})$	[26]
R23	$\text{NH}+\text{H}\to {\text{NH}}_{\text{2}}$	-	[29]
R24	$\text{NH}+\text{NH}\to \text{N}+{\text{NH}}_{\text{2}}$	$1.7\times {10}^{-12}\times {({T}_{\mathrm{g}}/300)}^{1.5}$	[26]
R25	$\text{NH}_{\text{2}}+\text{H}+M\to\text{NH}_{\text{3}}+M$ $M\ \text{=}\ \text{N}_{\text{2}}\text{,}\ \text{H}_{\text{2}}$	$1.4\times10^{-32}\mathrm{\ c}\mathrm{m}^6\cdot\mathrm{s}^{-1}$	[26]
R26	$\text{NH}+{\text{NH}}_{\text{2}}\to {\text{NH}}_{\text{3}}+\text{N}$	$1.66\times {10}^{-12}$	[26]
R27	${\text{NH}}_{\text{2}}+{\text{H}}_{2}\to {\text{NH}}_{\text{3}}+\text{H}$	$5.4\times {10}^{-11}\times \mathrm{e}\mathrm{x}\mathrm{p}(-6492/{T}_{\mathrm{g}})$	[26]
R28	$\text{NH}+{\text{H} }_{\text{2} }+M\to {\text{NH} }_{\text{3} }+M ,$ $M\ \text{=}\ \text{N}_{\text{2}}\text{,}\ \text{H}_{\text{2}}$	$6.5\times10^{-38}\times\left(T_{\mathrm{g}}/300\right)\mathrm{e}\mathrm{x}\mathrm{p}\left(\dfrac{1700}{T_{\mathrm{g}}}\right)\ \mathrm{c}\mathrm{m}^6\cdot\mathrm{s}^{-1}$	[26]
R29	${\text{N}}_{\text{2}}\left(\text{A3}\right)+{\text{NH}}_{\text{3}}\to {\text{N}}_{\text{2}}+{\text{NH}}_{\text{3}}$	$1.66\times {10}^{-10}$	[26]
R30	${\text{N}}_{\text{2}}\left(\text{B3}\right)+{\text{H}}_{\text{2}}\to {\text{N}}_{\text{2}}\left(\text{A3}\right)+{\text{H}}_{\text{2}}$	$2.5\times {10}^{-11}$	[26]
R31	${\text{N}}_{\text{2}}\left(\mathrm{a}\text{'1}\right)+\text{H}\to {\text{N}}_{\text{2}}+\text{H}$	$1.5\times {10}^{-11}$	[26]
R32	${\text{N}}_{\text{2}}\left(\mathrm{a}\text{'1}\right)+{\text{H}}_{\text{2}}\to {\text{N}}_{\text{2}}+{\text{H}}_{\text{2}}$	$2.6\times {10}^{-11}$	[26]
R33	$\text{N}+\text{NH}\to \text{H}+{\text{N}}_{2}$	$5\times {10}^{-11}$	[26]
R34	$\text{H}+\text{NH}\to {\text{H}}_{2}+\text{N}$	$5.4\times {10}^{-11}\times \mathrm{e}\mathrm{x}\mathrm{p}(-165/{T}_{\mathrm{g}})$	[26]
R35	$\text{H}+{\text{NH}}_{3}\to {\text{NH}}_{2}+{\text{H}}_{2}$	$8.4\times {10}^{-14}\times {\left({T}_{\mathrm{g}}/300\right)}^{4.1}\mathrm{e}\mathrm{x}\mathrm{p}(-4760/{T}_{\mathrm{g}})$	[26]
R36	${\text{N}}_{\text{2}}\left(\text{B3}\right)+{\text{N}}_{\text{2}}\to {\text{N}}_{\text{2}}+{\text{N}}_{\text{2}}$	$2.0\times {10}^{-12}$	[26]
R37	${\text{N}}_{\text{2}}\left(\mathrm{a}\text{'1}\right)+{\text{N}}_{\text{2}}\to {\text{N}}_{\text{2}}\left(\text{B3}\right)+{\text{N}}_{\text{2}}$	$1.9\times {10}^{-13}$	[26]
R38	${\text{N}}_{\text{2}}\left(\mathrm{a}\text{'1}\right)+{\text{N}}_{\text{2}}\left(\mathrm{a}\text{'1}\right)\to {\text{N}}_{\text{4}}^{\text{+}}+\mathrm{e}$	$1.0\times {10}^{-11}$	[26]
R39	${\text{N}}_{\text{2}}\left(\mathrm{a}\text{'1}\right)+{\text{N}}_{\text{2}}\left(\text{A3}\right)\to {\text{N}}_{\text{4}}^{\text{+}}+\mathrm{e}$	$4.0\times {10}^{-12}$	[26]
R40	${\text{N}}_{\text{2}}\left(\text{A3}\right)+\text{N}\to {\text{N}}_{\text{2}}+\text{N}\left({}_{}{}^{\text{2}}\text{P}\right)$	$4.0\times {10}^{-11}\times {(300/{T}_{\mathrm{g}})}^{2/3}$	[26]
R41	${\text{N}}_{\text{2}}\left(\text{A3}\right)+{\text{N}}_{\text{2}}\left(\text{A3}\right)\to {\text{N}}_{\text{2}}+\text{2N}$	$3.0\times {10}^{-11}$	[26]
R42	${\text{N}}_{\text{2}}\left(\text{A3}\right)+{\text{N}}_{\text{2}}\left(\text{A3}\right)\to {\text{N}}_{\text{2}}+{\text{N}}_{\text{2}}\left(\text{B3}\right)$	$3.0\times {10}^{-10}$	[26]
R43	${\text{N}}_{\text{2}}\left(\text{A3}\right)+{\text{N}}_{\text{2}}\left(\text{A3}\right)\to {\text{N}}_{\text{2}}+{\text{N}}_{\text{2}}\left(\text{C3}\right)$	$1.5\times {10}^{-10}$	[26]

 | Show Table

DownLoad: CSV

In this research, emission spectroscopy is used to diagnose the active species during the discharge process, and emission spectra covering the range of 280‒900 nm were collected under an exposure time of 0.5 s. The results are shown in figure 4. These spectra mainly consisted of bands of N₂ (C³Π_u→B³Π_g, Δν = 2, 1, 0, −1, −2, −3, −4), N₂ (B³Π_g→A³Σ$_{\mathrm{u}}^+$, Δν = 2, 1), ${\mathrm{N}}_{2}^+$ (B²Σ$_{\mathrm{u}}^+$→X²Σ$_{\mathrm{g}}^+$, Δν = 0) and NH radial (336.1 nm) bands. It is experimentally proved that NH radical exists in the ammonia synthesis reaction process. Moreover, from the collected spectra, a large number of excited states exist in the microwave-driven ammonia synthesis system, which creates conditions for the synthesis of NH radical. It is noted in the literature that vibrational excitation is the most energy-efficient activation pathway for ammonia synthesis at mild conditions using plasma, which requires only $0.44\; {\mathrm{kW}}\cdot {\mathrm{h}}\cdot {\mathrm{kg}}^{-1}_{{\mathrm{NH}}_3}$ of energy [28]. Meanwhile, the characteristic parameters in Specair software are continuously adjusted to fit the characteristic temperature of the plasma. By comparing the experimental spectra of N₂ (C³Π_u→B³Π_g, Δν = −2, 380.1 nm) with the calculated fitted spectra, we obtained the vibrational temperature (T_vib) and rotational temperature (T_rot) of the plasma, as shown in figure 5.

Figure  4.  Emission spectrum of atmospheric pressure pulse-modulated microwave plasma during ammonia synthesis process. 280‒900 nm, N₂ flow rate: 7 L·min⁻¹, H₂ flow rate: 1 L·min⁻¹, microwave incident power: 105.86 W, reflected power: 18.77 W, pulse modulation frequency: 25 kHz, pulse duty cycle: 50%.

DownLoad: Full-Size Img PowerPoint

Figure  5.  Experimental and simulated spectra of N₂ (C³Π_u→B³Π_g, Δν = −2, 380.1 nm) are compared. H₂ flow rate: 1 L·min⁻¹, microwave incident power: 105.86 W, reflected power: 18.77 W, pulse modulation frequency: 25 kHz, pulse duty cycle: 50%.

DownLoad: Full-Size Img PowerPoint

3.2   Effect of mode of ventilation

The challenge in this study is that when N₂ and H₂ are mixed and introduced into the resonant cavity for discharge, the discharge bright spot is observed only at the tip, and the plasma jet cannot be successfully generated at the tip. To overcome this challenge, we first introduced pure N₂ into the resonant cavity at a flow rate of 4 L·min⁻¹ to induce the formation of a plasma jet at the tip of the needle. Subsequently, we gradually injected H₂ into the N₂ gas. During the process of increasing the hydrogen flow rate from 0 to 1 L·min⁻¹, the jet length gradually shortened and finally disappeared, as shown in figure 6(a). The underlying cause of this phenomenon is the quenching effect of hydrogen on the metastable state nitrogen atoms [27].

Figure  6.  Photographs of microwave plasma jet with the H₂ injected from (a) reactor branch port, and (b) quartz tube branch port (at the arrows).

DownLoad: Full-Size Img PowerPoint

Therefore, in order to deeply investigate the effect of pulse-modulated microwave plasma on ammonia synthesis, a method of injecting H₂ from the tail (at the arrows) of the nitrogen jet was used in this study. As shown in figure 6(b), with the increasing flow rate of H₂ injected from the tail of the flame, the quenching effect shortened the length of the jet but did not trigger the quenching phenomenon. At the same time, the plasma jet injected at the tip of the needle spreads upward due to the gas flow.

3.3   Effect of different microwave powers on ammonia synthesis

Discharge parameters are important factors affecting the effectiveness of ammonia synthesis. Therefore, we explore the influence of microwave parameters on the ammonia synthesis at the optimal nitrogen flow rate and H₂ flow rate. Since the plasma ejected at the tip of the needle is formed by a locally enhanced electric field resonance, the pulse absorbed power below 80 W is not sufficient to excite the particles to produce a plasma jet, so the microwave absorbed power is controlled between 80 W and 130 W in this work. The ammonia synthesis rate decreased from 2.93 to 1.41 μmol·min⁻¹ during the increase of microwave absorbed power from 84.42 to 124.47 W. N₂ conversion and EE decreased continuously with the increase of microwave absorption power, the results are presented in figure 7. It is pointed out by some researchers that there is an ionization enhancement effect of pulse-modulated microwave discharges from the point of view of lowering the plasma temperature [34]. As shown in figure 8(a), the higher the microwave absorption power, the higher the intensity of the emission spectra of NH radial, N₂ (C³Π_u→B³Π_g, Δν = −2) and N₂ (B³Π_g→A³Σ$_{\mathrm{u}}^+$, Δν = 2), and the higher the number of active particles within the system. In the process of ammonia synthesis, plasma excitation process is mainly through the reactions R1‒R3, R5, R6, R8 (table 2). In consideration of the venting methodology employed in this setup, the plasma is initially generated in nitrogen, resulting in a substantial concentration of excited state ${\mathrm{N}}^* _2$. Consequently, the intermediate NH radicals are more probable to be generated through both the L1 and L2 pathways. Figure 8(b) shows the trend of distribution of T_vib and T_rot for different microwave absorbed powers. The T_vib range from 3600±20 to 3450±20 K and the T_rot range from 2010±20 to 2160±20 K. Under the action of the microwave field, when the microwave energy matches the T_vib and T_rot energy levels of the molecules, the molecules will rapidly absorb the microwave energy and convert it into thermal energy, resulting in a rapid increase in the temperature of the gas. As the microwave absorbed power increases, more molecules absorb the microwave energy to collide with other particles in the system and transfer their energy to other molecules [28]. Therefore, in ammonia synthesis process, the gas temperature in the system rises rapidly through microwave energy absorption and energy collision transfer. The microwave absorption power is too high, and the temperature in the discharge area may rise above the optimal reaction temperature for ammonia synthesis. Such a high temperature environment will promote the decomposition reaction of ammonia, resulting in a slowdown in the production rate of ammonia [27].

Figure  7.  Ammonia generation rate, N₂ conversion and EE at different microwave absorbed powers. N₂ flow rate: 7 L·min⁻¹, H₂ flow rate: 1 L·min⁻¹, pulse modulation frequency: 25 kHz, pulse duty cycle: 50%.

DownLoad: Full-Size Img PowerPoint

Figure  8.  (a) Intensity of emission spectra of NH radial, N₂ (C³Π_u→B³Π_g, Δν = −2) and N₂ (B³Π_g→A³Σ$_{\mathrm{u}}^+$, Δν = 2), (b) T_vib and T_rot for different microwave absorption powers. N₂ flow rate: 7 L·min⁻¹, H₂ flow rate: 1 L·min⁻¹, pulse modulation frequency: 25 kHz, pulse duty cycle: 50%.

DownLoad: Full-Size Img PowerPoint

3.4   Effect of different pulse modulation frequencies on ammonia synthesis

Subsequently, the impact of pulse modulation frequency on ammonia synthesis by microwave plasma jet at low microwave absorbed power (86.87 W) is examined. The pulse modulation frequency serves to determine not only the repetition rate of the pulse but also the average power delivered to the system per unit time.

As the pulse modulation frequency is increased from 10 to 100 kHz, the nitrogen conversion rate demonstrates an overall upward trajectory, reaching a maximum of 0.357×10⁻⁷ at a pulse modulation frequency of 80 kHz (figure 9). Additionally, the ammonia conversion and EE reached their respective peaks at this time, with values of 3.57 μmol·min⁻¹ and 8.09×10⁻² g·(kW·h)⁻¹, respectively. This indicates that the reactants were utilized to their maximum potential at a pulse modulation frequency of 80 kHz. As the pulse modulation frequency increased from 10 to 100 kHz, the intensity of the emission spectra of NH radial, N₂ (C³Π_u→B³Π_g, Δν = −2) and N₂ (B³Π_g→A³Σ$_{\mathrm{u}}^+$, Δν = 2) particles within the system exhibited a continuous increase, as illustrated in figure 10(a). This indicates that, despite the reduction in pulsed microwave power from 50 to 5 μs (with a duty cycle of 50%), a considerable number of excited particles remain in the discharge channel during the interval when no microwave power is applied. In particular, the nitrogen-excited particles have a sufficiently long lifetime of up to 2 ms. This phenomenon is responsible for the persistence of the discharge channel and the presence of optical radiation upon cessation of microwave input [34–36]. The experimental results demonstrate that the reduction in the duration of the microwave power input cycle increases the number of excited particles in the discharge region, thereby significantly enhancing the probability of generating NH radicals (reactions R19‒R21) and NH₃ molecules (reaction R29). The highest yield of NH₃ is obtained when the pulse modulation frequency is 80 kHz and the microwave power input occurs every 6.25 μs. At this juncture, the vibrational temperature of the plasma is approximately 3600 K, while the rotational temperature is approximately 2100 K, as illustrated in figure 10(b). Between 80 kHz and 100 kHz, there is a discernible decline in both the vibrational and rotational temperatures of the plasma. The factors affecting the T_rot are primarily the collision and radiation processes occurring within the plasma. Higher T_rot are conducive to collisional excitation, radiation, and the transfer of energy among particles within the system. However, excessive T_rot can lead to system instability [28]. Additionally, a slight wobble in the tail of the microwave plasma jet was discerned in the experiments conducted beyond 70 kHz. This phenomenon manifested with greater clarity at higher frequencies of pulse modulation. This is one of the reasons why the EE and nitrogen conversion did not continue to increase after 80 kHz.

Figure  9.  Ammonia generation rate, N₂ conversion and EE at different pulse modulation frequencies. N₂ flow rate: 7 L·min⁻¹, H₂ flow rate: 1 L·min⁻¹, microwave incident power: 104.21 W, reflected power: 17.34 W, pulse duty cycle: 50%.

DownLoad: Full-Size Img PowerPoint

Figure  10.  (a) Intensity of emission spectra of NH radial, N₂ (C³Π_u→B³Π_g, Δν = −2) and N₂ (B³Π_g→A³Σ$_{\mathrm{u}}^+$, Δν = 2), (b) T_vib and T_rot for different pulse modulation frequencies. N₂ flow rate: 7 L·min⁻¹, H₂ flow rate: 1 L·min⁻¹, microwave incident power: 104.21 W, reflected power: 17.34 W, pulse duty cycle: 50%.

DownLoad: Full-Size Img PowerPoint

3.5   Effect of different pulse duty cycles on ammonia synthesis

In figure 11, it is shown that the ammonia synthesis rate, EE and nitrogen conversion rate all reach their maximum values at a duty cycle of 80%, which are 2.58 μmol·min⁻¹, 5.86×10⁻² g·(kW·h)⁻¹, and 18.48×10⁻⁸, respectively. When the duty cycle reaches 100%, the output mode of the power supply becomes continuous microwave, which changes the whole system to microwave discharge ammonia synthesis. Compared with the duty cycle of 80%, the ammonia synthesis rate, EE, and nitrogen conversion are reduced by about 22%. In other words, the traditional continuous microwave output mode is not the optimal condition in the context of microwave discharge ammonia synthesis. This is the fundamental reason why we introduced pulse modulation for the microwave discharge process for ammonia synthesis.

Figure  11.  Ammonia generation rate, N₂ conversion and EE at different pulse duty cycles. N₂ flow rate: 7 L·min⁻¹, H₂ flow rate: 1 L·min⁻¹, microwave incident power: 108.18 W, reflected power: 18.25 W, pulse modulation frequency: 25 kHz.

DownLoad: Full-Size Img PowerPoint

The duty cycle determines the delivery time and time interval of the microwave energy in the discharge system. By adjusting the duty cycle, it is possible to achieve precise control of the energy delivery [34]. When the microwave pulse is turned on, the microwave power is coupled into the plasma, where electrons absorbed energy and nitrogen molecules (N₂) collide to produce excited nitrogen molecules ${\mathrm{N}}^*_2$ (i.e. R9). The ventilation mode of the device determines that the dominant reaction for ammonia generation is the collision of excited nitrogen molecules ${\mathrm{N}}^*_2$ with hydrogen molecules (H₂) (i.e., R18‒R22). When the microwave pulse is turned off, that is, when the microwave power stops coupling to the plasma, the electrons cannot continue to derive energy from the microwave field. The heating of the gas by the microwave field can be reduced at this time [34].

As shown in figure 12(a), the intensity of the emission spectra of NH, N₂ (C³Π_u→B³Π_g, Δν = −2) and N₂ (B³Π_g→A³Σ$_{\mathrm{u}}^+$, Δν = 2) are generally increasing within the system during the process of increasing the duty cycle from 10% to 100%. In this process, there are a large number of excited particles (N₂ (A), N₂ (B), N₂ (C)) in the plasma. The collision of particles is more frequent and intense in the system, and the number density of particles is increasing, which accelerates the reactions of generating ammonia gas (R25‒R29) [3, 26]. However, at the same time, the collisions between particles (R29‒R37, R41‒R43) are also intensified, which leads the system towards an unfavorable direction for ammonia synthesis. According to the distribution of the simulated T_vib and T_rot (figure 12(b)), the optimal conditions for ammonia synthesis are approximately 2100 K and 3750 K for the T_rot and T_vib of the system, respectively. Moreover, according to the experimental results, the duty cycle affects the T_rot to a much greater extent than the T_vib.

Figure  12.  (a) Intensity of emission spectra of NH radial, N₂ (C³Π_u→B³Π_g, Δν = −2) and N₂ (B³Π_g→A³Σ$_{\mathrm{u}}^+$, Δν = 2), (b) T_vib and T_rot for different pulse duty cycles. N₂ flow rate: 7 L·min⁻¹, H₂ flow rate: 1 L·min⁻¹, microwave incident power: 108.18 W, reflected power: 18.25 W, pulse modulation frequency: 25 kHz.

DownLoad: Full-Size Img PowerPoint

4.   Conclusion

This study investigates the one-step direct synthesis of NH₃ at room temperature and atmospheric pressure using pulsed-modulated microwave plasma technology. A high-flow pulsed-modulated microwave plasma jet is employed, enabling microwave-assisted ammonia synthesis. The experiment involved the introduction of hydrogen at the tail of the nitrogen jet, a strategy that effectively prevented discharge extinction, a phenomenon caused by the mixing of the two gases. The results demonstrate that the ammonia production rate and EE reach their peaks at a coupled microwave power of 84.42 W in the resonant cavity, with values of 2.93 μmol·min⁻¹ and 6.64×10⁻² g·(kW·h)⁻¹, respectively. In comparison with the 80% duty cycle condition, the ammonia synthesis rate, EE, and nitrogen conversion rate decrease by approximately 22% under the 100% duty cycle, suggesting that the continuous microwave output mode is not the optimal mode for microwave discharge-assisted ammonia synthesis. Furthermore, a detailed analysis of the ammonia synthesis mechanisms and reaction pathways in the plasma reveals that the vibrational excitation of the microwave plasma significantly promotes the ammonia synthesis process. The experiment also shows that the optimal condition for ammonia synthesis in this setup occurs at a rotational temperature of approximately 2100 K.