Ion measurement by millimeter and sub-millimeter in plasma
In high-temperature plasmas, it is not easy to diagnose the state of ions and fast ions because they cannot be measured directly by a thermometer. As a non-contact measurement method to solve this problem, we are developing a Collective Thomson Scattering (CTS) diagnostic using electromagnetic scattering phenomena. In addition to the 77 GHz and 154 GHz bands, we have developed the world's first submillimeter-wave scattering measurement system in the 300 GHz band (in collaboration with Fukui Univ.).
Right: Results of velocity space tomography reconstruction by deep learning (reconstruction results). Measurement signal is observed as quantity projected onto measurement line of sight (two-line measurement). Reconstruction result obtained from two sightline measurements reproduces velocity distribution of original ions well.
In high-temperature plasmas, it is not easy to diagnose the state of ions and fast ions because they cannot be measured directly by a thermometer. As a non-contact measurement method to solve this problem, we are developing a Collective Thomson Scattering (CTS) diagnostic using electromagnetic scattering phenomena. In addition to the 77 GHz and 154 GHz bands, we have developed the world's first submillimeter-wave scattering measurement system in the 300 GHz band (in collaboration with Fukui Univ.).
Signal analysis has been performed for the anisotropic velocity distribution of fast ions. This study has succeeded in tomography on velocity space (velocity space tomography) using deep learning. As a result, it was newly found that it is possible to obtain anisotropic ion velocity distributions. This achievement contributes to the fast-ion diagnostic method, investigating the state of α-particles produced in fusion plasmas.
Fusion plasmas generate energy by sustaining the plasma burning with alpha particles produced by the fusion reaction between deuterium and tritium in the plasma. Therefore, it is necessary to know the state of the energetic α-particles in the high-temperature plasma. Since it is challenging to measure α-particles in high-temperature plasmas, a non-contact method, Cooperative Thomson Scattering (CTS), has been developed using electromagnetic scattering phenomena.
The research team studied a method to evaluate the ion temperature and velocity distribution from the scattering phenomena of millimeter/submillimeter-wave beams by the plasma. In addition to the 77 GHz and 154 GHz bands, the team has been preparing for CTS measurement in a 300 GHz band that enables CTS diagnostics at high-density. The Far-Infrared Center of the University of Fukui developed the 300-GHz gyrotron and a receiver system. The receiver system measures the thermal radiation from electrons in the 300-GHz band successfully. Next, we plan to conduct experiments to detect the scattered signals.
The measured scattered signal contains the plasma parameters at the intersection position of the incident and received beams. However, the scattered radiation signal becomes a projection on the line-of-sight direction in velocity space. Therefore, it was necessary to reconstruct the ion distribution in velocity space parallel and perpendicular to the magnetic field, using a tomography technique. Consequently, we proposed a new reconstruction method using deep learning. As a result, for the first time in the world, we have succeeded in reconstructing a two-dimensional distribution from a one-dimensional measurement signal, by learning the reconstruction rules from the model distribution to the measurement signal, and from the measurement signal to the original distribution.
This achievement brings us one step closer to measuring the velocity space of fast ions and α-particles produced in plasmas. In addition, we expect that this method will be applied to measurements in a wide variety of research fields, because even if a conventional inverse transformation is impossible and a forward transformation is possible, an inverse transformation is also possible.
This research was carried out in a collaboration led by Masaki Nishiura at the National Institute for Fusion Science and Professor Teruo Saito at the Far-Infrared Research Center, University of Fukui.
This research result was published in Journal of Instruments, IOP Publishing, on January 2, 2020.