Search

Turbulent transport in magnetically confined fusion plasma has conventionally been analyzed at the ion gyroradius scale based on the microturbulence theory. However, ion-scale turbulence analysis sometimes fails to predict the turbulent transport flux observed experimentally. Microturbulence at the electron gyroradius scale and cross-scale interactions between disparate-scale turbulences are possible mechanisms to resolve this issue. This overview discusses the recent progress in multiscale turbulence studies and presents future perspectives from recent experimental, theoretical, and numerical investigations. The following aspects are highlighted: (1) the importance of electron-scale effects in experiments, (2) the physical mechanisms of cross-scale interactions, (3) modeling electron-scale effects in quasilinear transport models, and (4) the impacts of cross-scale interactions on burning plasmas. Understanding multiscale turbulence is necessary to improve performance prediction and explore optimal operations for future burning plasmas.

This work presents the PORTALS framework (Rodriguez-Fernandez et al 2022 Nucl. Fusion 62 076036), which leverages surrogate modeling and optimization techniques to enable the prediction of core plasma profiles and performance with nonlinear gyrokinetic simulations at significantly reduced cost, with no loss of accuracy. The efficiency of PORTALS is benchmarked against standard methods, and its full potential is demonstrated on a unique, simultaneous 5-channel (electron temperature, ion temperature, electron density, impurity density and angular rotation) prediction of steady-state profiles in a DIII-D ITER Similar Shape plasma with GPU-accelerated, nonlinear CGYRO (Candy et al 2016 J. Comput. Phys. 324 73–93). This paper also provides general guidelines for accurate performance predictions in burning plasmas and the impact of transport modeling in fusion pilot plants studies.

The recent progress in high-temperature superconductor technologies has led to the design and construction of SPARC, a compact tokamak device expected to reach plasma breakeven with up to 25 MW of external ion cyclotron resonant heating (ICRH) power. This manuscript presents local (flux-tube) and radially global gyrokinetic GENE (Jenko et al 2000 Phys. Plasmas 7 1904) simulations for a reduced-field and current H-mode SPARC scenario showing that supra-thermal particles—generated via ICRH—strongly suppress ion-scale turbulent transport by triggering a fast ion-induced anomalous transport barrier. The trigger mechanism is identified as a wave-particle resonant interaction between the fast particle population and plasma micro-instabilities (Di Siena et al 2021 Phys. Rev. Lett. 125 025002). By performing a series of global simulations employing different profiles for the thermal ions, we show that the fusion gain of this SPARC scenario could be substantially enhanced by up to ∼80% by exploiting this fast ion stabilizing mechanism. A study is also presented to further optimize the energetic particle profiles, thus possibly leading experimentally to an even more significant fusion gain.

Perturbative transport experiments in magnetically confined plasmas have shown, for more than 20 years, that the injection of cold pulses at the plasma edge can trigger the increase of core temperature. Predictive heat transport simulations with the trapped gyro Landau fluid (TGLF) quasilinear transport model demonstrate that the increase of core temperature in some regimes, and lack thereof in other regimes, can be explained by a change in dominant linear micro-instability in Alcator C-Mod. The effect of density and plasma current on the cold pulse are well captured by TGLF, including the relative change in position of the temperature flex point as current density changes. Linear stability analysis of simulated density and current scans reveals a competition between trapped electron and ion temperature gradient modes as the main driver of the core transient response. These results further demonstrate that cold-pulse propagation and associated phenomenology in the cases studied are well explained within the local transport paradigm, without resorting to non-local effects. [ABSTRACT FROM AUTHOR]

Perturbative thermal diffusivity is measured with partial sawtooth-generated heat pulses on ASDEX Upgrade for the first time, and these measurements are used to validate the first nonlinear ion-scale gyrokinetic simulation that agrees with experimentally measured perturbative diffusivity. Recent work on Alcator C-Mod (Howard et al 2016 Phys. Plasmas23 056109) and DIII-D (Holland et al 2017 Nucl. Fusion57 066043) has shown that ion-scale gyrokinetics cannot adequately describe certain plasma conditions, and that in these instances multi-scale simulations may resolve the observed discrepancies. This work presents the first measurements of electron perturbative thermal diffusivity with partial sawteeth on ASDEX Upgrade (measuring values between 2 and 9 m2 s−1), and compares these measurements to those made with more established modulated electron cyclotron heating measurements, finding good agreement within experimental uncertainty. Perturbative diffusivity is found to scale inversely with collisionality across ASDEX Upgrade and Alcator C-Mod plasmas. Finally, perturbative diffusivity is used as a validation constraint in a study with the gyrokinetic code GENE, showing the first instance where an ion-scale gyrokinetic simulation can simultaneously match the experimental ion and electron heat fluxes and the perturbative thermal diffusivity. These results indicate that multi-scale effects may only be important in some plasmas, and the collisionality and ratio of high to low wavenumber linear growth rates may distinguish these cases. [ABSTRACT FROM AUTHOR]

Contemporary predictive models for heat and particle transport in tokamak plasmas are based on the assumption that local fluxes can be described in terms of local plasma parameters, where electromagnetic drift-wave-type turbulence is driven by local gradients and results in cross-field transport. The question of whether or not transport could be dominated by non-local terms in certain circumstances is essential for our understanding of transport in magnetically confined plasmas, and critical for developing predictive models for future tokamaks, such as ITER. Perturbative transport experiments using cold-pulse injections at low density seem to challenge the local closure of anomalous transport: a rapid temperature increase in the core of the plasma following a sharp edge cooling is widely observed in tokamaks and helical devices. Past work in Ohmic plasmas in Alcator C-Mod and in ECH plasmas in KSTAR found that the temperature inversions disappear at higher densities, above the intrinsic toroidal rotation reversal density. These observations suggested that the so-called ‘non-local’ heat transport effects were related to the intrinsic rotation reversal, and therefore to changes in momentum transport. In this work, new experiments and analysis at Alcator C-Mod show that intrinsic rotation reversals and disappearance of temperature inversions are not concomitant in Ohmic plasmas at high plasma current and in ICRH L-modes. This new data set shows that the correlation between transient temperature inversions and intrinsic rotation reversals is not universal, suggesting that ‘non-local’ heat transport and momentum transport effects may be affected by different physical mechanisms. [ABSTRACT FROM AUTHOR]

New multiscale gyrokinetic simulations predict that electron energy transport in a DIII-D ITER baseline discharge with dominant electron heating and low input torque is multiscale in nature, with roughly equal amounts of the electron energy flux Qe coming from long wavelength ion-scale (kyρs  <  1) and short wavelength electron-scale (kyρs  >  1) fluctuations when the gyrokinetic results match independent power balance calculations. Corresponding conventional ion-scale simulations are able to match the power balance ion energy flux Qi, but systematically underpredict Qe when doing so. Significant nonlinear cross-scale couplings are observed in the multiscale simulations, but the exact simulation predictions are found to be extremely sensitive to variations of model input parameters within experimental uncertainties. Most notably, depending upon the exact value of the equilibrium E  ×  B shearing rate γE×B used, either enhancement or suppression of the long-wavelength turbulence and transport levels in the multiscale simulations is observed relative to what is predicted by ion-scale simulations. While the enhancement of the long wavelength fluctuations by inclusion of the short wavelength turbulence was previously observed in similar multiscale simulations of an Alcator C-Mod L-mode discharge, these new results show for the first time a complete suppression of long-wavelength turbulence in a multiscale simulation, for parameters at which conventional ion-scale simulation predicts small but finite levels of low-k turbulence and transport consistent with the power balance Qi. Although computational resource limitations prevent a fully rigorous validation assessment of these new results, they provide significant new evidence that electron energy transport in burning plasmas is likely to have a strong multiscale character, with significant nonlinear cross-scale couplings that must be fully understood to predict the performance of those plasmas with confidence. [ABSTRACT FROM AUTHOR]

A new paradigm of zonal flow mixing as the mechanism by which zonal EB fluctuations impact the saturation of gyrokinetic turbulence has recently been deduced from the nonlinear 2D spectrum of electric potential fluctuations in gyrokinetic simulations. These state of the art simulations span the physical scales of both ion and electron turbulence. It was found that the zonal flow mixing rate, rather than zonal flow shearing rate, competes with linear growth at both electron and ion scales. A model for saturation of the turbulence by the zonal flow mixing was developed and applied to the quasilinear trapped gyro-Landau fluid transport model (TGLF). The first validation tests of the new saturation model are reported in this paper with data from L-mode and high-βp regime discharges from the DIII-D tokamak. The shortfall in the predicted L-mode edge electron energy transport is improved with the new saturation model for these discharges but additional multiscale simulations are required in order to verify the safety factor and collisionality dependencies found in the modeling. [ABSTRACT FROM AUTHOR]

Perturbative thermal diffusivity has been measured on Alcator C-Mod via the use of the extended-time-to-peak method on heat pulses generated by partial sawtooth crashes. Perturbative thermal diffusivity governs the propagation of heat pulses through a plasma. It differs from power balance thermal diffusivity, which governs steady state thermal transport. Heat pulses generated by sawtooth crashes have been used extensively in the past to study heat pulse thermal diffusivity (Lopes Cardozo 1995 Plasma Phys. Control. Fusion37 799), but the details of the sawtooth event typically lead to non-diffusive ‘ballistic’ transport, making them an unreliable measure of perturbative diffusivity on many tokamaks (Fredrickson et al 2000 Phys. Plasmas7 5051). Partial sawteeth are common on numerous tokamaks, and generate a heat pulse without the ‘ballistic’ transport that often accompanies full sawteeth (Fredrickson et al 2000 Phys. Plasmas7 5051). This is the first application of the extended-time-to-peak method of diffusivity calculation (Tubbing et al 1987 Nucl. Fusion27 1843) to partial sawtooth crashes. This analysis was applied to over 50 C-Mod shots containing both L- and I-Mode. Results indicate correlations between perturbative diffusivity and confinement regime (L- versus I-mode), as well as correlations with local temperature, density, the associated gradients, and gradient scale lengths (a/LTe and a/Ln). In addition, diffusivities calculated from partial sawteeth are compared to perturbative diffusivities calculated with the nonlinear gyrokinetic code GYRO. We find that standard ion-scale simulations (ITG/TEM turbulence) under-predict the perturbative thermal diffusivity, but new multi-scale (ITG/TEM coupled with ETG) simulations can match the experimental perturbative diffusivity within error bars for an Alcator C-Mod L-mode plasma. Perturbative diffusivities extracted from heat pulses due to partial sawteeth provide a new constraint that can be used to validate gyrokinetic simulations. [ABSTRACT FROM AUTHOR]

The transport of heat in laboratory and astrophysical plasmas is dominated by the complex nonlinear dynamics of plasma turbulence. In magnetically confined plasmas used for fusion energy research, turbulence is responsible for cross-field transport that limits the performance of tokamak reactors. We report a set of novel gyrokinetic simulations that capture ion and electron-scale turbulence simultaneously, revealing the dynamics of cross-scale energy transfer and zonal flow modification that give rise to heat losses. Multi-scale simulations are required to match experimental ion and electron heat fluxes and electron profile stiffness, establishing the applicability of the newly discovered physics to experiment. Importantly, these results provide a likely explanation for the loss of electron heat from tokamak plasmas, the ‘great unsolved problem’ (Bachelor et al (2007 Plasma Sci. Technol. 9 312–87)) in plasma turbulence and the projected dominant loss channel in ITER. [ABSTRACT FROM AUTHOR]

Core impurity transport has been investigated for a variety of confinement regimes in Alcator C-Mod plasmas from x-ray emission following injection of medium and high Z materials. In ohmic L-mode discharges, impurity transport is anomalous (Deff ≫ Dnc) and changes very little across the LOC/SOC boundary. In ion cyclotron range of frequencies (ICRF) heated L-mode plasmas, the core impurity confinement time decreases with increasing ICRF input power (and subsequent increasing electron temperature) and increases with plasma current. Nearly identical impurity confinement characteristics are observed in I-mode plasmas. In enhanced Dα H-mode discharges the core impurity confinement times are much longer. There is a strong connection between core impurity confinement time and the edge density gradient across all confinement regimes studied here. Deduced central impurity density profiles in stationary plasmas are generally flat, in spite of large amplitude sawtooth oscillations, and there is little evidence of impurity convection inside of r/a = 0.3 when averaged over sawteeth. [ABSTRACT FROM AUTHOR]

The need to fit smooth temperature and density profiles to discrete observations is ubiquitous in plasma physics, but the prevailing techniques for this have many shortcomings that cast doubt on the statistical validity of the results. This issue is amplified in the context of validation of gyrokinetic transport models (Holland et al 2009 Phys. Plasmas16 052301), where the strong sensitivity of the code outputs to input gradients means that inadequacies in the profile fitting technique can easily lead to an incorrect assessment of the degree of agreement with experimental measurements. In order to rectify the shortcomings of standard approaches to profile fitting, we have applied Gaussian process regression (GPR), a powerful non-parametric regression technique, to analyse an Alcator C-Mod L-mode discharge used for past gyrokinetic validation work (Howard et al 2012 Nucl. Fusion52 063002). We show that the GPR techniques can reproduce the previous results while delivering more statistically rigorous fits and uncertainty estimates for both the value and the gradient of plasma profiles with an improved level of automation. We also discuss how the use of GPR can allow for dramatic increases in the rate of convergence of uncertainty propagation for any code that takes experimental profiles as inputs. The new GPR techniques for profile fitting and uncertainty propagation are quite useful and general, and we describe the steps to implementation in detail in this paper. These techniques have the potential to substantially improve the quality of uncertainty estimates on profile fits and the rate of convergence of uncertainty propagation, making them of great interest for wider use in fusion experiments and modelling efforts. [ABSTRACT FROM AUTHOR]