Abstract
Recently, the application of ionic wind in atmospheric propulsion has gained significant attention, but existing devices face challenges in achieving sufficient thrust for practical applications. This study proposes a novel electrode structure combining a serrated single-ring emitter and multi-ring collector, which offers two key advantages: (1) The 1–2 times thrust increase is achieved through enhanced ion generation, improved ion drift efficiency, enabled by the sawtooth emitter and multi-ring collector design, (2) a compact, lightweight (17 g) design with good structural stability. We compared three electrode structures under varying conditions (voltage: 20–40 kV, electrode gaps: 60–120 mm, ring diameters: 60–100 mm). The sawtooth multi-ring structure achieved a maximum thrust of 164 mN/m under 40 kV and a 60 mm gap. Furthermore, through further optimization of the structural parameters of the sawtooth multi-ring, the thrust density achieved a 28.2% enhancement under equivalent operational conditions. This result highlight the potential of ionic wind propulsion for low-altitude flight, with further optimization promising greater efficiency.
Subject terms: Plasma physics, Aerospace engineering
Introduction
The phenomenon of ion wind can be traced back to 1709, when it was first discovered by Hauksbee1. Subsequently, in 1899, Chattock2 conducted an in-depth quantitative study of its generation mechanism, establishing a relatively comprehensive theoretical framework. After years of theoretical exploration, ion wind technology gradually found applications in various fields such as dust removal, heat dissipation, and air purification3,4. However, in the realm of propulsion technology, early scholars like Wilson, Gilmore, and Monrolin regarded ion wind as a suitable power source only for small aircraft such as drones due to its relatively low thrust output5–7. This conventional notion was challenged by Xu8 et al., who successfully developed a small drone demonstrator powered entirely by ion wind. Their work demonstrated that through technological innovation, the limitations of thrust-to-power ratio and thrust density could be partially addressed, enabling the potential application of ion wind in propulsion. Notably, their drone carried its own on-board power supply and achieved a propulsive force of 3.2 N, a flight speed of 4.8 m/s, and a thrust density of 3 N/m
. While this marked a significant milestone, the efficiency and flight duration of ion wind propulsion still lag behind traditional electric engines with propellers, as highlighted by recent studies such as Grosse9 et al. and ongoing efforts by groups like Undefined Technologies. To further enhance the thrust and thrust-to-power ratio of EAD devices, Xu et al. conducted experimental research on large-gap EAD devices. By adjusting the electrode gap between 50 and 300 mm, they achieved a thrust-to-power ratio of up to 15N/kW10. Additionally, they explored Dielectric Barrier Discharge (DBD) technology as an alternative to traditional corona discharge for ion generation. Their findings revealed that under a thrust of 50mN/m, the thrust-to-power ratio could be elevated to 20N/kW, while even at a thrust of 150 mN/m, the ratio remained at a high level of 10N/kW11. These advancements have enriched the theoretical framework of EAD propulsion technology and laid a foundation for its future application in broader domains. Today, the primary application areas of EAD thrusters encompass a diverse range of systems. These include fixed-wing aircraft that harness ion flow for forward propulsion, while generating lift through the incoming airstream5,12; corrective engines mounted on balloons operating at low pressures13; vertical take-off and landing (VTOL) aircraft14,15; EAD (Electroaerodynamic) micro-robots16,17; and EAD propellers18.
Electric propulsion has emerged as a promising trend in aeronautical propulsion, offering potential alternatives to traditional combustion engines19,20. In particular, ionic thrusters operating in the atmosphere have garnered significant interest due to their absence of moving parts, low noise output, and minimal maintenance requirements21. However, further research is necessary to enhance their thrust density and efficiency, which remain key challenges for practical applications. Recent work by Gomez-Vega22–24 et al. and others has focused on improving thrust-to-power ratios and exploring new configurations, highlighting the ongoing efforts to advance this technology.
The EAD unit generally comprises a high-voltage electrode (emitter) and a grounded electrode (collector). The emitter, typically a sharply curved surface (such as a pin or thin wire) with an extremely small radius of curvature. The collector, on the other hand, is usually a curved surface or a plane with a larger radius of curvature. When the emitter is positively charged and its voltage is sufficiently high to exceed the critical discharge electric field strength of the atmosphere, discharge occurs at the electrode tip.This ionization process splits the air molecules in the vicinity into electrons and positive ions, forming an ionization zone. Under the influence of Coulomb forces, the positive ions drift towards the collector, and this moving region is referred to as the drift zone. Meanwhile, the electrons attach themselves to neutral molecules during their movement, forming negative ions. Subsequently, nearby positive and negative ions may neutralize into neutral molecules due to electrostatic attraction.Within the drift zone, the positive ions are accelerated by the electric field during their drift and collide with neutral gas particles, propelling them towards the collector. This collective motion generates the ion wind. The majority of EAD thrusters uses thin metallic wires as emitting electrodes; however, these wires are fragile and cannot sustain bending loads. This can potentially lead to an early failure of the system. Increasing the diameter of the wires would increase the robustness of the emitters but, at the same time, it would increase the inception voltage, leading to a degradation of the performance of the thruster. Meanwhile, the thrust density of this structure also fails to meet the requirements of practical applications.
These conflicting requirements indicate that a possible experimental investigation can explore other emitting shapes, such as sawtooth, which have a small curvature radius at their edges, enhancing the local electric field strength and facilitating corona inception at lower voltages, as supported by Peek’s formula (Eq. 9), while also providing much higher structural integrity. In this study, a new type of sawtooth multi-ring electrode structure was designed (The multi-ring collector provides a larger surface area for ion collection, reducing ion recombination losses and increasing the effective ion drift velocity), which has a larger thrust per unit length compared with the ultra-thin long wire used in the classic atmospheric thruster configuration.
EAD propulsion
Ionic wind induced by corona discharge under direct current voltage is described by the following equations. The electric potential V is governed by Poisson’s equation25
 |
1 |
where 
is the space charge density, which is the measure of electric charge per unit volume, and 
is the dielectric permittivity of air.
The electric potential is represented by electric field intensity E as equation25
 |
2 |
The thrust acting on the EAD cell can be expressed as the volume integral of the product of the charge density 
and the electric field E at a given point26,27.
 |
3 |
The current arises due to the movement of charged particles in the drift zone with velocity v. Thus, the current density can be expressed as
 |
4 |
where 
is the air ion mobility, and the ion drift velocity is equal to the product of mobility and the electric field.
In the one-dimensional case, the current density can also be found using the Mott-Gurney law
 |
5 |
where d is the electrode spacing. Note that this equation assumes a 1D simplification, which is commonly used to model the current density in EAD systems.
However, corona discharge appears at a certain voltage value called the corona inception voltage. 
leads to a bias of the I - U curves, and the dependence of the current on the voltage may differ from the quadratic form28, and can be described by Townsend’s law:
 |
6 |
where C is a constant, V is the applied voltage, and 
is the inception voltage. Thus, the current can be written as:
 |
7 |
where S is the cross-section area of the drift region, k - power-law coefficient, which from Eq. (5) is equal to 2, but may differ in real systems.
 |
8 |
where 
is the emitter radius.
Peek’s formula defines the minimum electric field on a conductor required to initiate electrical discharge29
 |
9 |
where 
is the breakdown field strength of the air gap, 30 kV/cm can be taken at standard atmospheric pressure, 
is the coefficient of surface roughness of the electrode.
 |
10 |
where 
is the additional coefficient, 
and 
are the reference temperature (293 K) and reference pressure (1 atm) respectively.
The thrust generated by the dynamic structure based on ionic wind can be expressed as30:
 |
11 |
In addition to thrust, thrust power ratio is also an important parameter for evaluating the performance of a thruster. The thrust power ratio 
is used to determine the efficiency of converting electrical energy into thrust, and it can be calculated in conjunction with Eq. (7) to yield the power consumed by the thruster as:
 |
12 |
Therefore, its thrust power ratio is
 |
13 |
Experimental setup
The schematic of the experimental system is shown in Fig. 1. The system consists of 4 parts, namely, the electrostatic hydrodynamic propulsor, high-voltage direct current (HVDC) power supply, electronic balance, and lever mechanism.
Fig. 1.
Experimental setup.
The EAD unit is powered by a TD2200(0-50kv) rack-mounted high-voltage direct current (HVDC) power supply. Voltage and current measurements can be directly obtained from the high-voltage power supply, with accuracies of 0.01 mA for current and 0.01 kV for voltage. The power supply is equipped with high-voltage cables capable of operating up to 50 kV DC. Prior to reaching the collector, the cables are replaced with wires of 1 millimetre in diameter to minimize the impact of bending forces on weight measurements. Additionally, the wires were securely fixed to avoid errors caused by vibrations or movement. This ensures that the mechanical damping caused by the cables is minimized. To minimize parasitic effects, all high-voltage connections were insulated with polyimide tape (Kapton) wrapping for enhanced electrical insulation and mechanical stability, while maintaining a clear 1-meter radius around the experimental area free of unnecessary dielectric materials.
Thrust measurement is conducted on a one-meter-long “seesaw” device, which maintains the electronic balance at a sufficient distance to separate the strong electric field required for generating ion wind from the thrust measurement equipment. This separation eliminates the electrical effects from the high-voltage test site, allowing the electronic balance to operate normally. The length of the seesaw arm is experimentally selected with a significant margin to ensure that the measurements taken by the electronic balance are not affected by the electric field generated in the high-voltage area. The system has previously been tested under the voltage (50 kV) to comprehensively verify that the use of the electronic balance, whether under high voltage or not, does not cause data corruption. The experiments were conducted in a controlled environment with stable temperature and humidity to minimize the impact of variations in these factors on the performance of the force sensor and the ion wind generation process. The seesaw structure balances around a central pivot. One end is equipped with and powered by a high-voltage ion system, while a 200 g standard weight is hung at the other end on a model SN-FA3204 electronic analytical balance capable of measuring up to 320 g with an accuracy of 0.0001 g. Due to error considerations, readings for this experiment were only recorded to 0.01 g. Whenever the device is ready for thrust measurement and before high voltage is applied, the scale is zeroed. When the electrostatic hydrodynamic propulsor is mounted on this structure, The high-voltage area is equal to the arm length of the electronic balance, facilitating a direct reading of the thrust value on the electronic balance.
The EAD unit employs a single-ring-to-multiple-rings geometric structure,comprising a zigzag copper strip (250 mm in length with 50 teeth) attached to the outer edge of the emitter’s single ring (diameter D = 80 mm) and a multi-ring collector (with diameters of 60, 80, and 100 mm,Ring width=3 mm) whose front surface is fully covered with conductive adhesive tape. The sawtooth electrodes are fabricated by cutting and shaping metal plates using laser processing technology. Brass, as an electrode material, demonstrates excellent mechanical stability, electrical conductivity, and machinability. Under corona discharge conditions, it exhibits a certain level of corrosion resistance and wear resistance, while its electrical degradation rate remains relatively low31. Additionally, the surface oxide layer of brass can further mitigate the corrosion and degradation processes to some extent. Both the single ring and the multiple rings are fabricated using 3D printing. The distance d between the emitter and the collector varies from 60 mm to 120 mm, with a step size of 20 ± 0.2 mm. The base of the single-ring emitter and the multi-ring collector are both attached to a lightweight wooden board to facilitate thrust measurement and precise determination of the gap distance. The main parameters of the annular EAD unit are illustrated in Figure 2 and Figure 3.
Fig. 2.
Sawtooth multi-ring electrodes structure.
Fig. 3.
Multi-ring size diagram(left) Single-ring size diagram(right).
Results
All experiments were conducted at room temperature of 24 ± 1 
C and under standard atmospheric pressure, with positive voltage applied to the emitter. The curves in the graphs were fitted using the Allometric function, and the fitting process was optimized using the Levenberg-Marquardt algorithm. We conducted three independent measurements for each experimental condition and calculated the average as the final result.
When the electrode structure is of the patch type, Figure 4 demonstrates the relationship between thrust per unit length (in millinewtons per meter) and applied voltage (in kilovolts) for different electrode gaps. The graph indicates that, in general, thrust per unit length increases quadratically with the voltage. At the same applied voltage, thrust per unit length is decreases with 
.From the perspective of electric field intensity, this may be because, as the electrode gap increases, the electric field intensity decreases, which may result in weakened corona discharge intensity and reduced thrust. In terms of discharge current, this may occur because, when the electric field intensity decreases, the discharge current also decreases correspondingly. The magnitude of the discharge current directly reflects the movement of charges in the electric field, so a decrease in discharge current implies a reduction in the force experienced by charges in the electric field, leading to decreased thrust.
Fig. 4.
Patch thrust versus voltage.
When the electrode structure is of the single-ring to single-ring configuration, Figure 5 demonstrates the relationship between thrust per unit length (in millinewtons per meter) and applied voltage (in kilovolts) for different ring diameters and electrode gaps. The graph shows that, thrust per unit length increases quadratically with the voltage.
Fig. 5.
Single-ring thrust versus voltage.
If the ring diameter remains constant, thrust per unit length is decreases with 
at the same applied voltage. From the perspective of electric field intensity, when the electrode gap decreases, the distance between the high-voltage electrode and the grounded electrode shortens, resulting in an increase in electric field intensity. With increased electric field intensity, the positive ions generated by the ion generator experience a stronger electric field force, accelerating to gain greater kinetic energy. These high-speed ions collide with surrounding air molecules, generating a stronger ion wind and thus increasing thrust. Additionally, the decrease in electrode gap also affects ion concentration. When the electrode gap decreases, the electric field intensity increases, facilitating the production of more ions, thereby increasing ion concentration and improving thrust. In summary, the reduction in electrode gap enhances ion wind thrust, primarily due to the combined effects of increased electric field intensity and elevated ion concentration.
If the electrode gap remains constant, the thrust per unit length increases with the ring diameter under the same applied voltage. As the diameter of the ring electrode increases, the scope of the electric field distribution expands accordingly. While the peak value of the electric field intensity may decrease, the increased coverage area of the electric field allows more ions to be subjected to the force of the electric field, enabling more ions to be accelerated to higher speeds. Consequently, from an overall perspective, the thrust generated increases.
When the electrode configuration is of the single-ring to multi-ring type, Figure 6 presents the relationship between thrust per unit length (in millinewtons per meter) and applied voltage (in kilovolts) for various electrode gaps. In this scenario, the diameter of the emitter ring is set to D = 8 cm. The graph generally shows that thrust per unit length increases quadratically with the voltage, which is consistent with Eq. (11). As can be seen from the diagram, under the same conditions, the thrust per unit length of the sawtooth multi-ring structure is significantly higher than that of the wire-column structure Xu10.
Fig. 6.
Multi-ring thrust versus voltage.
At the same applied voltage, thrust per unit length is decreases with 
. Taking an applied voltage of 30 kV as an example, when the electrode gap increases from 6 cm to 8 cm and then to 10 cm, thrust per unit length decreases from 118.4 mN/m to 72 mN/m and further to 52 mN/m. When the collector changes from a single ring to multiple rings, under the same applied voltage and ring diameter, the thrust per unit length of the multi-ring configuration increases significantly, approximately 1.5 times that of the original single-ring configuration. The fundamental reasons for the performance differences between multi-ring and single-ring structures is: (1) Multi-ring structures expand the electric field, enabling more efficient ion generation and drift over a larger area compared to single-ring structures. (2) Multi-ring collectors reduce ion recombination losses by providing multiple collection points, enhancing thrust. (3) Multi-ring designs offer better mechanical stability under high-voltage conditions, ensuring consistent performance. In the future, further optimization of teeth shape and multi-ring geometry can be conducted to enhance electric field distribution, boost ion generation, and reduce aerodynamic drag.
Figures 7 and 8 illustrates the measured variation of collector current with respect to voltage. As evident from Figs. 7 and 8, the function exhibits an approximate quadratic exponential characteristics, which is consistent with Eq. (7).
Fig. 7.
Single-ring current versus voltage.
Fig. 8.
Multi-ring current versus voltage.
Figures 9 and 10 illustrates the relationship between the thrust-to-power ratio (in mN/W) and the applied voltage (in kV) for various ring diameters and electrode gaps. As predicted by Eqs. (11) and (12), the thrust-to-power decreases as 
, resulting from the quadratic scaling of thrust (
) and cubic scaling of power (
) at 
. This trend is experimentally observed as decreasing efficiency at higher voltages, primarily due to enhanced energy dissipation mechanisms including Joule heating and ion recombination losses. As can be seen from the diagram, under the same conditions, the thrust-to-power ratio of the sawtooth multi-ring structure is double that of the single-ring structure, while it is comparable to the wire-column structure Xu10.
Fig. 9.
Single-ring thrust-to-power versus voltage.
Fig. 10.
Multi-ring thrust-to-power versus voltage.
Figures 11 and 12 llustrates the relationship between thrust per unit length (in mN/m) and current per unit length (in mA/m) for various ring diameters and electrode gaps. The graph demonstrates that the function T(I) has a non-linear character. From Eq. (11), T should be directly proportional to I. This may be attributed to the following reasons: (1) At higher currents, the density of ions in the drift zone increases, leading to a higher probability of ion recombination. This reduces the effective number of ions contributing to thrust, resulting in a non-linear increase in thrust with current. (2) As the current increases, the electric field distribution between the electrodes may become non-uniform due to space charge effects. This distortion can alter the ion drift velocity and reduce the efficiency of thrust generation. (3) At higher currents, energy losses such as Joule heating and collisions between ions and neutral gas molecules become more significant. These losses dissipate energy that would otherwise contribute to thrust, leading to a sub-linear relationship between thrust and current. (4) The surface condition of the electrodes (e.g., roughness, oxidation) can affect the corona discharge process. At higher currents, surface effects may become more pronounced, leading to deviations from the ideal linear relationship. (5) residual parasitic currents (despite polyimide insulation and controlled experimental environment) and collector-induced reverse ion emission10.
Fig. 11.
Single-ring thrust versus current.
Fig. 12.
Multi-ring thrust versus current.
The experimental results presented in Table 1 demonstrate the performance of various electrode configurations under different operating conditions. Effect of teeth angle: Configurations with a 30
angle generally produce higher thrust and better efficiency compared to those with 45
or 60
angles. This is likely due to the enhanced electric field concentration at smaller teeth angles, where the reduced curvature radius intensifies local field strength according to Peek’s formula (Eq. 9). For example, the 30
angle configuration achieves a thrust of 172 mN/m at 34 kV, while the 60
angle configuration only reaches 144 mN/m under the same conditions. Effect of number of teeth: The thrust demonstrates an inverse relationship with the number of teeth. This is because fewer teeth create larger inter-teeth gaps that reduce space charge accumulation, minimizing ion recombination losses and each teeth maintains stronger individual corona discharge due to reduced mutual electric field interference. Collector geometry: The airfoil collector outperforms the ring collector in terms of thrust generation, particularly at higher voltages. It is mainly because the design of the airfoil collector minimizes air resistance, enabling it to generate thrust more efficiently. For instance, at 34 kV, the airfoil collector configuration generates 200 mN/m of thrust, compared to 156 mN/m for the ring collector with the same emitter geometry.
Table 1.
Performance metrics of sawtooth emitter electrodes with different configurations and collectors.
| Electrode configuration (d = 6 cm) | Voltage (kV) | Current (mA/m) | Power (W) | Thrust (mN/m) | Thrust-to-power (N/kW) |
|---|---|---|---|---|---|
|
Sawtooth emitter-ring collector (50 teeth 45 |
20 | 0.2 | 4 | 48 | 12 |
| 22 | 0.28 | 6.16 | 60 | 9.74 | |
| 24 | 0.36 | 8.64 | 78 | 9.03 | |
| 26 | 0.4 | 10.4 | 92 | 8.85 | |
| 28 | 0.52 | 14.56 | 108 | 7.42 | |
| 30 | 0.72 | 21.6 | 124 | 5.74 | |
| 32 | 1 | 32 | 138 | 4.31 | |
| 34 | 1.4 | 47.6 | 156 | 3.28 | |
|
Sawtooth emitter-NACA 0018 airfoil collector (50 teeth 45 |
20 | 0.24 | 4.8 | 52 | 10.8 |
| 22 | 0.32 | 7.04 | 72 | 10.2 | |
| 24 | 0.36 | 8.64 | 88 | 10.18 | |
| 26 | 0.44 | 11.44 | 104 | 9.1 | |
| 28 | 0.52 | 14.56 | 124 | 8.5 | |
| 30 | 0.76 | 22.8 | 147 | 6.45 | |
| 32 | 1 | 32 | 164 | 5.12 | |
| 34 | 1.32 | 44.8 | 200 | 4.46 | |
|
Sawtooth emitter-ring collector (50 teeth 30 |
20 | 0.24 | 4.8 | 50 | 10.42 |
| 22 | 0.28 | 6.16 | 64 | 10.39 | |
| 24 | 0.36 | 8.64 | 80 | 9.26 | |
| 26 | 0.44 | 11.44 | 97.2 | 8.5 | |
| 28 | 0.56 | 15.68 | 116 | 7.4 | |
| 30 | 0.64 | 19.2 | 134 | 6.98 | |
| 32 | 0.76 | 24.32 | 156 | 6.4 | |
| 34 | 1.08 | 36.72 | 172 | 4.68 | |
|
Sawtooth emitter-ring collector (50 teeth 60 |
20 | 0.28 | 5.6 | 40 | 7.1 |
| 22 | 0.36 | 7.92 | 52 | 6.6 | |
| 24 | 0.44 | 10.56 | 60 | 5.7 | |
| 26 | 0.56 | 14.56 | 74 | 5.1 | |
| 28 | 0.72 | 20.16 | 92 | 4.6 | |
| 30 | 0.84 | 25.2 | 108 | 4.3 | |
| 32 | 1.12 | 35.84 | 128 | 3.6 | |
| 34 | 1.32 | 44.8 | 144 | 3.2 | |
|
Sawtooth emitter-ring collector (100 teeth 45 |
20 | 0.2 | 4 | 40 | 10 |
| 22 | 0.24 | 5.28 | 50 | 9.47 | |
| 24 | 0.32 | 7.68 | 66.8 | 8.7 | |
| 26 | 0.44 | 11.44 | 80 | 7 | |
| 28 | 0.56 | 15.68 | 100 | 6.38 | |
| 30 | 0.72 | 21.6 | 124 | 5.74 | |
| 32 | 1.12 | 35.84 | 138 | 3.85 | |
| 34 | 1.6 | 54.4 | 149 | 2.74 | |
|
Sawtooth emitter-ring collector (200 teeth 45 |
20 | 0.12 | 2.4 | 22 | 9.17 |
| 22 | 0.16 | 3.52 | 30 | 8.52 | |
| 24 | 0.24 | 5.76 | 44 | 7.64 | |
| 26 | 0.36 | 9.36 | 58 | 6.2 | |
| 28 | 0.48 | 13.44 | 80 | 5.95 | |
| 30 | 0.68 | 20.4 | 108 | 5.29 | |
| 32 | 1.04 | 33.28 | 134 | 4.03 | |
| 34 | 1.76 | 59.84 | 146 | 2.44 |
The experimental results presented in Table 2 demonstrate the performance of various electrode configurations with different ring spacing (b) and ring width (k) under varying voltage conditions. Effect of ring spacing (b): Increasing the ring spacing (b) from 8 mm to 10 mm generally leads to a decrease in thrust. For example, at 34 kV, the thrust decreases from 148 mN/m (b = 8 mm) to 138 mN/m (b = 10 mm). This is because with fixed outermost ring diameter, smaller spacing results in larger intermediate ring diameters, which in turn increases the effective emitter length when matching single-ring configurations. The extended emitter length directly enhances thrust output by providing more ion generation sites along the sawtooth edges. Effect of ring width (k): Increasing the ring width (k) from 2 mm to 5 mm results in a decrease in thrust. For example, at 34 kV, the thrust decreases from 164 mN/m (k = 2 mm) to 134 mN/m (k = 5 mm). This demonstrates that reduced ring width improves electric-field distribution, thereby enhancing both ion generation and drift efficiency.
Table 2.
Performance metrics of sawtooth emitter electrodes with varying ring spacing (b) and width (k).
| Electrode configuration ( d = 6cm) | Voltage (kV) | Current (mA/m) | Power (W) | Thrust (mN/m) | Thrust-to-power (N/kW) |
|---|---|---|---|---|---|
|
Sawtooth emitter-ring collector (50 teeth 45 |
20 | 0.24 | 4.8 | 44 | 9.2 |
| 22 | 0.28 | 6.16 | 58 | 9.4 | |
| 24 | 0.36 | 8.64 | 72 | 8.3 | |
| 26 | 0.44 | 11.44 | 84 | 7.34 | |
| 28 | 0.52 | 14.56 | 100 | 6.87 | |
| 30 | 0.64 | 19.2 | 120 | 6.25 | |
| 32 | 0.8 | 25.6 | 136 | 5.3 | |
| 34 | 1.04 | 35.36 | 148 | 4.18 | |
|
Sawtooth emitter-ring collector (50 teeth 45 |
20 | 0.24 | 4.8 | 44 | 9.2 |
| 22 | 0.28 | 6.16 | 56 | 9.1 | |
| 24 | 0.36 | 8.64 | 68 | 7.87 | |
| 26 | 0.48 | 12.48 | 82 | 6.57 | |
| 28 | 0.56 | 15.68 | 96 | 6.12 | |
| 30 | 0.68 | 20.4 | 112 | 5.5 | |
| 32 | 0.8 | 25.6 | 128 | 5 | |
| 34 | 0.96 | 32.64 | 142 | 4.35 | |
|
Sawtooth emitter - ring collector (50 teeth 45 |
20 | 0.2 | 4 | 40 | 10 |
| 22 | 0.24 | 5.28 | 52 | 9.85 | |
| 24 | 0.28 | 6.72 | 66 | 9.82 | |
| 26 | 0.36 | 9.36 | 80 | 8.55 | |
| 28 | 0.44 | 12.32 | 92 | 7.47 | |
| 30 | 0.56 | 16.8 | 106 | 6.31 | |
| 32 | 0.68 | 21.76 | 120 | 5.5 | |
| 34 | 0.8 | 27.2 | 138 | 5.07 | |
|
Sawtooth emitter - ring collector (50 teeth 45 |
20 | 0.24 | 4.8 | 52 | 10.83 |
| 22 | 0.32 | 7.04 | 68 | 9.66 | |
| 24 | 0.36 | 8.64 | 82 | 9.49 | |
| 26 | 0.48 | 12.48 | 96 | 7.69 | |
| 28 | 0.56 | 15.68 | 112 | 7.14 | |
| 30 | 0.76 | 22.8 | 124 | 5.44 | |
| 32 | 1.08 | 34.56 | 142 | 4.11 | |
| 34 | 1.48 | 50.32 | 164 | 3.26 | |
|
Sawtooth emitter - ring collector (50 teeth 45 |
20 | 0.24 | 5.6 | 44 | 7.86 |
| 22 | 0.32 | 7.04 | 54 | 7.67 | |
| 24 | 0.4 | 9.6 | 64 | 6.67 | |
| 26 | 0.48 | 12.48 | 78 | 6.25 | |
| 28 | 0.56 | 15.68 | 92 | 5.87 | |
| 30 | 0.68 | 20.4 | 108 | 5.3 | |
| 32 | 0.88 | 28.16 | 124 | 4.4 | |
| 34 | 1.08 | 36.72 | 138 | 3.76 | |
|
Sawtooth emitter - ring collector (50 teeth 45 |
20 | 0.24 | 4.8 | 40 | 8.3 |
| 22 | 0.32 | 7.04 | 52 | 7.4 | |
| 24 | 0.4 | 9.6 | 64 | 6.7 | |
| 26 | 0.44 | 11.44 | 72 | 6.3 | |
| 28 | 0.56 | 15.68 | 92 | 5.87 | |
| 30 | 0.64 | 19.2 | 104 | 5.42 | |
| 32 | 0.8 | 25.6 | 120 | 4.7 | |
| 34 | 1 | 34 | 134 | 3.94 |
The decrease in thrust power efficiency at higher voltages is indeed consistent with the mechanism of ion wind generation. To address this issue, we propose the following mitigation strategies: (1) Limiting the current can effectively reduce power loss while maintaining sufficient thrust. (2) Optimizing the geometric shape of the electrodes can achieve higher electric field strength at lower voltages, thereby reducing the reliance on high voltages. (3) Optimizing operating parameters (voltage, current) can maximize power efficiency while ensuring thrust. For example, identifying the optimal matching point between voltage and current can achieve higher thrust without significantly increasing power loss.
To scale this technology for practical applications such as silent micro-drones, several steps need to be taken to enable an ionic wind-propelled drone to achieve autonomous takeoff. Steps for Self-Sustained Flight: (1) Increase Thrust Density: Further optimization of the electrode geometry, such as refining the sawtooth multi-ring design, could increase thrust density. (2) Reduce Weight of the HV Power Supply: Developing lightweight, high-efficiency HV power supplies is critical. Advances in solid-state transformers and compact HV converters could reduce the weight of the power supply while maintaining the necessary voltage and current output. Increasing the energy density of the power source would allow for longer flight times without significantly increasing weight. (3) Improve Propulsion Efficiency: Enhancing the thrust-to-power ratio is essential for practical applications. Our current design still lacks sufficient thrust-to-power ratio, but further improvements could be made by optimizing the ionization and drift zones. (3) Integration with Drone Design: The ionic wind propulsion system must be integrated into the drone’s aerodynamic design to minimize drag and maximize lift. This could involve embedding the electrodes within the drone’s fuselage. Developing control systems to modulate the thrust output and stabilize the drone during flight is essential. This could involve real-time adjustment of the electrode voltage and gap to control thrust and direction. (3) Prototype Testing and Validation: Building and testing small-scale prototypes is a critical step toward achieving self-sustained flight. These prototypes would allow us to validate the thrust, efficiency, and stability of the propulsion system in real-world conditions. These steps collectively aim to bridge the gap between laboratory-scale experiments and real-world applications, enabling the development of efficient and autonomous ionic wind-propelled drones.
Our experiment was conducted under controlled room conditions, which to some extent limits the applicability of the experimental results in real-world environments. Variations in humidity and temperature can significantly affect the generation and thrust performance of ion wind, as confirmed by multiple studies. For example, research shows that increased humidity leads to a higher concentration of water molecules in the air, which affects ionization efficiency and ion mobility32–34. Additionally, changes in temperature can alter air density and molecular velocity, further influencing the dynamic characteristics of ion wind35,36.
Conclusions
A new type of ionic wind thruster with a sawtooth multi-ring electrode structure was demonstrated in this study. The performance of ionic wind thrusters with three different electrode configurations (patch-type, single-ring-single-ring, single-ring-multi-ring) was evaluated under varying voltages, ring diameters, and electrode gaps. Experimental results showed that the sawtooth multi-ring structure achieved a maximum thrust per unit length of 164 mN/m at 40 kV, with an electrode gap of 60 mm, through further optimization of the structural parameters of the sawtooth multi-ring, the thrust density was significantly enhanced, demonstrating its potential for practical applications.
The ion wind thruster demonstrates the potential to generate sufficient thrust, with the prospect of future applications in low-altitude flight. In subsequent work, the ion wind thrust system can be further optimized through design modifications such as further refinement of the sawtooth multi-ring geometry, altering the electrode microstructure and utilizing multi-stage acceleration37. Alternatively, propulsion efficiency can be enhanced by adding biased electrodes to assist corona discharge and creating separate ionization and drift zones. These improvement methods still need to be validated for their applicability one by one, and the materials used in the propulsion system also deserve further exploration. These aspects merit further investigation in future research.
Acknowledgements
This work has been supported by the National Natural Science Foundation of China (Grant No. 51877111).
Author contributions
Ronghui Quan conceived the experiments, Miaosen Hou and Jifan Zhang conducted the experiments, Miaosen Hou analyzed the results and wrote the article. All authors reviewed the manuscript.
Data availibility
The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.
Declarations
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Hauksbee, F. Physico-mechanical experiments. (1719).
- 2.Chattock, A. Xliv. On the velocity and mass of the ions in the electric wind in air. Lond. Edinb. Dublin Philos. Mag. J. Sci.48, 401–420 (1899).
- 3.Fujishima, H., Ueda, Y., Tomimatsu, K. & Yamamoto, T. Electrohydrodynamics of spiked electrode electrostatic precipitators. J. Electrostat.62, 291–308 (2004). [Google Scholar]
- 4.Yamamoto, T. et al. Electrohydrodynamically assisted electrostatic precipitator for the collection of low-resistivity dust. IEEE Trans. Indus. Appl.45, 2178–2184 (2009). [Google Scholar]
- 5.Wilson, J., Perkins, H. D. & Thompson, W. K. An investigation of ionic wind propulsion (Technical Report, 2009). [Google Scholar]
- 6.Gilmore, C. K. & Barrett, S. R. Electrohydrodynamic thrust density using positive corona-induced ionic winds for in-atmosphere propulsion. Proc. R. Soc. A Math. Phys. Eng. Sci.471, 20140912 (2015). [Google Scholar]
- 7.Monrolin, N., Plouraboué, F. & Praud, O. Electrohydrodynamic thrust for in-atmosphere propulsion. AIAA J.55, 4296–4305 (2017). [Google Scholar]
- 8.Xu, H. et al. Flight of an aeroplane with solid-state propulsion. Nature563, 532–535 (2018). [DOI] [PubMed] [Google Scholar]
- 9.Grosse, S., Moreau, E. & Binder, N. Modeling of the flight performance of a plasma-propelled drone: Limitations and prospects. Drones8, 114 (2024). [Google Scholar]
- 10.Xu, H., Gomez-Vega, N., Agrawal, D. R. & Barrett, S. R. Higher thrust-to-power with large electrode gap spacing electroaerodynamic devices for aircraft propulsion. J. Phys. D Appl. Phys.53, 025202 (2019). [Google Scholar]
- 11.Xu, H., He, Y. & Barrett, S.R. A dielectric barrier discharge ion source increases thrust and efficiency of electroaerodynamic propulsion. Appl. Phys. Lett.114 (2019).
- 12.Young, M., Keith, S. & Pancotti, A. An overview of advanced concepts for near space systems. In 45th AIAA/ASME/SAE/ASEE Joint Propulsion Conference & Exhibit. Vol. 4805 (2009).
- 13.Göksel, B. & Mashek, I.C. First breakthrough for future air-breathing magneto-plasma propulsion systems. J. Phys. Conf. Ser.825, 012005 (IOP Publishing, 2017).
- 14.Chung, C.F. & Li, W.J. Experimental studies and parametric modeling of ionic flyers. In 2007 IEEE/ASME International Conference on Advanced Intelligent Mechatronics. 1–6 (IEEE, 2007).
- 15.Khomich, V. Y. & Rebrov, I. E. In-atmosphere electrohydrodynamic propulsion aircraft with wireless supply onboard. J. Electrostat.95, 1–12 (2018). [Google Scholar]
- 16.Drew, D.S. & Pister, K.S. First takeoff of a flying microrobot with no moving parts. In 2017 International Conference on Manipulation, Automation and Robotics at Small Scales (MARSS). 1–5 (IEEE, 2017).
- 17.Hari Prasad, H. K. et al. A laser-microfabricated electrohydrodynamic thruster for centimeter-scale aerial robots. PloS one15, e0231362 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Ieta, A. & Chirita, M. Electrohydrodynamic propeller for in-atmosphere propulsion; Rotational device first flight. J. Electrostat.100, 103352 (2019). [Google Scholar]
- 19.Cao, W., Mecrow, B. C., Atkinson, G. J., Bennett, J. W. & Atkinson, D. J. Overview of electric motor technologies used for more electric aircraft (MEA). IEEE Trans. Indus. Electron.59, 3523–3531 (2011). [Google Scholar]
- 20.Brelje, B.J. & Martins, J.R. Electric, hybrid, and turboelectric fixed-wing aircraft: A review of concepts, models, and design approaches. Prog. Aerosp. Sci.104, 1–19. 10.1016/j.paerosci.2018.06.004 (2019). [DOI]
- 21.Masuyama, K. & Barrett, S. R. On the performance of electrohydrodynamic propulsion. Proc. R. Soc. A Math. Phys. Eng. Sci.469, 20120623 (2013). [Google Scholar]
- 22.Gomez-Vega, N., Xu, H., Abel, J.M. & Barrett, S. R.H. Performance of decoupled electroaerodynamic thrusters. Appl. Phys. Lett.118, 074101. 10.1063/5.0038208 (2021). https://pubs.aip.org/aip/apl/article-pdf/doi/10.1063/5.0038208/14545397/074101_1_online.pdf.
- 23.Gomez-Vega, N. & Barrett, S. R. Order-of-magnitude improvement in electroaerodynamic thrust density with multistaged ducted thrusters. AIAA J.62, 1342–1353 (2024). [Google Scholar]
- 24.Gomez-Vega, N., Kambhampaty, J.D. & Barrett, S. R.H. Mitigating reverse emission in electroaerodynamic thrusters. J. Phys. D Appl. Phys.55, 505202. 10.1088/1361-6463/ac9c0f (2022).
- 25.Rickard, M. & Dunn-Rankin, D. Numerical simulation of a tubular ion-driven wind generator. J. Electrostat.65, 646–654 (2007). [Google Scholar]
- 26.Khomich, V. Y., Malanichev, V. & Rebrov, I. Electrohydrodynamic thruster for near-space applications. Acta Astronaut.180, 141–148 (2021). [Google Scholar]
- 27.Pekker, L. & Young, M. Model of ideal electrohydrodynamic thruster. J. Propuls. Power27, 786–792 (2011).
- 28.Morrison, R. & Hopstock, D. The distribution of current in wire-to-cylinder corona. J. Electrostat.6, 349–360 (1979). [Google Scholar]
- 29.Weinberg, F. et al. Inducing gas flow and swirl in tubes using ionic wind from corona discharges. Exp. Fluids40, 231–237 (2006). [Google Scholar]
- 30.Stuetzer, O. M. Ion drag pressure generation. J. Appl. Phys.30, 984–994 (1959). [Google Scholar]
- 31.Ny, R., Boulloud, A. & Labbé, J. Influence of negative corona discharges on copper electrodes in room air. (1985).
- 32.Hu, Q. et al. Influence of air pressure and humidity on positive direct current corona discharge performances of the conductor in a corona cage. Int. Trans. Electr. Energy Syst.24, 723–735 (2014). [Google Scholar]
- 33.Zhang, B., He, J. & Ji, Y. Dependence of the average mobility of ions in air with pressure and humidity. IEEE Trans. Dielectr. Electr. Insul.24, 923–929 (2017). [Google Scholar]
- 34.Wang, X. & You, C. Effect of humidity on negative corona discharge of electrostatic precipitators. IEEE Trans. Dielectr. Electr. Insul.20, 1720–1726. 10.1109/TDEI.2013.6633702 (2013).
- 35.Xiao, G. et al. Characteristics of dc discharge in a wire-cylinder configuration at high ambient temperatures. J. Electrostat.72, 13–21 (2014). [Google Scholar]
- 36.Xiao, G. et al. Current analysis of dc negative corona discharge in a wire-cylinder configuration at high ambient temperatures. J. Electrostat.72, 107–119 (2014). [Google Scholar]
- 37.Rebrov, I. Cascade volumetric acceleration of electrohydrodynamic flows. J. Phys. Confer. Ser.1393, 012122 (IOP Publishing, 2019).
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.