$T_i/T_e$ Dependence of Core Turbulence and Transport in DIII-D QH-Mode Plasmas

TL;DR

This work systematically quantifies how the ion-to-electron temperature ratio $T_i/T_e$ shapes microturbulence-driven transport in DIII-D QH-mode plasmas using global gyrokinetic simulations with the code GTC. By separately varying $T_i$ (at fixed $T_e$) and $T_e$ (at fixed $T_i$), the study uncovers a consistent ITG stabilization and TEM destabilization as $T_i/T_e$ decreases, with nonlinear zonal flows providing strong transport suppression that scales with $T_i/T_e$ and eddy sizes that respond to $T_e$ through $\rho_s$. Impurity effects are found to be modest in their impact on saturated transport, while helium as a main ion species yields higher linear growth yet lower nonlinear transport, implying potential confinement benefits for ITER-like scenarios. The results highlight gyro-Bohm scaling as a key factor in interpreting transport differences across heating regimes and emphasize the role of temperature profiles and zonal flows in optimizing confinement for future fusion devices.

Abstract

This study investigates the effect of the ion-to-electron temperature ratio ($T_i/T_e$) on microturbulence driven transport in Quiescent H-mode (QH-mode) plasmas in the DIII-D tokamak. Utilizing the Gyrokinetic Toroidal Code (GTC) and the QH-mode equilibrium, we perform linear and nonlinear simulations to analyze transport properties and instability dynamics under variations of $T_i$ and $T_e$. Our results demonstrate that decreasing $T_i/T_e$ leads to a relative destabilization of trapped electron modes (TEM) over ion temperature gradient (ITG) modes, with the transition between these regimes dictated by $T_i/T_e$. When the electron temperature is increased at fixed ion temperature, we observe an increase in transport saturation levels. In contrast, decreasing the ion temperature at fixed electron temperature results in more modest transport enhancement. The radial correlation length, which characterizes eddy size, increases with rising $T_e$ and decreases with falling $T_i$, consistent with the observed trends in turbulent transport. Additionally, we examine the impact of impurity addition on turbulence and growth rates, finding that impurity presence does not significantly alter transport quantities compared to the impurity-free case. Finally, investigating helium as an alternative main ion species, we find that helium plasmas exhibit higher linear growth rates but result in lower transport saturation levels than deuterium plasmas, suggesting potential confinement benefits. These findings provide quantitative insights into the temperature ratio dependence in QH-mode plasmas and highlight the role of temperature profiles and zonal flows in influencing plasma confinement.

Figures (12)

• Figure 1: Time evolution of plasma parameters for the discharge #157102. We chose a small time window around $2420$ms in the above discharge marked with the grey box to simulate plasma. The QH mode operation starts around $1050$ms when EHOs appear. There is a sporadic ELM at $\sim 1500$ ms and then the EHOs reappear strongly. This can also be seen in part (b), which shows the "quiet" $D_{\alpha}$ profile when EHOs are present. More details about the experiment can be found in Ref chen2016rotational.
• Figure 2: (Left) The original temperature profile obtained from the experiment. The ion temperature ($\sim 10.24$ keV) is higher than the electron temperature ($\sim 3.54$ keV) due to NBI injection. The impurity ion temperature (impurity taken here to be Carbon and its temperature is denoted by $T_z$ in the above figure) is the same as the main ion temperature for the reasons described in Sec \ref{['sec:impstudy']}. All the quantities are normalized by on-- axis electron temperature. The grey dotted lines mark the simulation domain $\psi/\psi_w \in [0.05,0.9]$ where $\psi$ is the flux surface. (Right) We plot the temperature gradient normalized by on-axis electron temperature as a function of $\psi/\psi_w$.
• Figure 3: (Left) Experimental electron density profile used in the simulations, normalized to the on-axis electron density, $n_{e_{\text{mag}}}$. The grey dotted lines mark the simulation domain $\psi/\psi_w \in [0.05,0.9]$. We are using deuterium plasma in our simulation. The electron density profile is the same whether the impurities (impurity taken here to be Carbon and its density is denoted by $n_z$ in the above figure) are present or absent. Due to the quasineutrality condition in Eq. \ref{['eq:quasineutrality']}, the ion density differs when the impurity is present or absent. When no impurities are present, then $n_i$(w/o imp) = $n_e$, shown in blue and orange colors. When the impurities are absent, $n_i$(w/ imp)= $n_e$- $6\times n_z$, shown in green color in the above figure. (Right) We plot the density gradient normalized by on-axis electron density as a function of $\psi/\psi_w$.
• Figure 4: (a) Contour plots of the electrostatic perturbed potential in the linear phase on $\zeta=0$ poloidal plane at $t=37.5R_0/C_s$,(b) nonlinear phase without ZF, (c) nonlinear phase with ZF, both at $t=57.50 R_0/C_s$. The black curves indicate the inner and outer simulation boundaries.
• Figure 5: (Left) Poloidal spectrum in the linear regime at $t=25R_0/C_s$ for the four $T_i/T_e$ ratios, where the maximum amplitude is normalized to 1. The poloidal mode number increases as we decrease the $T_i/T_e$ ratio, with electron temperature fixed while only the ion temperature is varied. (Right) We plot the radial profile in the linear regime at $t=25R_0/C_s$ for the four $T_i/T_e$ ratios, also normalized to a maximum amplitude of 1. All modes peak at the same location of $\psi/\psi_w=0.29$. Additionally, we observe a secondary peak around $\psi/\psi_w=0.1$, identified as an ITG mode for $T_i/T_e=2.90$ and a TEM mode for $T_i/T_e=1.0, 1/2.90$. For $T_i/T_e=1.45$, the mode diamagnetic frequency is zero at that location.