Spatial structures of different particles in helicon plasma

doi:10.21203/rs.3.rs-3309639/v1

Download PDF

Research Article

Spatial structures of different particles in helicon plasma

https://doi.org/10.21203/rs.3.rs-3309639/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

The spatial structures of different particles (high-energy electron excited ionic and low-energy electron excited neutral particles) in both discharge and plume plasmas of a helicon source have been characterized. These two different populations show different intensity jumps and growth rates with increasing RF input powers. Filters of 480 nm band pass and 600 nm high pass have been used to distinguish these two populations. Results show that the plasmas are generated from both ends of the antenna and transmitted to the middle of the antenna to form an axial particle path. When the wave mode reaches, the axial particle path is formed. The radial distribution of the high-energy electrons is highly concentrated on the center line while that of the low-energy electrons is relatively uniform. The axial distribution of the high-energy electrons is asymmetric, that is the maximum density is located on the one end of the antenna. The reversed magnetic field could reverse their axial distributions. In the plume plasma, the high-energy electrons are highly directional and expand further axial distances compared with the low-energy electrons. The most probable energies of ions derived from the RFEA remain almost the same with an average value of 45 eV. But the IEDF shapes and the relative ion densities change greatly at 700W. It is believed that the IEDFs and ion density jumps are more likely related to the high-energy electron density jumps.

Due to the advantages of high density, electrodeless design, and multiple operation modes, helicon plasmas are attracting more attention as a thruster or a plasma source in spacecraft propulsion [1–4]. In helicon plasmas, the electron characteristics are significantly important because electrons are the medium for the input powers and the plasmas. The electron dynamics determine the ionization and acceleration of the ions and lead the establishment of plasma potentials, such as the current-free double layers [5–8]. Moreover, the ion characteristics are also very important because ions are the contributors of effective thrust. It is thus of interest to understand the relationship between the electron and ion dynamics in such systems.

Much effort has been made to characterize the electron and ion dynamics. Initial studies suggested that the ion beams were created by formation of a double layer (DL) [9–10]. A DL is a spontaneous freestanding structure that can appear anywhere in a plasma. This region also coincides with a steep drop in the plasma density and in the magnetic field [11]. Recent experiments found evidence that the accelerating structure is not a true DL but an acceleration region with the Debye lengths on the order of 1000 [12–15].

Many groups have reported observations of high energy electrons streaming from upstream into downstream at the same time the ion beams appear. Hot electrons have been observed traveling along magnetic field lines into a magnetic nozzle [16–17]. Two electron temperature populations have been identified within the plasma [18–19]. It was deduced that the downstream electrons came from upstream electrons that have sufficient energy to overcome the potential trap [20]. There are two possible explanations for the energetic electrons: RF skin heating effects in the source and energization by plasma instabilities. The RF skin heating effects hypothesis suggests that the ions are accelerated because energetic electrons were created at the edge of the plasma, and then stream downstream and through the expansion regions with few collisions [17]. A spatially distinct region of ion acceleration must arise through ambipolar fields inside of the annulus of energetic electrons.

Based on the recent research described above, many studies focused solely on the plasma upstream or downstream of the acceleration region [21–22]. There is a gap in understanding exactly how the energetic electrons are produced and how they influence/create the ion beam. To explore the relationship of the energetic electrons and ion acceleration in expanding plasmas, we present spatial structures of the high- and low-energy electrons both in upstream and downstream plasmas and investigate the relationship between electrons and ions. The direct images show that the electrons were generated from the side ends of the antenna rather than from the edge of the discharge tube. The images also show that the axial distribution of the high-energy electrons is asymmetric. The maximum density is located on the one end of the antenna after the wave mode is formed. The reversed magnetic field could reverse the axial distributions. This axial asymmetric of electron density could cause the ion acceleration in expanding plasmas.

In this paper, the experiments were performed in a 1.5-m long and 1.0-m inner diameter stainless-steel vacuum chamber, as shown in Fig. 1. The discharge chamber (tube) consisted of a 4 cm diameter and 40 cm long quartz tube. The quartz tube was surrounded by a water-cooled 15.5 cm long helical antenna. A 13.56 MHz RF source was connected to this helicon antenna. A uniform, axial magnetic field of 0–1200G (with equal current of 0–70 A) in the plasma source was produced by two water cooled electromagnets. The argon mass flow rate was 20 standard cubic centimeter per minute (sccm) during the discharge. The vacuum system was pumped down to a base pressure of 5×10^–3 Pa by a turbo-molecular-rotary pump which was connected to the side of the diffusion chamber downstream. The working pressure was about 5×10^–2 Pa (~ 0.4 mTorr) and it was measured using an ionization gauge.

An optical emission spectroscope (OES) (AvaSpec-ULS2048CL-8-EVO) with wavelengths ranging from 200 nm to 1100 nm was used for the optical experiments. The spectroscope has 8 channels, and the average spectral resolution was 0.1 nm. An intensified charge coupled device (ICCD) (DH334T-18U-E3) camera with 1024×1024 pixel array was used to picture the spatial structure of the discharge plasma. This camera was available to match wavelength range requirements from 120 nm to 1100 nm. Along with the camera, the 480 nm band pass, the 810 nm band pass (with 10 nm full width at half maxima, FWHM), and the 600 nm high pass filters were used in the experiment to transmit the required light and to reject all other unwanted radiation.

A four-grid retarding field energy analyser (RFEA) was positioned on the main axis at the position of 10 cm from the exit plane of the discharge tube [23–25]. The whole assembly was 1.8 cm in length, 1.4 cm in width, and 0.7 cm in height. The four grids consisted of a (in order) floating, negatively biased electron repelling, positively biased ion retarding, and secondary electron suppression, all before a collection plate. The electron repelling grid was maintained at -40 V for nominal test cases while the ion retarding grid was swept from 0-150V. Its aperture hole about 0.3 cm faced the plasma source to measure the current I versus discriminator voltage (V_d) characteristic and obtain its derivative, the ion energy distribution function (IEDF).

A. OES emission spectra

Figure 2 shows the emission lines with their electron excitation energy thresholds of an argon RF discharge. The OES was set outside the discharge tube and was facing the centerline of the tube. Wavelengths of emission lines and their excitation energies were compared with the NIST Atomic Spectra Database Lines in order to meet the international standards (https://physics.nist.gov/PhysRefData/ASD/lines_form.html).

As shown in Fig. 2, all the emission lines with wavelengths ranging from 200 nm to 1100 nm were divided into two different regions by their different excitation energy thresholds. Firstly, the emission lines with wavelengths ranging from 200 nm to 600 nm are mainly high-energy electron excited ionic (Ar II) lines (≥ 19 eV). Meanwhile, the emission lines with wavelengths ranging from 600 nm to 1100 nm are mainly low-energy electron excited neutral lines (Ar I) lines (mainly 13–15 eV). Four lines with the highest signal-to-noise-ratio (SNR) in one emission spectrum were used to establish a comparison with different working conditions. These are 434.8 nm, 480.6 nm, 763.5 nm, and 811.5 nm lines. The lines of 434.8 nm and 480.6 nm are high-energy electron excited ionic lines. The lines of 763.5 nm and 811.5 nm are low-energy electron excited neutral lines. Their spectra details are listed in Tables 1, in which wavelengths, ion types, transition probabilities (A_ki), statistical weights (g_k), excited energy levels (E_i, E_k), and the electron configurations of the emission lines have been included.

Table 1

Spectra details of the 434.8 nm, 480.6 nm, 763.5 nm, and 811.5 nm lines.
No.	Ion	Wavelength (nm)	A_kig_k (10⁸ s^–1)	E_i (eV)	E_k (eV)	Lower level (conf., Term, J)	Upper level (conf., Term, J)
1	Ar II	434.8	9.36	16.644	19.494	3s²3p⁴(³P)4s, ⁴P, 5/2	3s²3p⁴(³p)4p, ⁴D⁰,7/2
2	Ar II	480.6	4.68	16.644	19.223	3s²3p⁴(³P)4s, ⁴P, 5/2	3s²3p⁴(³p)4p, ⁴P⁰,5/2
3	Ar I	763.5	1.23	11.548	13.172	3s²3p⁵(²P⁰_3/2)4s, ²[3/2]⁰, 2	3s²3p⁵(²P⁰_3/2)4p, ²[3/2], 2
4	Ar I	811.5	2.30	11.548	13.075	3s²3p⁵(²P⁰_3/2)4s, ²[3/2]⁰, 2	3s²3p⁵(²P⁰_3/2)4p, ²[5/2], 3

B. Different intensity jumps of ionic and neutral lines

Figure 3 shows the emission spectra with different input RF powers (200–1200 W) under different magnetic field strengths. Figure 3 (a) shows the emission spectra with a central magnetic field strength of 800 G. Figure 3 (b) shows the emission spectra with a central magnetic field strength of 1000 G. From these figures, we can see that the intensity growth rate of the high-energy electron excited ionic population (434.8 nm and 480.6 nm lines) is larger than that of the low-energy electron excited neutral population (763.5 nm and 811.5 nm lines) with the increase of input powers. This phenomenon is more obvious with 1000 G condition. The intensity of the ionic population is higher than the neutral population at 1000 W and 1200W.

Figure 4 shows the emission intensity variations with different input powers of the 434.8 nm, 480.6 nm, 763.5 nm, and 811.5 nm lines under different magnetic field conditions (800G, 1000G, and 1200G). Figure 5 shows the normalized emission intensities of these four lines. These values are normalized by taking the data of 200 W as reference. The results show that the low-energy electron excited neutral populations (763.5 nm and 811.5 nm lines) jump at 400 W and reach the helicon wave mode, and then become to a nearly saturation level. However, the high-energy electron excited ionic populations (434.8 nm and 480.6 nm lines) jump firstly at 400 W and reach a wave mode (W1 mode). After that, they jump for the second or/and the third time (W2 or W3 mode). Instead of reach to a saturation level, their intensity increase almost linearly with the increasing input power. Moreover, the applied magnetic field strengths have an important influence on their relative emission intensity. The relative intensities of the low-energy electron excited neutral lines (763.5 nm and 811.5 nm lines) decrease significantly with the increase of magnetic field strengths, while that of the high-energy electron excited ionic lines (434.8 nm and 480.6 nm lines) increase slightly with the increase of magnetic field strengths.

C. ICCD images with different filters

1. Energy coupling with different RF input powers

From the above emission spectra, it can be seen that there are mainly two different particles in the Helicon discharge plasma. These are the high-energy electron excited ionic lines (434.8 nm and 480.6 nm) and the low-energy electron excited neutral lines (763.5 nm and 811.5 nm). Moreover, these two parts show different intensity jumps with the increase of RF input power. This phenomenon may allow us to have a deeper understanding of the energy coupling between RF input power and plasma. In order to distinguish these two different particles, two different filters (480 nm band pass and 600 nm high pass) were used for ICCD imaging experiments. Originally, 810 nm band pass filter was intended to be used, but there was large fringe noise in the image. Finally, the 600 nm high pass filter was used for a high-quality image.

Figure 6 shows the spatial structure of the high-energy electron excited ionic and low-energy electron excited neutral particles with different RF input powers. For high-energy electron excited ionic particles (480 nm band pass), they are firstly generated at both ends of the antenna and their radial distribution is relatively uniform, see the 200-W case. With the increase of input powers (300 W), they expand to the middle of the antenna to form an axial particle path. After the wave mode has been reached (500 W), their axial intensity distribution is asymmetric. The maximum axial intensity locates at one end of the antenna. In addition, their radial distribution concentrates towards the center dramatically, and the maximum radial intensity locates on the central axis of the discharge tube.

As for the low-energy electron excited neutral particles (600 nm high pass), they are also firstly generated at both ends of the antenna. They expand to the middle of the antenna to form an axial particle path with the increase of input powers. After the wave mode has been reached, the axial particle path is formed. But the higher axial intensities locate at both ends of the antenna. The radial distribution concentrates towards the center slightly.

2. Plume plasma properties

Figure 7 shows plume plasma structure of the ionic (480 nm band pass) and neutral particles (600 nm high pass) with different RF input powers. The results show that the high-energy electron excited ionic particles are highly directional compared with that of the low-energy electron excited neutral particles. The ionic particles could reach longer axial distances than the neutral ones.

3. Reversed magnetic field

Figure 8 shows the effects of reversed magnetic field on the plasma structure. Both the intensity distributions of the high-energy electron excited ionic and low-energy electron excited neutral particles are axially reversed. Moreover, the radial intensity distributions are more concentrate towards the centerline with the negative magnetic field.

For electric propulsion, ion energy distribution is more important because ions are the contributors of effective thrust. The results described above show the different spatial structures and intensity jumps of the high-energy electron excited ionic and the low-energy electron excited neutral particles. Therefore, we want to establish a relationship between the ion energy distribution and the relatively high- and low-energy electrons.

Figure 9 shows the normalized IEDFs from the RFEA tests with the RF input power from 200 W to 1200 W. They were normalized as their integral area to be 1. The IEDFs from 200 W to 600 W are lower and wider. And the IEDFs from 700 W to 1200 W are more concentrated. Figure 10 shows the most probable energies of ions with the increasing RF input powers. They almost remain the same with an average value of 45 eV. Figure 11 shows the relative ion density jumps from the RFEA and the emission intensity jumps from the OES. It shows that the relative ion density jumps at 700 W. Moreover, the IEDF shapes also change greatly at 700W. Therefore, it is believed that the IEDF jump is more likely related to the jump of high-energy electrons.

In this paper, an OES, filters, ICCD camera, and RFEA have been used in order to distinguish the particles with different energies, obtain their spatial structure, and establish a relationship between them. The emission spectrum of the helicon discharge plasma has two sections. One is the high-energy electron excited ionic lines and the other one is low-energy electron excited neutral lines. Their intensity jumps and growth rates show significant difference with the increase of the RF input powers. Therefore, we use two filters and the ICCD camera to obtain the spatial images of the high- and low-energy electrons. The discharge plasma and the plume plasma structures have been studied and the effects of the reversed magnetic field on the discharge plasma have also been investigated.

The plasmas are generated from both ends of the antenna and transmitted to the middle of the antenna to form an axial particle path. When the wave mode is reached, the axial particle path is formed. When the wave mode is reached, the radial distribution of the high-energy electrons is concentrated on the center line and the axial distribution of them is asymmetric. The maximum density is located on the one end of the antenna. The distribution of the low-energy electrons is relatively more uniform. The axial plasma structure almost reversed with a reversed applied magnetic field.

The most probable energies of ions remain almost the same with an average value of 45 eV with no other acceleration method. However, the IEDF shapes change greatly at 700W and the relative ion density jumps also at 700 W. This means that the IEDF and ion density jumps are more likely related to the high-energy electron density jumps.

Competing interests

This work has no competing interests of a financial or personal nature with others.

Authors' contributions

Zun Zhang: Writing and editing.

Jikun Zhang: Resources, Methodology, Investigation.

Yuzhe Sun: Data processing and picture drawing.

Funding

This work was supported by National Natural Science Foundation of China (No. 11805011).

Availability of data and materials

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Chen FF (1997) Plasma Sources Sci Technol 24:014001
Charles C (2009) J Phys D Appl Phys 42:163001
Takahashi K et al (2014) Plasma Sources Sci Technol 23:044004
Zhao G et al (2018) Plasma Sci Technol 20:075402
Takahashi K et al (2017) Phys Plasmas 24:084503
Ghosh S et al (2015) Plasma Sources Sci Technol 24:034011
Doyle SJ (2018) et al Plasma Sources Sci Technol 27:085011
Takahashi K et al (2009) Appl Phys Lett 94:191503
Charles C, Boswell R (2003) Appl Phys Lett 82:1356–1358
Sun X, Biloiu C, Hardin R, Scime EE (2004) Plasma Sources Sci Technol 13:359–370
Chen F (2006) Phys Plasmas 13:034502
Aguirre EM, Thompson DS, Scime EE, Good TN (2017) Phys Plasmas 24:123510
Bennet A, Charles C, Boswell R (2018) Phys Plasmas 25:023516
Zhang X, Aguirre EM, Thompson DS, McKee J, Henriquez M, Scime EE (2018) Phys Plasmas 25:023503
Aguirre EM, Bodin R, Yin N, Good TN, Scime EE (2020) Phys Plasmas 27:123501
Takahashi K, Akahoshi H, Charles C, Boswell RW, Ando A (2017) Phys Plasmas 24:084503
Aguirre EM, Scime EE, Thompson DS, Good TN (2017) Phys Plasmas 24:123510
Sung YT, Li Y, Scharer JE (2016) Phys Plasmas 23:092113
Sung YT, Li Y, Scharer JE (2015) Phys Plasmas 22:034503
Takahashi K, Charles C, Boswell RW (2007) Phys Plasmas 14:114503
Foster JE, Gallimore AD (1997) J Appl Phys 81:3422
Kortshagen U, Pukropski I, Zethoff M (1994) J Appl Phys 76:2048
Hopwood J (1993) Appl Phys Lett 62:940
Charles C, Boswell R (2003) Appl Phys Lett 82:1356
Ichihara D, Uchigashima A, Iwakawa A, Sasoh A (2016) Appl Phys Lett 109:053901

No competing interests reported.

Download PDF

Editorial decision: Revision requested
04 Dec, 2023
Reviews received at journal
29 Sep, 2023
Reviewers agreed at journal
19 Sep, 2023
Reviewers invited by journal
18 Sep, 2023
Editor assigned by journal
01 Sep, 2023
Submission checks completed at journal
01 Sep, 2023
First submitted to journal
30 Aug, 2023

You are reading this latest preprint version

Spatial structures of different particles in helicon plasma

Status:

Version 1

Abstract

Figures

I. Introduction

II. Experiment apparatus

III. Results

IV. Discussion

V. Conclusion

Declarations

References

Additional Declarations

Status:

Version 1