Session TP15: Poster Session: Magnetic Confinement: DIII-D Tokamak II (9:30am - 12:30pm)

Effect of realignment on detachment in DIII-D small angle slot divertor

Analysis using the newly aligned small angle slot (SAS) divertor shows robustness of detachment onset to small divertor misalignments. The toroidal alignment of the SAS divertor in DIII-D was recently improved with respect to the outer strike point (OSP) location, as confirmed by toroidally separated Langmuir probes. Preliminary analysis of matched pre- and post-alignment discharges shows no significant difference in detachment onset for the OSP in the outer divertor corner. Detachment onset is rapid and observed across diagnostics for discharges with ion $\nabla B$ drift into the SAS; detachment onsets at $n_e\approx 6.0\times10^{13}$ cm$^{-3}$ pre-alignment, and $\approx 6.0-6.6\times10^{13}$ cm$^{-3}$ post-alignment. For ion $\nabla B$ drift out of the divertor, target $J_{sat}$ measured in the near scrape off layer rolls over at $n_e\approx 4.7\times 10^{13}$ cm$^{-3}$ in both pre- and post-alignment discharges. Probe-measured $T_e$ drops from $\approx 17$ eV at $\approx 3.7\times 10^{13}$ cm$^{-3}$; pre-alignment $T_e$ drops immediately to $<10$ eV but post-alignment $T_e$ drops more slowly to $<10$ eV near $J_{sat}$ rollover. These observations suggest that local variations in divertor conditions may have little effect on global detachment.   

First SOLPS-ITER modelling with drift effects of the new V-shape small angle slot divertor on DIII-D

This paper is the first numerical assessment of the detachment performance of a new DIII-D divertor configuration, the “SAS-V”, using SOLPS-ITER with drifts. Recent modelling indicates that a symmetric V-end slot divertor can reach detachment at lower plasma densities relative to other divertors for both toroidal field directions, significantly reducing detachment asymmetries resulting from particle drift effects. To test this promising finding, the flat-end of the currently installed small-angle slot (SAS) divertor in the DIII-D tokamak will be replaced with a V-end target, creating a new divertor shape, the “SAS-V”. In this contribution, the detachment performance of SAS-V is systematically studied using SOLPS-ITER with drifts for a range of plasma conditions (density, input power, current) typical for DIII-D. Scans of the strike point position in the slot, the angle between strike line and target, and divertor leg length are also performed to find the magnetic configuration that optimizes detachment. The results of this extensive modelling effort will guide future SAS-V experiments and improve our understanding of the strong interplay between divertor geometry, particle drifts and detachment physics.   

Numerical experiments towards the design of an optimized SAS-2 divertor for DIII-D.

Numerical experiments using the SOLPS-ITER code with full drift terms are carried out to guide the design of the next generation Small-Angle Slot (SAS) divertor for the DIII-D tokamak, referred to as SAS-2. Experimental results from DIII-D showed that a SAS divertor can enhance neutral compression as well as energy and momentum dissipation, thereby mitigating the heat load at the divertor target. Recent SOLPS-ITER modeling of the SAS divertor has highlighted the prominent role of ExB drifts in setting particle flux patterns in and out of the slot. New slot geometries are currently being investigated in view of the SAS-2 divertor, with particular interest in V-shaped slots and in-slot pumping for improved local particle control. In this contribution we assess the effect of key design parameters and local plasma conditions on the SAS-2 divertor energy and momentum dissipation performance. Scans of slot angle, slot width, strike point position, pumping speed, heating power, gas puff flow rate and heat flux width are carried out to optimize the SAS-2 divertor design, improving, at the same time, our understanding of the complex physics interplay between slot geometry, magnetic configuration and ExB drifts.   

Diagnosing Slot Divertors For Physics Understanding

Here we examine specific diagnostic requirements and strategies for closed divertor geometries needed to validate models of divertor dissipation. A slot geometry with its increased closure can achieve improved divertor performance with respect to increased energy dissipation and detachment dynamics at lower upstream densities, as recently demonstrated using the SAS divertor on DIII-D (Guo NF 2019). The physics behind this behavior is complex, representing an intimate coupling of particle retention, plasma drifts, divertor magnetic geometry, and neutral flux recycling from the target surfaces. This environment is difficult to model accurately given the limitations of existing codes (SOLPS-ITER, UEDGE), setting a premium on good diagnostic measurements. Key parameters include (but are not limited to) plasma temperature and density, heat and particle flux, and neutral particle distributions. Spatial resolutions are set by the dissipation scale, which can be comparable to the slot dimensions. Unfortunately, the small solid angle and poor access make these measurements very challenging to deploy. We will discuss these measurement requirements, compare them with the existing SAS diagnostic suite, and suggest improvements for planned upgrades (SAS-1VW, SAS-2) on DIII-D.   

MHD stability constraints on divertor heat flux width in DIII-D

The radial width of heat flux flowing into the DIII-D divertor is found to expand at high input power and plasma density, exceeding existing empirical scaling but consistent with MHD stability limits. At low heating power the SOL width remains consistent with empirical scaling dependent only on the midplane poloidal field, even for high density divertor detachment. At high heating power a higher separatrix density and pressure, are required to achieve divertor detachment. For separatrix density approaching half of the Greenwald density limit the separatrix pressure gradient saturates at a level near the MHD limit. However, any increased transport associated with SOL broadening does not excessively propagate inward to degrade the edge pedestal pressure and overall core plasma performance if excessive detachment is avoided. The components of the midplane pressure profile are measured with Thomson scattering and CER for main ion and CVI ion temperature and density. The separatrix normalized pressure stability limit is evaluated with the ideal MHD code BALOO and found to be relatively constant across the data set. The sensitivity of this analysis to edge profile fitting is also explored. The implications for divertor heat flux dissipation in future devices are assessed.   

DIVIMP-WallDYN predictions of tungsten migration and transport in slot versus V-shaped divertors

Time-dependent, mixed-material DIVIMP-WallDYN modeling is presented that predicts favorable tungsten (W) erosion and leakage properties from a V-shaped divertor, versus a comparable slot divertor. Plasma backgrounds are obtained from SOLPS 5.1 with drifts, assuming identical upstream conditions but differing in outer divertor geometry: a small-angle slot (similar to the current SAS in DIII-D), and a V-shape (similar to the proposed SAS-V in DIII-D). Material migration simulations are performed for a toroidally-symmetric W tile in the outer divertor of an otherwise carbon (C) machine, similar in setup to the planned SAS1-VW experiment in DIII-D. For favorable ion grad-B drift direction, in both cases the initial W surface reaches a mixed W/C equilibrium within 10 seconds, remaining at least 60{\%} W throughout the common flux region. Both gross and net erosion of W are markedly lower in the V-shaped case, primarily due to the lower electron temperature near the strike point. Furthermore, a lower fraction of the eroded W leaks out of the divertor in the V-shaped case, primarily due to stronger near-target deuterium flows. The poloidal extent of the W tile is shown to not be a major driver of W migration behavior. In both geometries, simulations show these W migration patterns are largely recovered even when W is initially covered by incidentally-deposited C layers.   

Effect of particle drifts on tungsten transport and leakage in the new V-shaped Small Angle Slot divertor in DIII-D

The DIII-D Small Angle Slot (SAS) divertor will be coated with tungsten (W) to investigate W sourcing and leakage from a closed slot divertor during the next experimental campaign. The SOLPS-ITER code package with drifts, coupled with the DIVIMP impurity Monte Carlo code, finds that the addition of drifts shifts the previously predicted location of primary W sourcing, which will indirectly impact leakage due to spatial differences in plasma parameters throughout the slot. Differences in impurity transport based on the distance from the strike point for no-drift and drift cases are presented. Changes to the baffling in the SAS divertor, creating a V-shape, are predicted to facilitate detachment with the ion Bx$\nabla $B drift direction toward the slot. SOLPS-ITER simulations with the V-shape show better neutral confinement near the strike point, leading to lower target electron temperatures. This effect lowers the impact energy of particles incident on the W plasma-facing wall, thereby reducing gross erosion. Predicted increases in near-target electron density and strong near-target flows with the V-shape may also reduce the W impurity leakage fraction. Results are compared to previous simulations conducted using the original SAS geometry in both B$_{\mathrm{t}}$ directions.   

Melting of leading edges and surfaces of high-and low-Z plasma facing components in the DIII-D Divertor.

Plasma facing component (PFC) edge and surface melting is a serious concern for ITER as it can cause PFC damage, plasma contamination and dust production. Melting of tungsten leading edges was observed during experiments in the lower divertor of DIII-D. W blocks misaligned by 0.3 mm and 1 mm with respect to the divertor tile level were exposed near the outer strike point during deuterium and helium L- and H-mode discharges using the DiMES manipulator. FIB SEM analysis showed an evidence of W recrystallization under the edges, and formation of cracks up to 100 microns wide was observed. Micro-scale melting was also observed at the toroidal edge of the block raised by 1 mm, indicating the potential importance of finite Larmor radius effects for edge thermal loading. Additional data on leading edge melting were obtained during the Metal Rings Campaign, where W-coated molybdenum inserts in the lower divertor tiles bowed during plasma exposure, forming leading edges. Re-solidified melt layers were observed at the edges, their shape being consistent with motion in jxB direction with j driven by electron emission. Plans for exposure of an aluminum block as a beryllium proxy to benchmark ITER-relevant MEMOS-U modeling of melt layer dynamics will also be presented.   

Micro engineered trench targets to capture deuterium ion impact angle distributions on DiMES at DIII-D

Recent DIII-D DiMES experiments have directly measured material deposition patterns affected by ion shadowing in 30x30x2 or 3 $\mu $m deep trenches. These trenches were fabricated via focused ion beam (FIB) milling on a Si surface partially coated with Al to clearly reveal the polar and azimuthal ion impact angle distributions (IADs). The angle of incidence of ions impacting a PFC surface is important in estimating the PFC erosion lifetime as the sputtering yield is highly sensitive to pitch angles below 10-20 degrees. These sample surfaces were exposed to L-mode D discharges for a total of 30 s using the DiMES facility at DIII-D. The areal impurity Al and C concentrations on the trench floors were measured by energy-dispersive X-ray spectroscopy (EDS). Net erosion profiles, calculated by a Monte-Carlo model, Micro-Patterning and Roughness (MPR) code, with input of predicted IADs, reproduced the characteristic shape of the experimental concentration profiles. The measurement revealed that the pitch angle of ions impacting on the surface peaks at 5-10 degrees while ions were entering the sheath region with a pitch angle of 1-2 degrees.   

A New UV Spectrometer for Measurements of Tungsten Erosion {\&} Re-deposition

A new high-resolution and high-throughput ultraviolet spectrometer has been constructed to resolve the most promising W line radiation arising from erosion during W sourcing experiments in the DIII-D divertor. Spectral surveys in the Compact Toroidal Hybrid experiment between 200 and 400 nm have identified over 60 neutral and singly ionized W emission lines in the UV region, which can be combined with atomic predictions to determine the net erosion and re-deposition of W on plasma facing surfaces. The importance of W metastable level populations requires that multiple W emission lines be monitored simultaneously to accurately characterize erosion rates. The new UV spectrometer has a maximum resolving power of 1.6 {\AA} at 250 nm with better than 1 kHz temporal resolution. The details of the fiber-coupled collection optics and instrument shielding required for installation on the DIII-D tokamak will be presented along with expected signal levels for DIII-D plasma conditions.   

An Update on W I and W II Atomic Data for use in Erosion, Redeposition, and Transport Studies

In support of tungsten experiments on DIII-D, we present a summary of the excited state ionization calculations for neutral W and recommend a set of rate coefficients for use in plasma modeling and diagnostics. For redeposition measurements, emission from higher charge states must also be used, so, we also report on a new $R$-matrix calculation for electron impact excitation of W$^{\mathrm{+}}$, comparing with spectral observations from the Auburn CTH experiment, as well as preliminary $R$-matrix with pseudostates calculations for the electron-impact ionization of W$^{\mathrm{+}}$ and W$^{\mathrm{2+}}$. Spectroscopic techniques to measure erosion, re-deposition, and transport for tungsten plasma facing components require accurate atomic data, with the near-neutral ion stages being the most critical. The available atomic data has been undergoing significant improvements in the past few years using large-scale quantal calculations. This work updates the progress on these calculations and reviews the current status.   

Impact of Divertor Closure on Super-H mode Plasmas in DIII-D

Integration of a high pressure pedestal and high performance core with a radiative divertor is assessed in DIII-D Super H-mode (SH) plasmas. SH is a promising regime for future devices due to the high pedestal pressures able to be obtained via increased shaping and density. The peeling limited pedestal in the SH regime allows for high densities in the scrape off layer and pedestal foot, without degradation of the pedestal height. Previous DIII-D experiments have shown significant temperature reduction at the divertor plate and near-detachment conditions with nitrogen seeded SH-modes in an open divertor. In this poster, similar plasmas operating in the upper, more closed, divertor configuration in DIII-D are compared to those operating in the lower divertor. The SH pedestal in the closed divertor is more resilient to deuterium gas puffing than the open divertor configuration, showing less degradation in both the temperature and density pedestal with increased gas. The closed divertor configuration can tolerate fueling up to \textasciitilde 145 Torr-L/s while maintaining a high performance core. The open divertor displays higher absolute separatrix densities and collisionalities with similar pedestal pressure due to increased pumping efficiency in the closed divertor configuration.   

Inter-ELM Recovery and Pedestal Fueling from Main-ions and Impurities in Low Frequency ELMy H-mode Discharges in DIII-D

Time-dependent, ELM-synchronized analysis of the inter-ELM buildup of the electron density pedestal in DIII-D has quantified the role of fueling from main-ions (deuterium) and the dominant impurity (carbon) in a low ELM frequency DIII-D discharge. The inter-ELM density pedestal recovery can be dominated by impurity influx and impurity ionization, with more electrons from C than D, and must be included when assessing electron particle transport in the H-mode pedestal. Establishment of the main-ion density profile may precede the impurity influx, consistent with onset of a neoclassical impurity pinch very early in the ELM cycle, which then fuels the pedestal by impurity ionization. Temporally resolved profiles show a factor or two increase in the electron and impurity density, with temperature and rotation displaying rapid profile establishment. These observations are expected from considerations of thermal transport being driven largely from the core outwards, whereas the particle transport is dominantly driven from the edge inwards by strong fueling sources. Inclusion of the impurity ionization source may help determine if there is an inward pinch of electrons in the formation of the edge particle transport barrier.   

Modifications of Nonlinear interactions in the Pedestal leading to ELM onset\textunderscore

Edge Localized Modes (ELM) are bursty relaxations of the pedestal. The nonlinear mechanism describing the onset of these bursty relaxations remains elusive. We investigate the relationship between the occurrence of ELMs and the associated inter-ELM pedestal localized modes on DIII-D. Specifically, we will discuss cases where NBI-induced pedestal perturbations modify the nonlinear interactions that systematically lead the ELM onset. We report analysis of the effects of the NBI (at fixed applied counter-torque) on the inter-ELM fluctuations, the ELM onset, and accordingly on the ELM frequency. First, we observe that single ELMs occur at multiples of the beam modulation periods. Second, we find that the NBI (low frequency modulation) leads to the bifurcation of pedestal localized low frequency modes ( \textasciitilde 50 kHz), consistent with the alteration of the mode's drive. Third, we identified using BES a beam-induced density perturbation localized to the pedestal leading to the ELM onset. Connections of these three observations with the ELM onset will be proposed, and similarities with other actuators will also be discussed.   

Internal measurement of low-frequency magnetic fluctuations in DIII-D ELMy H-mode pedestal

Faraday-effect polarimetry has showed high-frequency (100-500 kHz) broadband magnetic turbulence identified as micro-tearing modes in DIII-D ELMy H-mode pedestal [1]. Here we report low-frequency (1-100 kHz) magnetic fluctuations also observed in the pedestal for the same plasma conditions. The low-frequency magnetic fluctuations have line-averaged (lower-bound) $\delta $b/b$=$2.5x10$^{\mathrm{4}}$ and \textbar $\delta $b/b\textbar /\textbar $\delta $n/n\textbar $=$0.025 at the peak frequency \textasciitilde 10 kHz, where $\delta $b is line-averaged radial magnetic fluctuation amplitude measured by polarimetry and $\delta $n is local density fluctuation measured at the same frequency by Beam-Emission-Spectroscopy (BES). BES measurements establish that the low-frequency density fluctuations are localized in the pedestal and propagate in the ion diamagnetic direction (plasma frame). These observations indicate the existence of an ion direction, electromagnetic instability in the ELMy H-mode pedestal. BALOO and ELITE calculation suggest kinetic-ballooning modes and peeling-ballooning modes, which are unstable and have magnetic feature, are potential candidates for the measured fluctuations. [1] J. Chen et al., APS-DPP invited talk, Fort Lauderdale (2019).   

Variation of ELM Frequency with NBI Heating Delay in DIII-D

The impact of increase in heating power with respect to the current ramp-up and flat top on the pedestal structure is not yet understood. In DIII-D we observe that if the NBI heating time is altered with respect to reaching the $I_{p}$ flat-top, ELM frequency ($f_{ELM})$ can vary by a factor of 2 from \textasciitilde 30Hz to \textasciitilde 60Hz while the pedestal width and height are similar prior to the first ELM. Fueling is same (20 Torrl/s) prior to L-H transition and onwards. While all discharges show low frequency quasi-coherent magnetic fluctuations (\textless 100 kHz), discharges with low $f_{ELM}$ show broadband fluctuations at high frequency (300-400 kHz) in the magnetic spectrogram, whereas for those with higher $f_{ELM}$ a quasi-coherent mode is observed at 220 kHz. Magnetic fluctuations characteristics and role of the L-H mode transition time with respect to the $I_{p}$ flat top and its effects on the pedestal formation will be investigated to address the differences in $f_{ELM}$. We will investigate time-dependent evolution of the pedestal values before onset of the first ELM and pedestal stability during the ELMing regime. Understanding the impact of the heating onset with respect to the $I_{p}$ ramp-up on the pedestal and confinement might open opportunities for accessing similar confinement regimes with potentially smaller or no ELMs.   

Phase Contrast Imaging and Analysis of Turbulence Dynamics in the H-mode edge on DIII-D

Recent analysis of H-mode edge plasma turbulence measured by Phase Contrast Imaging (PCI) suggests a significant difference in the flow of energy in the fluctuation spectrum compared to L-mode plasmas, with an H-mode spectrum dominated by large sinks in the ITG range while L-mode turbulence is consistent with a cascade to high wavenumber. Earlier analysis of PCI measurements of the line-integrated density fluctuations in H-mode [Rost et al, Phys. Plasmas 21 (2014) 062306] represented the turbulence by spectral moments such as wavenumber and correlation length and generated implausible results even for the H-mode edge, including $k_r>k_{\theta}$ and sub-mm radial correlation lengths. Re-analysis with a new model was able to successfully match the experimental data by assuming a non-Gaussian turbulence spectrum. A Gaussian turbulence spectrum results from strong wave-wave interactions and a cascade to damped short wavelength modes, while the non-Gaussian spectrum results from large energy sinks in the ITG range. While validation metrics used in comparing theoretical models to diagnostic measurements often focus on quantities derived from spectral moments, such as correlation lengths, these results suggest that the full 2-d spectrum is vital to such comparisons   

Collisionality Dependence of 2D Inter-ELM Pedestal Fluctuation Measurements

Localized 2D measurements of density fluctuations in the H-mode pedestal of DIII-D plasmas reveal a range of broadband modes that vary temporally and spatially during the inter-ELM cycle. These measurements are obtained with Beam Emission Spectroscopy and a new higher radial resolution {\$}($\backslash $Delta R$\backslash $sim 0.8 cm){\$} Charge eXchange Imaging (CXI) prototype diagnostic. Fluctuation characteristics will be presented as a function of collisionality, which has been predicted to impact the growth rate of microtearing modes (MTM) and other pedestal-localized instabilities in ELM'ing H-mode plasmas. MTMs are predicted to cause electron thermal transport in the pedestal and other regions of high-beta plasmas. A future optimized CXI diagnostic will measure carbon density fluctuations at the pedestals with up to 3-4 mm radial resolution and complement the Beam Emission Spectroscopy (BES), which is limited to about 1 cm radial resolution. This will enhance sensitivity to fluctuations localized to the narrow pedestal range, potentially enabling the detection of MTM as well as other pedestal instabilities, and their dynamics in between ELMs. Preliminary analysis with prototype data indicates that CXI's enhanced radial resolution resolves multiple simultaneously coexisting inter-ELM instabilities.   

M3D-C1 modeling of pellet ELM triggering in low-collisionality discharges

Fully 3D nonlinear, as well as 2D linear M3D-C1 simulations are used to model ELM triggering by small pellets in the ITER relevant, peeling-limited pedestal stability regime. A critical pellet size threshold is found in both experiment and modeling depending on pedestal conditions, pellet velocity and injection direction. Using radial injection at the outboard midplane, the threshold is determined by M3D-C1 for multiple time slices of a DIII-D low-collisionality discharge that has pellet ELM triggering. Experimental observations show that a larger pellet size than the standard 1.3 mm diameter is necessary for ELM triggering; 1.8 mm pellets triggered several ELMs in cases where a smaller pellet failed. The M3D-C1 simulations are in reasonable agreement with these observations. While the 2D linear simulations give insight into the change of growth rates for various toroidal modes with pellet size, the 3D nonlinear simulations apply a pellet ablation model that mimics the actual injection with good match to the experiment.   

Automating Interpretive Edge Transport Modeling on DIII-D with UEDGE

DIII-D discharges are analyzed with automatic interpretive UEDGE modeling to determine transport profiles in the pedestal and scrape-off layer (SOL). This new workflow enables automatic interpretive UEDGE simulations that match observed temperature and density profiles, given flux boundary conditions and neutral density constraint. Understanding of edge and SOL transport analysis is key for meeting the challenges of a practical fusion reactor, but fluid simulations of the SOL have been hereto time consuming to obtain. Using the OMFIT framework to combine UEDGE with recently developed automatic grid generators and the CAKE automatic kinetic equilibrium reconstruction workflow has allowed the batch analysis of a much larger number of shots and time slices than was previously practical. This automatic workflow has been used to produce scans both from experimental and hypothetical profiles in order to evaluate sensitivities to model parameters. The results show qualitative agreement on detachment and ELM cycle divertor density evolution, but differs from experiment on the quantitative values, such divertor plate density or the density and temperature condition at which plasma detachment occurs. These discrepancies are explored in order to refine modeling parameters.]   

Power scaling of energy confinement in the wide pedestal quiescent H-mode at DIII-D

The recently discovered wide pedestal quiescent H-mode (WPQH) at DIII-D is characterized by stationary operation without ELMs, formation of a wider and higher pedestal than standard QH-mode, and pedestal broadband fluctuations. Unlike the conventional H-mode where energy confinement time ($\tau_{E})$ decreases with increasing heating power, $\tau_{E}$ in WPQH is constant when neutral beam heating is varied from 3.7 to 5.5 MW. In these experiments $\beta_{N}$ increased by 43{\%} but the pedestal height remained constant. Since core profiles in H-mode are generally considered stiff and strongly dependent on the pedestal, it is important to understand energy transport in the pedestal of WPQH. Here, we will present TRANSP power balance analysis to investigate the distribution of power in ion and electron channels. Furthermore, we will present gyrokinetic GENE simulations to understand the underlying microturbulence mechanisms of the core and pedestal. Preliminary results from local linear simulations in the mid pedestal exhibit the presence of low-k, electromagnetic modes with ballooning parity, and real frequencies in the ion diamagnetic direction, consistent with kinetic ballooning modes (KBM). Further analysis of the mid and high-k turbulence as well as global linear simulations will be presented.   

Using synthetic diagnostics to analyze NIMROD simulations of DIII-D QH-mode

Understanding of H-mode operation regimes without edge localized modes (ELMs) requires validated simulation of the nonlinear fluctuation dynamics. The extended-MHD NIMROD code is used to simulate the dynamics of an Edge Harmonic Oscillation (EHO) during DIII-D quiescent H-mode (QH-mode) discharge 163518. EHOs observed in nonlinear MHD simulations have $n=1$ and $n=2$ as dominant modes akin to the observed dynamics in DIII-D. Using the diagnostics in DIII-D to constrain the most accurate equilibrium reconstruction creates a MHD-stable initial state. Hence, the experimental equilibrium for the DIII-D discharge 163518 is modified to include two levels of instability drive by increasing the experimental pressure gradient and associated bootstrap current. A synthetic beam-emission spectroscopy (BES) diagnostic shows that the amplitude of the experimental density perturbations is between the computed density perturbation amplitude for the two levels of instability drive. We discuss how validation of edge instabilities requires an understanding of transport, instability, and careful consideration of how codes are mathematically formulated to create the proper numerical experiments, and how local measurements can help guide this effort.   

Study of edge He transport during resonant magnetic perturbations (RMPs) in DIII-D using a finite difference approximation of the continuity equation

An impurity transport model has been developed to analyze reductions of up to 2x in effective helium particle confinement time observed during application of RMPs to suppress and mitigate edge localized modes (ELMs). This reduction during RMPs was measured in the core, edge, and pump plenum. Higher measured helium pressure and concentration at the pump suggests it was better retained there during RMPs, in excess of that due to deuterium pump-out, enabling faster pumping. A reservoir approximation of the continuity equation with diffusion and convection fits the measurements well. Transport profiles from the fits are consistent with increased helium transport near the separatrix during RMPs. An analysis with EMC3-EIRENE suggests this transport could result from new flows in the perturbed separatrix. These findings provide evidence that RMPs in future devices may provide impurity exhaust that meets or exceeds that due to the ELMs, reducing ash in a burning regime, and increasing fusion gain.   

Measurements of tearing mode effects on pedestal plasma fluctuations and divertor footprint currents

A linear single fluid M3D-C1 model of the plasma response to a rotating \textasciitilde 53 kHz 4/3 tearing mode (TM) confirms the experimentally observed connection between core TM and enhanced pedestal electron temperature modulation as well as the plasma floating potential in the divertor region. Analogous to the plasma response from an external resonant magnetic perturbation, the internal magnetic perturbation field from the island is analytically generated with a current filament model inside the plasma and applied to the M3D-C1 plasma response simulation. The non-resonant perturbation from the core island excites a peeling response at the edge, which leads to an enhanced electron temperature modulation at the TM's frequency. The temperature fluctuation response in M3D-C1 is loaded into the synthetic diagnostic and matches the Electron Cyclotron Emission Imaging (ECEI) measurement in the pedestal. The TM's magnetic perturbation at the edge is \textasciitilde 1.5 Gauss measured by mid-plane Mirnov coils and thus exerts a non-negligible millimeter-level distortion on the mid-plane separatrix. Using field line tracing results from the TRIP3D code, the divertor footprint is found to be modulated at \textasciitilde 53 kHz by the TM's magnetic perturbation. The divertor footprint defines the interaction area of hot electrons from the foot of the pedestal during the presence of error fields and thus significantly changes the local plasma floating potential, which appears as fluctuations on the divertor Langmuir probe currents   

Enhanced Accessibility and Absorption of Helicon and Lower Hybrid Waves in Tokamak Plasmas Via n$_{\mathrm{\vert \vert }}$ Upshift From Poloidal Inhomogeneity

Two high-power systems for non-inductive current drive in the mid-radius region with waves in the lower hybrid range of frequencies are being implemented on the DIII-D tokamak. One system will launch fast waves at 0.48 GHz (`helicons') from a 30-element antenna on the outboard side of the torus; the other will launch slow waves at 4.6 GHz from a grill on the high-field side of the torus. In each case the aim is to couple 1 MW of power to DIII-D plasmas at a launched value of the parallel index of refraction n$_{\mathrm{\vert \vert \thinspace }}\approx $ 3. The high value of n$_{\mathrm{\vert \vert }}$, with its challenge for efficient coupling, is needed for core accessibility at high density at the toroidal field of DIII-D (B$_{\mathrm{T}}$\textless 2.2 T) for the 4.6 GHz slow wave, while for the 0.48 GHz helicon, a value of n$_{\mathrm{\vert \vert }}$ at least this high is needed for strong Landau damping at attainable electron pressures in DIII-D. (Lower values of n$_{\mathrm{\vert \vert \thinspace }}$would be appropriate for reactor plasmas for both waves.) Analysis verifies that the evolution of n$_{\mathrm{\vert \vert }}$ along the ray is decisively affected by poloidal inhomogeneity for both waves. The similarity in behavior can be traced to the essentially geometric origin of the effect. Consequences for wave accessibility to the core and the location of strong damping are analyzed for the conditions of the upcoming DIII-D experiments.   

Off-axis Current Drive via High Field Side Lower Hybrid Current Drive in DIII-D

A high field side launch lower hybrid current drive (HFS LHCD) system is nearing completion. The guiding physics criteria is to drive off axis current drive, $\rho \sim$0.6-0.8, with peak current density approaching 0.4 MA/m$^2$ in DIII-D AT discharges. HFS launch position was selected to improve wave penetration, allow for single pass absorption and off-axis deposition. In AT discharges, good wave penetration is achieved because the poloidal upshift balances the toroidal down shift as the wave penetrates into the plasma. Near the damping location, the wavenumber upshifts quickly resulting in localized, $\sim$0.2a, absorption and current drive. The technical challenges to implement HFS LHCD coupler in DIII-D were substantial. The coupler and waveguides enter on the low field side and follow the vacuum vessel contour up to the center post near the mid-plane. The expected disruption loads, 400C bake, and detailed RF structures compelled the use of an additive manufactured, high temperature copper alloy, GRCop-84. Further, the waveguide routing utilizes two compact vacuum-RF flanges tested to withstand $>$2 years accumulated bake time and 5 years worth of thermal cycling. The latest simulations, design and system status will be presented.   

Full-wave Simulation of High Field Side Scrape Off Layer Reflectometry and Lower Hybrid Coupling on DIII-D

The scrape off layer (SOL) density profile plays an important role in the coupling of lower hybrid waves into the core plasma for non-inductive current drive in tokamaks. When studying this wave coupling, it is crucial that the SOL density profile is accurately measured in experiments and proper numerical techniques are used in simulation. For the upcoming high field side (HFS) LHCD experiment on DIII-D, a compact reflectometer has been designed to measure the SOL density profile near the launcher with high temporal and spatial resolution. The final reflectometer design is presented here, including full-wave simulation of the reflectometry measurement using COMSOL. For the simulation of the LH-SOL interaction, full-wave simulations are required as ray-tracing assumptions are no longer valid in the SOL due to the proximity to the cutoff layer and the small length scales of the density profile in the SOL relative to the lower hybrid wavelength. To study the SOL's effect on LH coupling, the full-wave code PETRA-M is used to simulate the DIII-D HFS lower hybrid launcher and the SOL. This is coupled to the ray-tracing package GENRAY/CQL3D for simulation of core physics. Preliminary results of this simulation work and its insight for upcoming experimental runs on DIII-D is presented.