Effect of Microwave Antenna Material and Diameter on the Ignition and Combustion Characteristics of ADN-Based Liquid Propellant Droplets

Abstract

Microwave-assisted ignition is an emerging high-performance ignition method with promising future applications in aerospace. In this work, based on a rectangular waveguide resonant cavity test bed, the effects of two parameters (material and diameter) of the microwave antenna on the ignition and combustion characteristics of ADN-based liquid propellant droplets were investigated using experimental methods. A high-speed camera was used to record the droplet combustion process in the combustion chamber, the effect of the microwave antenna on the propellant combustion response was analyzed based on the emission spectroscopy method, and finally, the loss of the microwave antenna was evaluated using a scanning electron microscope. The experimental results show that the droplet has the lowest critical ignition power (179 W) when the material of the microwave antenna is tungsten, but the ignition delay time is higher than that of copper. A finer diameter of microwave antenna is more favorable for plasma generation. At a microwave power of 260 W, the ignition delay time of the droplet with a microwave antenna diameter of 0.3 mm is 100 ms lower than that of 0.8 mm, which is about 37.5%. In addition, this study points out the mechanism of microwave discharge in the droplet combustion process. The metallic microwave antenna not only collects the electrons escaping from the gas discharge, but also generates a large amount of metallic vapor, which provides charged particles to the plasma. This study provides the possibility for the application of microwave-assisted liquid fuel ignition.

1. Introduction

The development of green propellants has been receiving attention from researchers in various countries, and the development of pollution-free and recyclable next-generation green propellants is of great scientific significance and application value. Ammonium dinitramide-based liquid propellant, a green single-component propellant, has the advantages of high performance, low toxicity, and high stability [1,2,3,4]. Currently, the Swedish PRISMA mission satellites have successfully completed in-space, on-orbit tests, demonstrating the superiority of ADN-based liquid propellants in terms of safety and storage [5,6]. However, conventional catalytic combustion ignition methods are commonly used for ADN-based thrusters. This method has limitations: one aspect is that cold start is not possible, and the catalytic bed needs to be preheated for at least 30 min [7]; on the other hand, the catalytic particles are prone to sintering and deactivation in high-temperature environments [8]. In recent years, microwave-assisted ignition has become a promising strategy, and the main principle is to realize the ignition of auxiliary fuels by microwave plasma discharge [9,10,11,12,13].

The trend in microwave-assisted ignition methods is generally to design new ignitors to replace conventional spark plugs, and to achieve large-area ignition in space by plasma breakdown of a combustible gas. For example, Hwang et al. investigated the effect of microwave-assisted plasma ignition on laminar flame development in a constant volume combustion chamber (CVCC) [14]. The direct use of microwave plasma to penetrate liquid fuels, on the other hand, faces technical difficulties in practice, which requires a quantum increase in the electric field strength of the device. Yu. A. Lebedev et al. studied microwave discharges in liquid hydrocarbons, and found that the discharges ignite at the tip of a microwave antenna [15,16]. The microwave radiation energy is not sufficient to directly ignite the liquid fuel, and it is necessary to use microwave focusing to achieve a high electric field intensity in a localized area.

The interaction between microwave and medium is the key factor of microwave energy in the propagation and conversion process. A metal microwave antenna as a good conductive medium can be used to focus the microwave [17]. Fragge et al. ignited kerosene droplets with the help of an aluminum microwave antenna and investigated the critical ignition power under different microwave antenna structures [18]. Metallic microwave antennas may be involved in chemical reaction processes in addition to electrical conductivity, and metal atoms (Cu, Al) released by plasma erosion can control the combustion reaction rate [19,20].

The objective of this study is to find out the optimal microwave antenna material and diameter for achieving stable droplet ignition. In this paper, the electric field distribution inside the combustion chamber is simulated, and the effects of microwave antenna materials and diameters on droplet ignition and combustion characteristics are investigated using experimental methods. Finally, the morphological characteristics of the microwave antenna during the combustion process were analyzed using scanning electron microscopy to assess the effect of the metal atoms released after the erosion of the microwave antenna on the combustion reaction. This study uncovers the mechanism of microwave plasma discharge in the droplet ignition and combustion process, which opens up the possibility of potential applications.

2. Experimental Setup and Methods

2.1. Experimental Apparatus

The details of the microwave ignition experimental setup have been presented in a previous paper [21] and are only briefly summarized in this paper. The schematic diagram of the experimental setup is shown in Figure 1. A solid-state microwave source provides microwave energy with a peak power of 300 W and a frequency of 2.45 GHz, and the uncertainty of the output power is ±3 W. A high-speed color video camera (Photron, Tokyo, Japan, FASTCAM Nova S9, with a maximum resolution of 1024 × 1024) and a macro lens (ZHONGYI OPTICS Co., LTD, Shenyang, China, 85 mm f/2.8 1-5X Super Macro) were used to capture images of the combustion of droplets. A miniature fiber optic spectrometer (Ocean Optics, Dunedin, FL, USA, Ocean USB 2000+) was used to acquire the emission spectra of the flames, with the spectrometer detecting a range of 200–900 nm. All the experiments were carried out in a combustion chamber with ambient temperature and pressure conditions. The resonant cavity was modified from a BJ22 rectangular waveguide that had a length of 109.22 mm and a width of 54.61 mm. The cavity was made of pure aluminum and closed at the end, with the dominant mode of transmission being the TE₁₀ mode. A microwave antenna is used to focus the electric field inside the combustion chamber, as shown in Figure 2. It is made of a good conductive and wear-resistant metal (tungsten, copper, brass). The antenna was placed at the highest intensity of the electric field in the combustion chamber, i.e., 1/4 wavelength (30.5 mm) from the center of the end of the combustion chamber. Before the start of each experiment, a droplet with a volume of 2 μL was injected into the top of the antenna using a micro-syringe, ensuring that the droplet had an ellipsoidal shape.

2.2. Quality Factor

Resonant cavities often choose to operate in a single mode in order to achieve a high-quality factor in practical applications, and such microwave resonant cavities can usually be simulated with series or parallel RLC lumped elements equivalent circuits. In this study, the parallel equivalent circuit simulation is mainly used, which can be categorized into two cases, unloaded and loaded. Figure 3 shows the parallel RLC collector circuit with unloaded.

The input impedance is:

$$Z_{\mathrm{in}}={\left({\displaystyle \frac{1}{R}}+j\omega C+{\displaystyle \frac{1}{j\omega L}}\right)}^{-1}$$

(1)

where $R$ is the resistance, $\omega$ is the angular frequency, $L$ is the inductance, and $C$ is the capacitance.

The input power transferred to the resonant cavity is:

$$P_{\mathrm{in}}={\displaystyle \frac{1}{2}}\left|I^2\right|\left(R+j\omega L-j{\displaystyle \frac{1}{\omega C}}\right)$$

(2)

where $V$ is the voltage of the resonant cavity equivalent circuit.

The power dissipated by the resistor R is:

$$P_{\mathrm{loss}}={\displaystyle \frac{1}{2}}{\displaystyle \frac{{\left|V\right|}^2}{R}}$$

(3)

The average electric field stored energy in capacitor C is:

$$W_{\mathrm{e}}={\displaystyle \frac{1}{4}}{\left|V\right|}^{2}C$$

(4)

The average magnetic field stored energy in the inductor L is:

$$W_{\mathrm{m}}={\displaystyle \frac{1}{4}}{\left|I_{\mathrm{L}}\right|}^{2}L={\displaystyle \frac{1}{4}}{\left|V\right|}^2{\displaystyle \frac{1}{{\omega }^{2}L}}$$

(5)

where $I_L$ is the current flowing through the inductor.

$$P_{\mathrm{in}}=P_{\mathrm{loss}}+2j\omega \left(W_{\mathrm{m}}-W_{\mathrm{e}}\right)$$

(6)

The circuit resonates when the average electric field stored energy is equal to the average magnetic field stored energy. At this point, the pure impedance R is the input impedance of the circuit and ${\omega }_0$ is the resonant frequency. The Q of the parallel resonant circuit is:

$$Q=\omega {\displaystyle \frac{2W_{\mathrm{m}}}{P_{\mathrm{loss}}}}={\displaystyle \frac{R}{{\omega }_{0}L}}={\omega }_{0}RC$$

(7)

In practice, the resonant circuit must be coupled with other external circuits, a process that will result in a reduction in the Q of the entire resonant circuit. The Q value of the resonant cavity coupled to the external circuit is called the on-load Q value and is generally denoted by Q_L.

In the presence of a load, the quality factor of the whole circuit is:

$${\displaystyle \frac{1}{Q_{\mathrm{L}}}}={\displaystyle \frac{1}{Q_{\mathrm{e}}}}+{\displaystyle \frac{1}{Q}}$$

(8)

2.3. Electric Field Simulation

To accurately analyze the effect of the microwave antenna on the spatial electric field distribution in the resonant cavity, the electric field in the waveguide and combustion chamber was simulated in this study using COMSOL Multiphysics software, version 6.0 [22]. The simulation conditions consider the presence of walls and impedance matching, the microwave input power is set to 100 W. Both electric field and plasma are present in the resonant cavity, and modern self-consistent models of microwave discharges are used in this study [23,24]. The simulation temperature was set to 293.15 K and the pressure was set to one standard atmospheric pressure.

The microwave propagates along the cavity within the rectangular waveguide and the transmission mode is TE₁₀ mode. The transmitted power inside the rectangular waveguide is:

$$P={\displaystyle \frac{1}{2}}\mathrm{Re}{\displaystyle {\int }_0^a{\displaystyle {\int }_0^b-E_y{H}_x^{*}}}dxdy={\displaystyle \frac{ab}{4}}{\displaystyle \frac{E^2}{Z}}$$

(9)

where a is the long side of the rectangular waveguide, b is the narrow side of the rectangular waveguide, E is the electric field strength, and Z is the equivalent impedance of the waveguide:

$$Z={\displaystyle \frac{\eta }{\sqrt{1-{(\lambda /2\mathrm{a})}^2}}}$$

(10)

where $\eta$ is the transmission efficiency.

The return loss is used to represent the ratio of the power of the forward signal to that of the reflected signal, and can be seen as a function of the absorption-reflection relationship of the component to determine the power, as shown in Equation (3):

$$L_r=10\lg {\displaystyle \frac{P^{+}}{P^{-}}}=10\lg {\displaystyle \frac{P^{+}}{{\left|\Gamma \right|}^2{P}^{+}}}=-20\lg \left|\Gamma \right|$$

(11)

where Γ is the reflection coefficient, P⁺ is the forward power, and P⁻ is the reverse power.

Figure 4 shows the return loss for the three reduced region lengths. When L = λg/4, the return loss is larger, and there is no obvious inflection point within the selected frequency. The return loss gradually decreases with the rise of the input frequency, and this situation illustrates that its optimum frequency is not within the selected frequency; when L = 3λg/4, the peak inflection point also does not occur within the selected frequency, and it is still in the decreasing trend under the condition of 2.6 GHz; L = λg/2. When L = λg/2, the return loss of the conductor is minimized and is between −26 dB and −30 dB.

Figure 5 shows the distribution of electric field strength in the combustion chamber for different microwave antenna diameters and shapes. The microwave antenna directly affects the electric field distribution inside the combustion chamber. The peak electric field strength is 2.43 × 106 V/m for the microwave antenna diameter of 0.3 mm and 1.35 × 108 V/m for the microwave antenna shape of tip. This is due to the large curvature of the antenna tip, which has a large charge surface density, and the electric field strength at the edge of the antenna tip is the largest.

3. Results and Discussion

3.1. Effect of Microwave Antenna Materials on the Ignition and Combustion Characteristics of Droplets

Theoretically any conductive material can be used as a microwave antenna, but materials with different conductivity for microwave antennas have an important effect on the generation of plasma discharges. In addition, the instantaneous heat generated by propellant decomposition and combustion is extremely high, and the thermodynamic properties of the material should be considered. The larger the melting point, boiling point, and thermal conductivity, the less susceptible the material is to galvanic corrosion. In this work, the ignition and combustion properties of tungsten, copper, brass, and molybdenum when used as microwave antennas are compared and repetitive experiments are performed.

Figure 6 demonstrates the process of ADN-based liquid propellant flame profile development for different microwave antenna materials at a microwave input power of 260 W. The continuous microwave energy feed facilitates the expansion of the liquid droplets and micro-explosions, while the ADN decomposes a large amount of nitrogen oxides at high temperatures. Figure 6a shows the obvious plasma ball generation, where the fire nucleus arises from the top of the tungsten filament and gradually expands, and finally floats upwards to extinguish. Figure 6b shows the droplet combustion process when copper is used as the microwave antenna. From the figure, the flame is yellowish green, which may be the copper atoms involved in the combustion reaction. Under the action of microwave energy, the droplet starts to evaporate and decompose, and bubbles appear inside the droplet. The bubbles expand to the maximum value at 38.5 ms and start to break, and the combustible gas released by the bubbles is ignited by the plasma. Figure 6c shows the image of the droplet combustion when brass is used as the microwave antenna. During the combustion process, a clear tip-discharge phenomenon occurs, and the top of the brass breaks down with a clear filamentary plasma generation. Due to the pushing effect of the plasma, the droplet could not be maintained at the top of the antenna, and there was no large-scale fire in the combustion chamber. The image of the droplet combustion when molybdenum wire is used as the microwave antenna is shown in Figure 6d. Due to the poor electrical conductivity of molybdenum, the droplet took a much longer time in the evaporation stage. The droplet needs to continuously absorb microwave energy to reach the ignition point of the combustible gas.

Figure 7 shows the variation of ignition delay time and combustion duration with microwave power for different microwave antenna materials. The ignition delay time is defined as the time from the expansion of 10% of the droplet diameter to the appearance of a distinct flame core, and the combustion duration refers to the time from the appearance of a distinct flame core in the combustion chamber to the disappearance of the flame. As shown in Figure 7a, the ignition delay time tends to decrease with increasing microwave power when tungsten wire is used as the microwave antenna. The combustion duration at a microwave power of 300 W is 55 ms lower than that at 220 W, a 69% decrease. The copper microwave antenna has the lowest ignition delay time as shown in Figure 7b. This is since the copper wire is easy to tip-discharge to generate plasma, the propellant is accelerated to decompose at high temperatures, and the combustible gas is more easily ignited. Moreover, copper may be involved in the chemical reaction of the propellant, facilitating the generation of key intermediate products. Figure 7c shows the droplet ignition delay time and combustion duration when a microwave antenna made of brass is used. The lowest ignition delay time, about 76 ms, was observed for a microwave power of 300 W. The ignition of the ADN-based liquid propellant droplet became deteriorated when molybdenum wire was employed as the microwave antenna, as shown in Figure 7d. This may be due to the fact that the molybdenum wire inhibits the redox reaction of the propellant, and the molybdenum wire reacts with the nitrogen oxides decomposed by the ADN, which is detrimental to the combustion of the droplets [25].

Figure 8 shows the critical ignition power of the droplet for different microwave antenna materials. The critical ignition power is defined as the difference between the input power and the reflected power:

$$P_e=P_{in}-P_{re}$$

(12)

The experimental results show that the critical ignition powers of tungsten, copper, brass, and molybdenum are 179 W, 183 W, 185 W, and 203 W. The critical ignition power is affected by the conductivity and thermal conductivity of the microwave antenna, and the higher conductivity means that it is easier to generate plasma. The thermal conductivity of the microwave antenna has an uncertainty on the ignition of the droplet. Copper has a better thermal conductivity than tungsten, while tungsten has the lowest critical ignition power of the droplet. Part of the energy generated by microwave radiation is absorbed by the antenna, part of the energy is absorbed directly by the droplet, and the remaining energy is transferred to the walls of the combustion chamber and other areas. When the material of the microwave antenna is copper, most of the microwave energy is absorbed by the antenna. In addition, the electrical conductivity of conventional metallic materials deteriorates at high temperatures, which is not favorable for plasma-assisted droplet ignition.

3.2. Effect of Microwave Antenna Diameter on the Ignition and Combustion Characteristics of Droplets

Figure 9 shows the images of droplet combustion at microwave antenna diameters of 0.3 mm, 0.5 mm, and 0.8 mm. The microwave antennas were fabricated from structurally stable tungsten wires and the top of the antenna used for each experiment was polished smooth. The experimental results show that the combustion characteristics of the droplet are better at a microwave antenna diameter of 0.3 mm, as evidenced by a larger flame profile area and a stable and bright flame. Previous studies have illustrated that droplets are ignited by plasma, and the thin microwave antenna diameter (0.3 mm) means that plasma generation is easier. When the microwave antenna diameter is 0.3 mm, the first generation of plasma is at 49 ms, while when the microwave antenna diameter is 0.8 mm, the first generation of plasma is at 128 ms.

Figure 10 shows the ignition delay time and combustion duration for different microwave antenna diameters varying with microwave input power. The microwave input power directly affects the absorption of microwave energy by the flame, and the ignition delay time of the droplet decreases with the increase of microwave power. When the microwave power is 200 W, the droplet ignition delay time at a microwave antenna diameter of 0.3 mm is 13.9% lower than that of 0.5 mm and 33.3% lower than that of 0.8 mm. Reducing the diameter of the microwave antenna is very effective in reducing the ignition delay time of the droplet. It is worth noting that the ignition delay should be minimized as much as possible in practical applications, while ensuring that the antenna can work in a high-temperature environment for a long time.

Figure 11 shows the critical ignition power of the droplet for different microwave antenna diameters, and the material of the microwave antenna is tungsten. The critical ignition power of the droplet increases with the increase of the microwave antenna diameter. The critical ignition power at microwave antenna diameter of 0.3 mm is only 161 W, which is 19.5% lower than that at antenna diameter of 0.8 mm. The experimental results are consistent with the simulation results in Section 2.2, and the thinner diameter of the microwave antenna implies that the electric field strength on the surface of the antenna is larger, which is more favorable for the excitation of the plasma to ignite the propellant.

3.3. Microwave Plasma Discharges and Antenna Losses

In the whole microwave-assisted ignition process, the observed microwave plasma discharge can be divided into gas discharge and metal discharge. In gas discharge, air in the combustion chamber undergoes high-energy electron collisions, leading to ionization and the formation of plasma. Metal discharge involves a metal microwave antenna emitting electrons at high temperatures, which are accelerated by a magnetic field. This acceleration causes collisions between electrons and neutral metal atoms, resulting in the dissociation and formation of a high density of metal ions. The experimental results show that the gas discharge and the metal discharge are not completely separated, and both jointly promote the combustion of the droplet during the droplet combustion.

The temporal emission spectra of the plasma discharge and flame are shown in Figure 12, counting three distinct emission spectra during microwave-assisted ignition. In the phase between the microwave input and the burning of the droplet, the air is pierced to produce an intense bright white light. The emission spectra of Na and K were detected in the flame, which could be caused by alkali metal salts in the air. Salts from both oxygen and on the experimental setup leads to the production of Na and k bands. The discharge during this period is mainly dominated by gas discharge, and the corrected spectrum exhibits a distinct C₂ band. The droplets are induced by microwave energy and begin to frequently micro-explode and decompose large amounts of combustible gases. The propellant was successfully ignited by the plasma effect, and this stage exhibited mainly spectral features of ADN and methanol flames. In the later stage of combustion, the metal microwave antenna was vaporized at high temperature due to the large amount of heat released from the propellant. An obvious plasma sphere was observed in the experiment, and the plasma discharge at this time was dominated by the metal discharge, accompanied by the burning of residual propellant. The metal microwave antenna not only collects the electrons escaping from the gas discharge, but also generates a large amount of metal vapor, which provides charged particles to the plasma sphere.

Oxidization and corrosion of metallic microwave antennas at high temperatures are issues that must be considered during the ignition process. Tungsten is the most suitable material for microwave antennas due to its high hardness and high melting point (tungsten has a melting point of 3410 °C). In order to observe the involvement of tungsten wire in the droplet ignition process, the surface of the burned tungsten wire was examined using a scanning electron microscope. As shown in Figure 13, the surface of the tungsten filament was eroded, and visible holes were formed (Figure 13b–d). When the tungsten wire is struck and discharged under high temperature conditions, the following morphological features will appear: (1) molten area: the energy generated by the microwave discharge will cause the surface of the tungsten wire to melt locally, and a certain area of molten area will appear. (2) Air holes: around the molten zone on the surface of tungsten wire, many small air holes will appear. This is due to the gas generated by the microwave discharge in the tungsten wire surface localized expansion formation. (3) Oxide: The experiment is carried out under atmospheric conditions. Due to the involvement of oxygen in the heating process, the tungsten wire is oxidized at high temperature to produce tungsten trioxide.

4. Conclusions

In this paper, the effect of two parameters (material and diameter) of the microwave antenna on the ignition and combustion characteristics of ADN-based liquid propellant droplets is investigated using experimental methods. The effect of the microwave antenna on the combustion response of the propellant was analyzed by combining a high-speed camera system and emission spectroscopy methods, and finally the loss of the microwave antenna was evaluated by using scanning electron microscopy. The main conclusions are as follows:

(1)

It is found that the droplets have lower ignition delay time and critical ignition power when the materials of microwave antenna are tungsten and copper. However, tungsten or tungsten alloy are more suitable for microwave antennas considering the wear resistance of the material.

(2)

A finer microwave antenna diameter (0.3 mm) is more favorable for plasma generation, and the flame profile area during propellant combustion is larger, and the flame is stable and bright. Reducing the diameter of the microwave antenna is very effective in reducing the ignition delay time and critical ignition power of droplets.

(3)

Microwave plasma combustion is the result of the joint action of gas discharge and metal discharge. In the pre-combustion stage of the droplet, the gas discharge is mainly dominated, while the metal discharge is mainly dominated in the post-combustion stage. The metal atoms on the microwave antenna control the combustion reaction by providing charged particles to the gas plasma.