Skip to Main content Skip to Navigation

Study of tokamak plasma disruptions and runaway electrons in a metallic environment

Abstract : Nuclear fusion is regarded as one of the most promising candidates for humankind’s future energy sources. Tokamaks are the devices currently closest to achieve reactor- relevant fusion power. The nuclear fusion power increases with the size of the tokamak and a large plasma current is required for better confinement. In tokamaks, disruptions are unfavorable events in which the plasma energy is lost in a very short timescale causing damage to the tokamak’s structures. Disruption loads increase with energy and plasma current. Thus, they are a major threat to the robust operation of future large tokamaks, including ITER. There are three consequences of disruptions: thermal loads, electromagnetic (EM) loads and runaway electrons (RE). Runaway electron beams carry the risk of in-vessel component damage. In larger machines, higher plasma current increases the runaway formation. The prevention, control and mitigation of the runaway electrons are areas that are considered as hot topics in nuclear fusion research. The current strategy for runaway electrons is to avoid their generation by a massive material injection (MMI) of deuterium or high-Z noble species (Ne, Ar, Kr, Xe). If their generation cannot be avoided, a second MMI will be used to mitigate the generated RE beam. Material can be injected via either Massive Gas Injection (MGI) or Shattered Pellet Injection (SPI, currently adopted by ITER). After the first MMI to prevent RE generation, a cold dense background plasma of MMI impurities is formed. In its presence, the second MMI aimed at mitigating the runaway electron beam may be inefficient due to poor penetration, as observed in the JET tokamak. Therefore, understanding the physics of the interaction between the runaway electron beam and the mitigation MMI in the presence of a cold background plasma is an essential study for a reliable runaway electron beam mitigation scenario. This study will be the focus of this PhD thesis. The background plasma is characterized through its electron temperature. For this, a method based on VUV spectroscopy is developed. In this method, synthetic line ratios are constructed using Photon-Emissivity Coefficient from the ADAS atomic model and the background plasma temperature profile is estimated by fitting the experimental line ratios with the synthetic line ratios. Background plasma in the JET tokamak is hotter (Te ≈6-18 eV) than on other tokamaks (DIII-D, Te ≈1-2 eV). The electron temperature of the background plasma increases with the gas amount used to trigger the disruption and electron density in the far scrape-off layer. When the background plasma is created using argon SPI, the electron temperature have no dependence on the pre-disruptive plasma temperature but is found to weakly correlate with the pre-disruption electron density. In addition, intact SPI pellets produce hotter background plasma. When argon SPI is used as a mitigation injection, it produces hotter background plasma than MGI. A 0D/1D power balance of the runaway electron beam and the background plasma is performed to confirm the temperature measurements. In the power balance, the dominant physical processes like collisional power transfer, synchrotron and bremsstrahlung radiation, electric field acceleration, line radiation of the background plasma are considered. The background plasma temperature predicted by the 0D power balance model is in good agreement with measurements from VUV spectroscopy. The collisional power transfer between the runaway electrons and the background plasma is found to be the primary power source heating the background plasma to high temperatures. The results of a 1D radial diffusion code, adapted for the JET tokamak, are presented. The model is sensitive to initial guesses of the species densities and the geometrical wall radius. The 1D diffusion model predicts higher electron temperature and density when rate coefficients from ADAS atomic model are used compared to CRETIN atomic model (used in the code by default). As compared to the temperature estimated from VUV spectroscopy, the simulated argon background plasma temperatures are much lower and they decrease when the argon amount increases. When a deuterium SPI in argon background plasma is simulated, a drop in the argon line brightness after the entry of deuterium SPI is predicted, consistent with experimental VUV measurements. However, the model predicts an increase in electron density after deuterium SPI entry, inconsistent with experimental measurements. The model predicts a drop in temperature after deuterium SPI entry but not low enough for the argon recombination conditions. On the other hand, the model predicts low electron temperature and density, supporting argon recombination in DIII-D. The over-prediction of electron density and temperature may be due to the presence of higher radiated power (one of model’s inputs) in JET (∼1-4 MW) than on DIII-D (≤100 kW). A large fraction of non-thermal radiation due to the runaway electrons is considered to explain this observation in JET compared to DIII-D.
Complete list of metadata
Contributor : Cédric Reux Connect in order to contact the contributor
Submitted on : Wednesday, May 18, 2022 - 10:54:46 PM
Last modification on : Friday, May 20, 2022 - 3:19:26 AM


Files produced by the author(s)


  • HAL Id : tel-03671885, version 1



Sundaresan Sridhar. Study of tokamak plasma disruptions and runaway electrons in a metallic environment. Plasma Physics [physics.plasm-ph]. Université d'Aix-Marseille, 2020. English. ⟨tel-03671885⟩



Record views


Files downloads