Observation of Electron Temperature Bursts in the Edge Region of RT-1 Dipole Plasma

Abstract

A triple Langmuir probe was applied to the edge plasma of RT-1 to perform simultaneous measurements of electron temperature T_e and electron density n_e. In addition to relatively slow fluctuations commonly observed in RT-1 plasmas, abrupt T_e enhancement events accompanied by a decrease in n_e were observed. Since fluctuations play an important role in radial transport and structure formation in dipole plasmas, the observed intermittent events are expected to provide useful insight into fluctuation phenomena in the edge plasma.

The properties of edge plasmas have important effects on a wide range of plasma phenomena, including confinement, transport, and plasma-wall interactions. In dipole magnetic fields, radial transport driven by low-frequency fluctuations in the edge region has been observed to self-organize plasma structures peaked in the strong-field region [1, 2]. Simultaneous measurements of fluctuations in the electron temperature $T_{\mathrm{e}}$ and electron density $n_{\mathrm{e}}$ are essential for understanding fluctuation-driven transport processes. In particular, separating the temperature and density components of fluctuations with high temporal resolution is important for identifying the nature of low-frequency fluctuations and distinguishing them from intermittent events occurring on different time scales. Intermittent phenomena have been reported in ECR plasmas in linear devices [3, 4]. In this study, we focus on local fluctuation phenomena [5, 6] associated with structure formation and performed edge plasma measurements in RT-1 (Ring Trap 1) [2] using a triple Langmuir probe [7]. We report intermittent electron-temperature enhancement events observed through simultaneous measurements of $T_{\mathrm{e}}$ and $n_{\mathrm{e}}$.

RT-1 is a plasma confinement device that produces a dipole magnetic field using a magnetically levitated superconducting coil, and studies are being carried out toward the realization of advanced fusion [1] and antimatter plasmas [8]. As shown in Fig. 1, a radially movable triple Langmuir probe was installed in the edge region to independently measure $T_{\mathrm{e}}$ and $n_{\mathrm{e}}$ with high temporal resolution. The triple probe was operated using the standard method [7]. As shown in the figure, the temporal evolution of $T_{\mathrm{e}}$ and $n_{\mathrm{e}}$ was obtained from the relationship among the three electrode currents $I_1$, $I_2$, and $I_3$ using two power supplies. The probe electrodes were made of tungsten wires with a diameter of 1 mm. The wires were inserted into a four-bore ceramic insulating tube (SSA-S) with an inner diameter of $\varphi$ 1.2 mm, leaving the 5 mm tips exposed to the plasma. The separation between probe tips was 3 mm, and the minimum surface-to-surface distance was about 2 mm. For typical edge parameters, $T_{\mathrm{e}}$ ≃ 3 eV and $n_{\mathrm{e}}$ ≃ 1.5 × 10¹⁵ m⁻³, the Debye length is estimated to be ${\lambda }_{\mathrm{D}}$ ≃ 0.33 mm. The probe spacing therefore corresponds to approximately 6–9 ${\lambda }_{\mathrm{D}}$. Although this distance is comparable to the estimated sheath thickness and the absolute values of $T_{\mathrm{e}}$ and $n_{\mathrm{e}}$ may include systematic uncertainty, the present analysis focuses mainly on reproducible relative temporal variations of the triple probe signals rather than on the absolute calibration of $T_{\mathrm{e}}$ and $n_{\mathrm{e}}$. To reduce electrostatic noise and provide mechanical support, the lower part of the ceramic tube was covered with a stainless-steel tube. At the other end of the ceramic tube, the tungsten wires were connected to Junflon wires through thin nickel plates by spot welding, and the electrical signals were transmitted to a digitizer outside the vacuum vessel through current feedthroughs. The probe could be moved radially using a probe drive system, allowing insertion up to 80 mm from the wall and measurements over the range $R$ = 920–1,000 mm. The plasma was produced by electron cyclotron resonance heating (ECH) using 2.45 GHz microwaves.

Fig. 1.  (a) Top and cross-sectional views of RT-1, (b) schematic and picture of triple Langmuir probe structure, and (c) its electric circuit.

Figure 2(a) shows the temporal evolution of (a-1) $T_{\mathrm{e}}$ and (a-2) $n_{\mathrm{e}}$ measured with the triple probe at $R$ = 920 mm, slightly inside the separatrix. The discharge was produced in helium at 1.8 mPa with 2 kW heating. In the $T_{\mathrm{e}}$ signal, intermittent burst-like increases in the electron temperature were repeatedly observed, superposed on low-frequency fluctuations around 1 kHz commonly observed in both $T_{\mathrm{e}}$ and $n_{\mathrm{e}}$ in RT-1 plasmas [6]. These low-frequency fluctuations are not clearly resolved in Fig. 2(a) because of the time and amplitude scales of the plot. As shown in the enlarged view, the typical duration of the electron-temperature bursts was several milliseconds, and in many cases these bursts were accompanied by a decrease in the electron density. Figures 2(b) and (c) show the histograms of the peak values of $T_{\mathrm{e}}$ and $n_{\mathrm{e}}$ extracted from the time-series data in Fig. 2(a). The $T_{\mathrm{e}}$ histogram shows a tail component corresponding to intermittent bursts, which is not observed in the $n_{\mathrm{e}}$ histogram. The occurrence frequency of the bursts tended to increase with increasing microwave heating power. Under the present experimental conditions, no distinct magnetic fluctuation synchronized with the electron-temperature bursts was detected.

Fig. 2.  (a) Temporal evolution of (1) electron temperature $T_{\mathrm{e}}$ and (2) density $n_{\mathrm{e}}$. Histograms of peak values extracted from the time-series data of (b) $T_{\mathrm{e}}$ and (c) $n_{\mathrm{e}}$.

Figure 3 compares the fluctuating components of $T_{\mathrm{e}}$ and $n_{\mathrm{e}}$, defined as deviations from their time-averaged steady components. The overall trend of the plot is downward, indicating an inverse correlation between $T_{\mathrm{e}}$ and $n_{\mathrm{e}}$. However, the behavior of the electron-temperature bursts, indicated by the green ellipse, deviates from the linear fit obtained after excluding the tail component. This result cannot be explained solely by a simple high-temperature, low-density structure with approximately constant pressure, and may suggest a contribution from a locally heated high-temperature component.

Fig. 3.  Correlation between the deviations of electron temperature and density from their time-averaged values. The blue line is a linear fit excluding the burst component, and the green ellipse highlights the burst events.

In summary, by separately measuring the electron temperature and density in the edge plasma of RT-1 using a triple Langmuir probe, we found intermittent changes in plasma parameters. These events were characterized mainly by burst-like increases in the electron temperature accompanied by decreases in the electron density. Clarifying the physical mechanism responsible for these events and evaluating their influence on plasma confinement remain subjects for future work.

This work was supported by JST Moonshot R&D Program No. JPMJMS25A6, JSPS KAKENHI No. JP26K22304, and NIFS Collaboration Research Program Nos. NIFS26KFGS002 and NIFS25KIPS009.

References

• [1]   A.  Hasegawa, Comments Plasma Phys. Control. Fusion 11, 147 (1987).
• [2]   Z.  Yoshida et al., Plasma Fusion Res. 1, 008 (2006).
• [3]   K.  Terasaka et al., Phys. Plasmas 25, 052113 (2018).
• [4]   Y.  Yamada et al., Plasma Fusion Res. 20, 1201027 (2025).
• [5]   H.  Saitoh et al., Nat. Commun. 15, 861 (2024).
• [6]   H.  Saitoh et al., Nucl. Fusion 66, 096013 (2026).
• [7]   S.-L.  Chen and  T.  Sekiguchi, J. Appl. Phys. 36, 2363 (1965).
• [8]   M.R.  Stoneking et al., J. Plasma Phys. 86, 155860601 (2020).