Since 18 of December 2019 conferences.iaea.org uses Nucleus credentials. Visit our help pages for information on how to Register and Sign-in using Nucleus.

10–15 May 2021
Virtual Event
Europe/Vienna timezone
The Conference will be held virtually from 10-15 May 2021

Plasma current ramp-up with 28 GHz second harmonic electron cyclotron wave in the QUEST spherical tokamak

14 May 2021, 08:30
4h
Virtual Event

Virtual Event

Regular Poster Magnetic Fusion Experiments P7 Posters 7

Speaker

Takumi Onchi (Kyushu University)

Description

Plasma current ramp-up experiment with 28 GHz second harmonic electron cyclotron (EC) wave has been in progress in the QUEST spherical tokamak. Through oblique injection of the EC wave, the observation presents that (i) the highest level of plasma current $I_{p}$ as non-inductive EC ramp-up is attained with almost zero loop voltage, $V_{loop} \sim$ 0, by long-pulse radio frequency (RF) injection for 1.3 s. Furthermore, (ii) $I_{p}$ reaches 100 kA with $V_{loop}$ < 0.5 V. Relativistic Doppler shift performs the key role to drive plasma current when parallel (to toroidal magnetic field) refractive index of the incident beam $N_{//}$ is large. As a result of the beam steering control and finely adjusted gas fueling, (iii) bulk electrons are heated efficiently and the electron temperature reaches $T_{e}$ > 500 eV at the density $n_{e}$ $\sim$ 1 x 10$^{18}$ m$^{-3}$ with the incident RF power of no more than 150 kW. Obtained results indicate both of high $I_{p}$ and high $n_{e}T_{e}$ can be attained, on condition that dual beams with large and small $N_{//}$ are available, through control of a resonant velocity-region.

High current through non-inductive EC ramp-up has been explored in QUEST whose major and minor radii of the device are $R_{0}$ = 0.68 m and $a_{0}$ = 0.40 m, respectively [1,2]. The world highest EC ramp-up current of $I_{p}$ > 80 kA was achieved solely by RF injection with the power of 230 kW for 1.2 s even without the polarization control [3]. As a next step, a new experiment has been conducted by RF injection with relatively low-power (150 kW) and long-pulse (maximum: 1.6 s) to ramp up $I_{p}$ effciently as $V_{loop}$ $\sim$ 0. The steering antenna can adjust the incident angle to vary $N_{//}$. To examine the effect of local EC heating on the non-inductive current ramp-up, the beam is sharply-focused as the power density reaches 20 MW/m$^{2}$ and its polarization is optimized as extra-ordinary: X-mode.

Obliquely injecting 28 GHz focused beam polarized as X-mode, the world highest level of plasma current through non-inductive EC ramp-up has been attained efficiently in QUEST. As specific examples, the waveforms are presented in Fig. 1, where $N_{//}$ = 0.75 at $R$ = 0.32 m. $I_{p}$ is ramped up with almost zero loop voltage by means of a slow ramp-up of vertical field for 1.3 s in #40811. Hard X-ray (HXR) emitted from energetic electrons around $R$ = 0.32 m was observed, as the HXR count in 50-100 keV range increases with $I_{p}$ before $t$ = 3.0 s. Even with low loop voltage $V_{loop}$ < 0.5 V after non-inductive ramp-up, plasma current is driven further through ohmic heating, and it reaches $I_{p}$ > 100 kA in #40527. Since $I_{p}$ has not been saturated as shown in Fig. 1, it would increase with longer gyrotron pulse. The plasma size is relatively large after $I_{p}$ is ramped up (0.23 m < $R$ < 0.90 m). Even though obtained bulk electron temperature so far is as low as $T_{e}$ < 100 eV, such acquirement of high current and high density would connect to the next heating scenario, for instance, using neutral beam injection.

The symmetry of current drive to co- and counter-directions is broken strongly due to the presence of energetic electrons. It is likely that one of the mechanisms to obtain high $I_{p}$ in QUEST is such symmetry breaking. Figure 2 shows radial dependence of the electron velocity pitch ($v_{//}$ / $v_{\perp}$), satisfying the condition of harmonic resonances (2nd-4th), where $N_{//}$ = 0.75 at $R$ = 0.32 m. $v_{//}$ and $v_{\perp}$ are parallel and perpendicular velocities to magnetic field, respectively. Here, $v_{//}$ is positive/negative when the electron moves in the co/counter toroidal field direction. The current would be generated not only in the co-direction due to the up-shifted resonances but also in the counter direction due to the down-shifted. Increasing the energy of energetic electrons to a typical value $T_{ee}$= 50 keV observed in the experiment, compared to $T_{ee}$= 1 keV, zero crossing points of the curves are shifted inward to the high field side. Owing to large Doppler shift, power of the wave launched from the low field side is absorbed in wide area. As presented in Fig. 2, the area of $n$th upshifted resonance overlaps with that of ($n$ + 1)th down-shifted resonance. The absorption of $n$th harmonic resonance is higher than that of ($n$ + 1)th resonance. Furthermore, around the central post, the strong relativistic effect cuts out the down-shifted 2nd harmonic resonance as presented in Fig. 2. Hence, up-shifted resonance can be superior to down-shifted resonance over wide range in plasma.
The calculation through the ray tracing code TASK/WR presents the RF power is efficiently absorbed mostly into energetic electrons. The single-pass absorption is estimated as 28 $\%$ when $N_{//}$ = 0.75 at $R$ = 0.32 m and the ratio of energetic electrons ($T_{ee}$= 50 keV) is 3 $\%$. About 1/3 of the absorption tends to occur in wide space with the effect of 3rd and 4th resonances due to the relativistic Doppler shift.

Effective bulk electron heating is attempted controlling resonant velocity-region by incident $N_{//}$. Injecting EC wave normal to toroidal field, wave absorption through Doppler shift effect is less significant to energetic electrons. Owing to the beam steering control and finely adjusted gas injections, the number of energetic electrons is reduced even during the EC plasma ramp-up, where $I_{p}$ < 30 kA. Whereas, the size of closed flux surface would be small (0.23 m < $R$ < $\sim$0.60 m) when $I_{p}$ is low. The waveforms are presented in Fig.3, where $N_{//}$ = 0.11 at $R$ = 0.32 m. $I_{p}$ is ramped up to < 30 kA with $V_{loop}$ < 0.1 V. HXR, that the energy range is 50-100 keV, increases with $I_{p}$ ramp-up. Even though the HXR count is much lower than that of the high-$I_{p}$ case, it is likely that energetic electrons work for the ramp-up of current. Multi-pass power absorption through reflections may be also related to the current maintenance.
$T_{e}$ increases gradually and exceeds 500 eV (at $R$ = 0.34 m) around the 2nd resonance layer when $n_{e}$ = 1 x 10$^{18}$ m $^{-3}$at $t$ = 2.7 s. Assuming $T_{e}$ = 500 eV, the RF power absorption arising only at the 2nd resonance layer is calculated as 7 $\%$ by TASK/WR. Heating power density per volume around the 2nd resonance layer is estimated as 0.4 MWm$^{-3}$ with incident RF power of 150 kW at 7 $\%$ absorption. As a result, significant bulk heating is achieved with control of the incident $N_{//}$ in small but effectual closed flux configuration.

[1] H. Idei, et al., Nucl. Fusion 57 (2017) 126045.
[2] H. Idei, et al., Nucl. Fusion 60 (2020) 016030.
[3] H. Idei, et al., Fusion Eng. Des. 146 (2019) 1149-1152.

Waveforms of (a) RF power, (b) plasma current, (c) loop voltage, (d) line-integrated density, and (e) count of hard x-ray (50-100 keV), where $N_{//}$ = 0.75 at $R$ = 0.32 m.

Radial profile of the velocity pitch ($v_{//}$ / $v_{\perp}$) of 1 keV (gray) and 50 keV (red) electrons satisfying $n$th harmonic relativistic Doppler resonance conditions. The inner limiter is located at $R$ = 0.23 m.

Waveforms of (a) plasma current and loop voltage, (b) line-integrated electron density and H$\alpha$ emission, (c) count of HXR (50-100 keV), and (e) electron temperature and density at $R$ = 0.34 m measured through Thomson scattering. EC wave was injected as $R$ = 0.11 at $R$ = 0.32 m.

Affiliation Kyushu University
Country or International Organization Japan

Primary author

Takumi Onchi (Kyushu University)

Co-authors

Dr Hiroshi Idei (Kyushu Univ.,Japan) Masaharu Fukuyama (Kyushu University) Daichi Ogata (Kyushu University) Tsuyoshi Kariya (Plasma Research Center, University of Tsukuba) Akira Ejiri (Graduate School of Frontier Sciences, The University of Tokyo) Mr Kyohei Matsuzaki (The University of Tokyo) Mr Yuki Osawa (The University of Tokyo) Mr Yi Peng (The University of Tokyo) Mr Ryuichi Ashida (Kyushu University) Mr Shinichiro Kojima (Kyusyu University) Dr Kengo Kuroda (Kyushu Univ.,Japan) Dr Makoto Hasegawa (Kyushu Univ.,Japan) Ryuya Ikezoe (Kyushu Univ.,Japan) Takeshi Ido (Kyushu Univ., Japan) Kazuaki Hanada (Advanced Fusion Research Center, Research Institute for Applied Mechanics, Kyushu University) Aki Higashijima (Kyushu Univ., Japan) Takahiro Nagata (Kyushu Univ., Japan) Shun Shimabukuro (Kyushu Univ., Japan) Mr Ichiro Niiya (Kyushu University) Prof. Kazuo Nakamura (IGSES, Kyushu University, Fukuoka, Japan) Nicola Bertelli (Princeton Plasma Physics Laboratory) Masayuki Ono (PPPL/Princeton University) Yuichi Takase (University of Tokyo) Atsushi Fukuyama (Kyoto University) Sadayoshi Murakami (Departement Nuclear Engineering, Kyoto University)

Presentation materials