Session NP11: Poster Session V: MFE:DIII-D and conventional tokamaks I;Heating and energetic particles;ITER, HBT-EP, and tokamak control HED: Measurements and analysis in HED plasmas Fundamental: Fundamental processes in plasmas Mini Conference:Experiments in Lab and Space MFE: Measurement and diagnostics techniques 9:30 AM - 12:30 PM

Upgrading DIII-D to Close the Gaps to Future Fusion Reactors

The DIII-D program is pursuing an ambitious plan to rapidly close critical design gaps to a Fusion Pilot Plant (FPP), including integrating performance and exhaust solutions, addressing plasma interacting material and technology issues through an expanded R&D effort, and resolving a high fusion gain path to ITER and a pulsed/steady-state FPP. Key to the DIII-D approach is major facility upgrades to access reactor-relevant physics regimes with increased flexibilities. A staged divertor program will allow for more plasma shaping, and higher current and density, to enable more reactor-relevant core-edge integration. The electron cyclotron heating (ECH) will be increased to 10 gyrotrons, with an extension to 20 gyrotrons by 2028, to furnish low-torque electron heating and profile control, while new reactor-relevant solutions for efficient off-axis current drive will be pioneered by high-field-side lower hybrid current drive, helicon waves and top launch ECH to enable FPP steady-state scenarios. Additional proposed facility upgrades include an expanded set of 3D coils, and disruption mitigation and materials testing capabilities. An exciting option for a negative triangularity path is being assessed.

Together these elements will enable reactor solutions to be pioneered and projected with confidence.   

Measurement of Stark-split beam and Carbon charge exchange emissions for simultaneous B-field and temperature/rotation analysis at DIII-D

A set of two newly designed, single-channel Czerny-Turner spectrometers [1] has been deployed at the DIII-D tokamak for simultaneous measurements of the full Motional Stark Effect (MSE) Split Deuterium or Hydrogen Beam Emission and the Carbon CVI charge exchange emission at high spectral (0.13 nm) and temporal resolution (2-10 kHz). The MSE emission allows for spectral fitting of the magnitude and direction of the local B-field, while the Carbon emission yields local ion temperature and toroidal rotation information. A new channel-specific lens-masking approach has been developed and installed to reduce the doppler broadening effect for the MSE emission. To that end, data has been collected from the 2023 DIII-D experimental campaign, focusing on measurements of fast changes in the magnetic field structure due to modifications of the bootstrap current – as well as rotation changes – during the ELM cycle. Initial results towards the evolution of the toroidal rotation and bootstrap current during ELMs will be presented. This requires fitting of the MSE emission, which involves generating tables of pyFIDASIM spectra and accounting for non-equilibrium excitation populations using a stark-shifted quantum collision model.

 

[1] Albosta R et al.: Rev Sci Instrum. 2022 Nov 1   

Turbulence-driven Transport and Spreading at the Edge of DIII-D L and H-Mode Plasmas.

Turbulent radial particle flux, heat flux, electron intensity flux, and turbulence energy flux are measured at the edge of DIII-D in L0mode, during the transition from L to H-mode, and the resulting H-mode plasmas. The radial turbulent particle and energy fluxes are outward both inside and outside the LCFS, but the turbelence energy flux is negative inside the LCFS and outwar/positive in the SOL> All the fluxes are reduced during the transition to, and during H-mode. This work aims to test theories of turbulencespreadingthat mayresult in a mechanism to control the SOL width, i.e. the ehat flux footprint in tokamak plasmas.

The rms levels of density and electric field fluctuations decay monotonically with radius towards the SOL and there is a 8X reduction of fluctuations and fluxes magnitude in H-mode.

The change of direction measured in the turbulence energy flux is consistent with the region of inward-traveling holes, marked by negative skewnessm slightly inside the LCFS thus the energy flux is heavily influene by the holes, consistent with holes propagating radially inwards carrying colder plasma at lower density. The hole-blob dynamics has been understood as the result of interchange instabiities in the edge   

Status of the HFS LHCD System on DIII-D

               The DIII-D High Field Side (HFS) Lower Hybrid Current Drive (LHCD) System will validate a novel approach for RF noninductive off-axis current drive for r/a~0.6-0.8 . The launcher centerpost placement improves wave accessibility and penetration while reducing plasma interaction issues and associated coupler damage.

A 2 MW power system with eight 4.6 GHz klystrons has been installed. The high voltage power supply, cooling water circuits, and instrumentation and control commissioning has been completed, while the eight klystrons commissioning and conditioning is underway. All the eight klystrons are operated with power injection in the dummy loads with power above 20 kW for each system. The pulse length was extended to 100 ms for all eight systems and klystron conditioning continues to increase pulse length and output power.  The system is expected to be ready for experiments in FY24.

The LHCD HFS launcher components and in-vessel waveguides were additively manufactured (AM) GRCop-84. The waveguide and launcher assembly has been temporary installed on a 1:1 scale representation of the DIII-D centerpost. The LHCD instrumentation and diagnostics include: optical arc detection, impurity monitor for plasma-material interaction, pyrometer to monitor thermal response , thermocouples, visible camera monitors, Langmuir probes, and reflectometer for density profile measurements.  The latest progress on klystron commissioning and coupler installation will be presented. 

    

Internal measurements of magnetic and density fluctuations from magneto-hydro-dynamic (MHD) to sub-ion cyclotron frequency range using radial interferometer-polarimeter in DIII-D

Tokamak transport is believed to be dominated by fluctuations such as those arising from Magneto-Hydro-Dynamic (MHD) instabilities, turbulence and Energetic-Particle-driven instabilities. Internal measurements of radial magnetic (δb) and density (δn) fluctuations from MHD to sub-ion cyclotron frequency range have been achieved with high resolution (δb/b~O(10^-5), δn/n~O(10^-6)) using a radial interferometer-polarimeter in the DIII-D tokamak. Coherent magnetic and density fluctuations with multiple harmonics in the 10-100 kHz range are usually observed before and after a sawtooth crash. Broadband magnetic and density fluctuations from 10 to 600 kHz exhibit a direct link to global plasma confinement: suppression of 10-100 kHz magnetic and density fluctuations closely correlates with low- (L-) to high-confinement (H-) mode transitions, whereas a ~50% reduction of 200-600 kHz magnetic fluctuation amplitudes is measured during transition from L- to improved-confinement (I-) mode. In sub-ion cyclotron frequency (3-4 MHz at ~1 T magnetic field), density fluctuations with a mixture of continuous and bursting behavior are measured to correlate with neutral beam injection. These measurements will help develop a better physics understanding to optimize transport in a fusion reactor.   

Top Launch ECCD Experiments in High-qmin AT Scenario Plasmas on the DIII-D Tokamak

Studies of many tokamak reactors have shown that non-inductive off-axis current drive is a requirement for a steady-state device in the AT regime. Following the successful demonstration of doubling off-axis ECCD via top launch ECCD compared to low field side (LFS) launch in L-mode and low-b H-mode plasmas on DIII-D using a top launcher installed at toroidal 300 deg, a second top launch system was installed at toroidal 90 deg. Current drive experiments in moderate betan (β_N~3) high qmin (>2) AT scenario plasmas using two top launch systems are conducted. Top launch ECCD deposition are measured by ECE system utilizing modulated ECH while top launch ECCD current drive are measured using MSE system utilizing constant EC waveform. Measurements are compared with prediction using ray tracing code TORAY and quasi-linear Fokker-Planck code CQL3D. Comparisons with current drive aimed at the same location via LFS launch in these plasmas are also been made. Both top launchers on DIII-D have fixed aiming and are identical. Numerical studies are carried out to investigate how to improve the absorption, current drive efficiency and narrow top launch ECCD deposition profile in these plasmas.   

Doppler backscattering measurements of radio frequency plasma waves at DIII-D

A novel Doppler backscattering system (DBS) has been developed to investigate radio frequency (RF) waves in DIII-D plasmas. This simultaneously measures density fluctuations from turbulence (f < 10 MHz) and RF waves within a selectable range (20-550 MHz). DBS has been used to probe ion cyclotron waves (f ~ 20 MHz), Alfvenic instabilities (e.g., CAEs), externally launched waves (476 MHz) from DIII-D's helicon antenna, and naturally occurring lower hybrid waves (~500 MHz). The mm-wave launch frequency (O or X-mode) can be remotely stepped over 60-90 GHz, with adjustable dwell time. It can also be steered horizontally and vertically, allowing probe location and wavenumber scan during a single discharge. DBS provides sub-ms temporal resolution and wavenumber coverage k_θ ~ 1-25 cm^-1. Recent data found broadband fluctuations near the helicon frequency (476 MHz) during high-power helicon experiments. Preliminary interpretation suggests this is due to the mm-wave backscattering from plasma turbulence modulated by the helicon wave. In long term, DBS aims to validate predictive modeling (GENRAY or AORSA) of helicon current drive in DIII-D plasmas, assessing wave propagation, amplitude, absorption, and current drive location.   

Effect of helicon RF injection on turbulence and transport in the pedestal and SOL regions of the DIII-D Tokamak*

Helicon waves are high harmonic fast waves (500-1500 MHz) designed for off-axis current drive in DIII-D and other tokamaks including DEMO. Helicon RF (~476 MHz) injection is found to modify the pedestal and SOL turbulence as well as the electron density profiles in H-mode plasmas. These modifications are seen 45-60° toroidally away from the helicon antenna with little-to-no change in turbulence 215° toroidally away, indicating a toroidally localized effect. GENRAY simulations of N_|| = 3 helicon indicate the wave makes at least two toroidal passes before it is fully absorbed. Flow velocities and density turbulence (k_θρ_s ≈ 0.15–0.61) amplitudes ñ (from Doppler backscattering) show an increase (~x10) in edge turbulent flow accompanied by smaller increases (≈10%) in the SOL ñ, with little-to-no change in flow velocities closer to the separatrix.  Density profiles show an increase in density with a decrease in density gradient in the SOL during helicon injection, with T_e showing no change. Divertor floor Langmuir probes show a slight reduction in both electron density and temperature during helicon wave injection. Density profile and fluctuation changes are significant and their effect on RF propagation, coupling, transport, etc. will be reported on.   

Overview of DIII-D Helicon Program

Experimental studies of the physics and technology of tokamak plasma heating and current drive with fast waves in the lower hybrid range of frequencies, referred to as "helicons", are ongoing at DIII-D, with power levels of up to 0.8 MW at 0.48 GHz from a klystron source applied to a 1.5-m-wide comb-line traveling wave antenna with 30 elements. Experiments in 2022-23 have focused on measuring the power deposition on electrons in the core of L- and H-mode plasmas preheated with neutral beams and ECH to increase helicon absorption. As designed, the antenna input impedance is "load resilient" to L-to-H transitions, to ELMs, and even to whether a plasma is present. Conditioning of the antenna was necessary for reliable power delivery; observed properties of this process are described. No deleterious effects such as significant impurity generation or a density rise have been observed to correlate with the antenna being powered. Accompanying posters in this session present details of the measurements, improvements in the technical aspects of the system, and of the development of new specific diagnostics to observe the effects of the helicon waves on the plasma in both the antenna near-field and core regions of the plasma.   

Power deposition measurements during high-power helicon experiments in DIII-D L-mode and H-mode plasmas

Helicon current drive, also called fast wave current drive in the lower hybrid range of frequencies, has long been regarded as a promising current drive tool for reactor grade plasmas. Experiments at DIII-D are underway with a 30-module traveling wave antenna of the comb-line type operating at 476 MHz with n_|| = 3, optimized for DIII-D high-beta target plasmas. To date, up to 0.7 MW has been applied to the input of the vacuum transmission line feeding the antenna in both L-mode and H-mode plasmas. Recent experiments have focused on power deposition measurements looking for a locked-in electron temperature response to modulated helicon power at various modulation frequencies. The core electron temperature was varied using core-absorbed ECH to alter the helicon first-pass absorption. Comparison shots were acquired with modulated ECH, both with strongly absorbed 2^nd harmonic X-mode and with weakly-absorbed 2^nd harmonic O-mode. Estimates of the absorbed power in the plasma are obtained from electron temperature fluctuation data cross-correlated with the modulated power. These estimates are compared to predictions from the GENRAY raytracing code.

[1] B. Van Compernolle, et al, Nucl. Fusion 61, 116034 (2021)   

Time-dependent integrated modeling of high power helicon experiments on DIII-D

Helicon current drive is a potential solution for driving off-axis current for advanced scenarios in reactor conditions. Dedicated DIII-D experiments have recently been conducted to assess power coupling, absorption, and current drive for a MW-level helicon system. Time-dependent modeling was performed with TRANSP in order to aid experimental interpretation and also validate physics models for future experimental design and physics studies. Within TRANSP, the helicon power deposition and current drive is calculated via the GENRAY ray tracing code, while temperature fluctuations are predicted with the TGLF turbulent transport model. In helicon modulation experiments, cross spectral analysis techniques are used to compare ECE measurements of temperature perturbations to predictions from the simulations. Reference discharges with modulated ECH (for which well-validated forward models exist) are used as a reliable benchmark for the helicon cases. A synthetic MSE diagnostic within TRANSP is used to predict the detectable signature of helicon current drive. Lastly, injected helicon power scans are performed in TRANSP in order to constrain estimates of the fractional power lost in the edge region via parasitic mechanisms.   

Parametric instabilities during high power helicon wave injection in DIII-D.

Helicon (whistler) waves at 0.476 GHz have been launched on DIII-D with powers up to 0.6 MW for heating and driving currents in the plasma core. Previously we have shown [1] that under experimental conditions in L-mode plasmas parametric decay instability (PDI) is expected in the edge plasma region, in agreement with measurements with magnetic pickup loops that show evidence of PDI corresponding to the ion cyclotron frequency and its harmonics at the outboard edge of the plasma. In one experiment, an energetic ion tail associated with PDI was observed. PDI has also been observed in H-mode plasmas. The dominant driver of PDI is the EXB and/or polarization drift velocity associated with the helicon wave which can drive ion-cyclotron quasi-modes and electrostatic sideband waves unstable [1].  We extend numerical calculations of growth rates and frequencies to different values of the edge plasma parameters, including PDI associated with possible parasitic excitation of the slow lower hybrid wave and predict strong PDI owing to the slow wave's large E_||. Estimation of convective thresholds and PDI saturation levels is in progress.

[1] M. Porkolab et al., in 24th Top. Conf. RF Power in Plasmas, Annapolis, MD, 2022.

^*Work supported by US DOE under DE-FC02-04ER54698 and DE-SC0016154.

    

Withdrawn

A thermal helium beam diagnostic (He-Beam) has been commissioned at the DIII-D tokamak. The diagnostic aims to measure the radial profiles of plasma density, temperature, and impurities in the DIII-D scrape-off layer surrounding the helicon antenna. These measurements are vital for evaluating plasma heating methods and the general characterization of the DIII-D boundary plasma. The diagnostic is composed of two spectrometers with eighteen lines of sight each. One spectrometer measures the emission lines of thermal helium, and using a collisional radiative model (CRM), the radial density and temperature profiles can be inferred. The second spectrometer is used for impurity spectroscopy. This contribution presents the details of the implementation and first analysis of three plasma discharges as examples. The data was acquired with 10 ms temporal resolution to be comparable with helicon antenna modulation. Changes in the line emission intensity of up to two orders of magnitude were detected between scenarios with and without the helicon antenna power injection. Preliminary results indicated minor changes in impurities associated with helicon antenna operation and, after system calibration, allow plasma parameter estimation based on a dedicated helium CRM. The first results for these scenarios will be presented and discussed.   

Effects of antenna misalignment and density turbulence on helicon and slow wave propagation in DIII-D

This study explores the behavior of helicon and slow wave propagation in the DIII-D tokamak using the Petra-M full-wave simulation code. At a frequency of 0.48 GHz in DIII-D, both slow and fast helicon waves can coexist in the scrape-off layer (SOL). It is crucial to minimize slow wave excitation since they can be easily damped in the SOL. To understand the excitation and propagation of slow waves, we conduct numerical surveys by scanning the angle between the magnetic flux and antenna surfaces, the antenna misalignment angle in the toroidal direction, and the plasma density in front of the antenna (n_ant). We also address the effects of turbulence near the last closed flux surface (LCFS) on the propagation of helicon and slow waves. We generate drift-wave turbulence with the XGC simulation code and plan to implement it in Petra-M. These small-scale spatial density fluctuations can affect both long-wavelength helicon and short-wavelength slow waves. Since the slow mode can reach the LCFS at a lower n_ant, turbulence is likely to impact slow wave propagation and absorption.   

Feedback control of helicon current drive power on the DIII-D tokamak

A feedback control system has been implemented to regulate the power output of the helicon current drive system on the DIII-D tokamak. A 1.2 MW klystron delivers 476 MHz RF injected to the plasma by a travelling wave antenna (TWA). A PID controller implemented with a National Instruments C-RIO FPGA allows user-defined control of the output power. The FPGA controls the klystron's output power by actuating a digitally-controlled variable attenuator to change the input power to the klystron. Modeling employing system identification indicated a discrete controller running faster than 20 kHz could provide 20-200 Hz modulation while rejecting 660 Hz disturbances. Software was developed to leverage the C-RIO as both a discrete PID controller running at 40 kHz and a data acquisition system digitizing up to 100 kS/s to monitor performance during tuning. The controller is tuned to match set point waveforms with 0.5 ms response time. This poster describes the controller's technical development, tuning, and potential applications.   

High-Power Helicon System Upgrades at DIII-D

  Helicon current drive is a promising new technology for plasma heating and current drive-in fusion reactors. The DIII-D tokamak has one such system, which operates at 476 MHz and can deliver up to 1.2 megawatts (MW) of power. However, the system has had problems reaching high power in the past.

  This year, the DIII-D team made several upgrades to the Helicon system, which allowed it to reach at the klystron output up to 0.7 MW for 2-second pulses and up to 850kW for 0.5s pulses. One of the upgrades was to pressurize the coaxial lines with dry air to 15 PSI. Another upgrade was to install a gas break at the output of the circulator to prevent any pressure from reaching the klystron window. The team also implemented an automatic voltage ramp up for the klystron filament.

  In the coming maintenance period, the DIII-D team plans to make additional upgrades and repairs to the Helicon system. They will replace a damaged bellows on the 150-degree side of the system. Additionally, they are working on a feedback loop controller to smooth out any power fluctuations from the klystron.

  These upgrades are expected to further improve the performance of the DIII-D Helicon system and make it a more reliable and versatile tool for plasma research.   

Reducing tungsten plasma-material interactions with boron and boron nitride powders in the DIII-D V-shaped divertor

Boron (B) and boron nitride (BN) powder injections into DIII-D’s V-shaped tungsten divertor is found to curtail heat fluxes, recycling, and tungsten (W) erosion without significantly impacting H-mode plasma performance. Experiments have been conducted using the closed SAS-VW divertor incorporating a W coated outer ring in H-mode (ion B×▽B directed into the divertor) at a neutral beam power of P_NBI = 6 MW and density of 5-6 x 10¹⁹ m^-3. Injection of boron powder at rates up to 25 mg/s at the OSP in the SAS-VW divertor showed no substantial reduction in heat flux but an ~20% decrease in neutron rates and stored magnetic energy. W deposition dropped up to 50<!-- Question, perhaps, is whether a 50% drop is consistent with the drop in Te at the surface, and the subsequent drop in the W physical erosion yield? -->% while B deposition increased by over 100% for shots with <10 mg/s B powder injection, observed through MiMES and DiMES probes. BN powder, however, increased the divertor neutral pressure tenfold and reduced peak parallel heat flux by 90% near the injection point<!-- …consistent with detachment (?) -->. BN has proved to be an effective heat dissipator in both the original SAS divertor with all-carbon walls and the high-Z SAS-VW divertor, outperforming pure B powder in the closed divertor at equivalent rates [1]<!-- Is this because N is higher Z, and so a better radiator at divertor-relevant temperatures? -->. These findings confirm powder injection as a suitable method for controlling plasma-material interactions in tungsten environments in future fusion machines.   

Multi-Chord Upgrade for UV Spectroscopic Profile Measurements in the DIII-D Lower Divertor

The recently commissioned High-Resolution Ultraviolet Spectrometer (HRUVS) is being upgraded from a single sightline to multiple viewing chords to quantify the spatial distribution of fuel and impurity emission of both low- and high- Z ions and neutrals. The upgrade will additionally provide simultaneous spectroscopic profiles from the HRUVS & Multichord Divertor Spectrometer (MDS), thereby reducing repeat experimental shots needed for coverage of multiple wavelengths. Measured profiles of ion emission will allow for comparisons with predictions from boundary codes such as SOLPS-ITER, UEDGE, and OEDGE. The HRUVS is optimized for 200 to 400 nm wavelengths with high-throughput and ~0.16 Å resolution at 250 nm to investigate high-Z erosion and re-deposition physics. A new design is underway to improve the utilization of the 150R+2 viewport on DIII-D by increasing the clear aperture of the window while also improving the robustness and adjustability of the existing collection optics for the MDS and DiMES TV. The new HRUVS view at 150R+2 will provide ~12 sightlines of the lower divertor with a spatial resolution of ~1 cm. The temporal resolution will depend on individual spectral line intensities and plasma conditions but 10 to 100 ms is anticipated for most lines.   

An overview of the atomic data and spectroscopy of low charge states of W for use in Plasma Facing Component studies

Tungsten (W) spectroscopy is important for diagnosis of gross and net erosion from plasma-facing components using the SXB method. Recent years have seen a significant improvement in the quality of the atomic data for W, including electron-impact excitation data for neutral W[1], W+[2], and W2+. An overview is given of the new atomic data, along with spectral comparisons of W I, W II, and W III emission lines from the Auburn CTH experiment and the DIII-D tokamak, using a high-resolution high-throughput UV spectrometer. Of particular value are wavelength windows showing emission from multiple charge states, making UV observations critical. The new data is shown to be a large improvement in predicting line intensities compared with those from previously available atomic datasets, with the behavior of metastable states also being important. The largest atomic data uncertainty in the SXBs is the excited state ionization data, leading to a set of large-scale R-matrix with pseudostates calculations for W2+ reported here. This will be followed by calculations for W+ and neutral W.

 

    

Atomic calculations and comparison to experiment for Si+ and W2+ erosion diagnostics

New atomic physics work is being done to build diagnostics for the erosion and re-deposition of wall materials in fusion plasmas. A new ionization calculation for W²⁺, needed to determine W re-deposition rates, is currently underway using the R-matrix with pseudostates (RMPS) method. This calculation, when combined with an existing excitation calculation for W²⁺, will be used to generate ionization/photon (S/XB) coefficients, which can be combined with spectroscopic measurements to infer time-resolved erosion and re-deposition rates. To transform the collision cross sections from the RMPS calculation into accurate derived coefficients, a new code has been developed that separates each collisional cross section into separate ionization series and extends them to high energies. The techniques used for the W²⁺ calculation have been tested by applying them to a simpler, but still useful, ion: Si⁺. Both calculations have been compared to experimental ionization cross section measurements, which justifies confidence in the derived coefficients, such as the S/XB coefficients. These calculations are being benchmarked over a range of plasma conditions from experiments completed in the DIII-D tokamak using the DiMES removable sample probe and new UV spectroscopy capabilities.   

Fluctuation Measurements of Neoclassical Tearing Mode using Imaging Neutral Particle Analyzer on DIII-D

An upgrade to the Imaging Neutral Particle Analyzer (INPA) on DIII-D has been installed to add higher temporal resolution (1 MS/s) measurements using an array of photomultiplier tubes (PMTs) that view several key areas of energetic particle (EP) phase space. The new detectors measured EP transport resulting from a 5 cm wide Neoclassical Tearing Mode (NTM) that locked to the vessel wall. All three PMTs that detected the mode measured the m/n=2/1 fundamental, and a 3/2 harmonic was also picked up by one detector. Analysis of the data shows that the signal comes from charge exchange events near the edge of the plasma, where the INPA can detect a range of pitches from -0.75 ≤ v_‖/v ≤ -0.5 across the separatrix. Contributions to the signal come from both the prompt transport of the neutral beam ions as well as the slowing down distribution over an energy range from 10 keV to 50 keV. Modelling of the fast ion and edge neutral densities from TRANSP show the signal from the slowing down distribution is stronger for orbits with high pitch. Tracing of orbits through the equilibrium fields shows that the prompt signal is stronger for orbits with lower magnitude pitch.   

Identification of Alfvén eigenmodes using recurrent neural networks, a labelled database and CO2 interferometer data on DIII-D

Artificial Intelligence is developed to automatically detect Alfvén eigenmodes (AE) and these models achieve high performance (True Positive Rate = 90% and False Positive Rate = 14%). Using labels created from a curated database [Heidbrink, et al., NF ‘20], Machine Learning-based systems are trained using single chord and crosspower spectrograms to predict the presence 5 AEs (EAE, TAE, RSAE, BAE and LFM). Since resonant fast ions can drive AEs unstable and degrade the plasma performance or energy confinement, this work demonstrates the potential of applying Machine Learning methods to detect and control AEs. In this work, the advantages of using the CO₂ interferometer to detect AEs, and the results from a comparison between inputs (single chord and crosspower spectrograms) and another comparison between two different models (Reservoir Computing Network and Long Short-Term Memory Network) are presented. The highest performance is achieved by the Reservoir Computing Network trained with single chord spectrograms. Also, AE detection using any chord is feasible (the vertical chord passing near center is best). These models can be implemented into control algorithms that drive actuators for the mitigation of unwanted AE impacts. Supported by the U.S. Department of Energy under DE-FC02-04ER54698, DE-SC0021275, DE-SC0020337, DE-SC0014664, Army Research Office (ARO W911NF-19-1-0045), National Science Foundation under 1633631 and Ghent University Special Research Award No. BOF19/PDO/134.   

Effect of divertor conditions on ionization and neutral density profiles in the edge and SOL in DIII-D

We compare the ionization and neutral density profiles in the edge and scrape-off layer (SOL) in DIII-D plasmas with measurements obtained from the Lyman-Alpha Measurement Apparatus (LLAMA) diagnostic for different divertor conditions. The impact of divertor closure (USN with a closed divertor and LSN with an open divertor) and state of divertor conditions (attached/detached) are investigated in H-mode plasmas on both the low and high field side of the plasma below the midplane. For the LSN vs USN configurations, measurements are taken at different poloidal locations relative to the X-point, allowing to study the influence of the detachment on the ionization and neutral density profiles at different positions. Divertor closure plays an important role in trapping neutrals, impacting the pedestal and SOL density and temperature profiles and the density at the onset of detachment. Results will be compared with modelling SOLPS predictions from [1]. Understanding how ionization profiles relates to pedestal structure, divertor closure and detachment onset is crucial for closing the Integrated Tokamak and Exhaust Performance (ITEP) knowledge gap. This research is beneficial for our understanding of possible fueling scenarios for future machine operation like ITER and future fusion reactor.

[1] A.C. Sontag et al 2017 Nucl. Fusion 57 076025   

Magnetic field pitch angle measurement using Beam Emission Spectroscopy (BES) diagnostic on DIII-D

The magnetic field pitch angle is defined by the slope of the ratio between the poloidal and toroidal magnetic field, α=arctan(B_p/B_t). Multi-point time delay estimation can been applied for calculating the α profile if the line-of-sight is separated toroidally and poloidally. The inclination of turbulent structure is referred to the pitch angle by measuring the propagation of density fluctuation. The Beam Emission Spectroscopy (BES) diagnostic measures the two-dimensional density fluctuation by volume sampling the Doppler-shifted Balmer-alpha beam emission. The measured intensity is a convolution of the emission along the line-of-sight where the spatial resolution depends on the point spread function (PSF), and the observed volume is determined by the intersection of the neutral beam volume and the collection optics. On DIII-D, the tangential neutral beam gives rise to a toroidal separation for the BES measurement, which enable us to calculate the pitch angle assuming a Gaussian distribution of the beam emission in the observed volume. The pitch angle image in the plasma edge region is measured. The α images during the L-H transition and ELM are presented.   

Analysis of beam-driven instabilities near the ion cyclotron frequency in DIII-D plasmas

Frontier Science experiments in 2018 and 2021 studied instabilities above and below the ion cyclotron frequency Ω_i by injecting ten different neutral beam populations into L-mode plasmas with different thermal compositions of H, D, and 3He at five different values of toroidal field. Magnetic loops [1] diagnose mode properties. Below Ω_i, modes are more unstable at low field and the amplitude increases with increasing H concentration. Above Ω_i,the modes are stronger at high field and increasing H concentration shifts the dominant harmonic to higher values. Instability is driven more strongly for on-axis neutral beam injection (which creates a more centrally peaked beam ion profile) than for off-axis injection. Database results are compared with multi-species extensions of formulas by Lestz [2] below Ω_i and with Cook’s magnetoacoustic code [3] above Ω_i.

[1] G.H. DeGrandchamp et al., RSI 92 (2021) 033543.

[2] J.B. Lestz et al., NF 61 (2021) 086016.

[3] J.W.S. Cook,  PPCF 64 (2022) 115002.   

Stationary I-modes in multiple configurations in the DIII-D tokamak

By using β feedback to compensate for improving τ_E, longer, more stationary I-modes were obtained than in past DIII-D experiments.  These experiments were conducted as part of the 2022 Joint Research Target campaign on intrinsically non-ELMing enhanced confinement regimes. I-modes feature a temperature pedestal but low particle confinement.  Configurations tested were a LSN, low triangularity δ shape matching an ASDEX Upgrade I-mode discharge and a higher δ USN shape with better pumping in the closed divertor.  Both had B×∇B drifts away from the X-point, as usual for I-mode. In the LSN shape, the upper limit for heating power (3.5 MW) and pressure (β_N=1) was set by I-H transitions. In the USN shape, it was limited by available power (11 MW NBI) and β_N as high as1.6 was achieved. T_e and T_i pedestals formed in both configurations, with negligible changes in density from L-mode.  The Weakly Coherent Mode has not yet been identified in these discharges; further analysis is ongoing.  Many other turbulence and fluctuations changes were observed, including a reduction of density fluctuations at intermediate k, a decrease of magnetic turbulence at 100-500 kHz, and bursts of n~20 MHD in T_e on ECE Imaging.   

Deep neural network based real-time ELM prediction and reconstruction of turbulent flow based on the DIII-D BES measurement

Since ELM crashes cause losses and fast relaxation of the pedestal energy and profiles, it is crucial to predict them beforehand to achieve high-performance and safe tokamak operations. With 2D turbulence and eddy flow measurements of electron density given by the DIII-D BES system, we develop a neural network which is able to predict not only ELMs several msecs before their onsets but the first ELMs after the L-H transition under about 0.8 true positive rate. Since BES configuration with a time window containing the spatiotemporally fluctuating density is used as the network inputs, it is also necessary to know what information the network obtains from its input. Thus, we train another network to prove that the turbulent velocity fields are one of the information being able to be captured by the network. As a preliminary result, we use the time-delayed cross-correlation to generate the ground truth of radial velocity profiles as a training dataset. Therefore, we expect that our network is possible to extract the velocity information from the BES signals and provide various velocity-related characteristics of plasmas in real time.   

Integrated modeling of Core, Edge pedestal and SOL using super H-mode experiments in DIII-D

A theory-based integrated modeling of Core, Edge pedestal, and Scrape-Off-Layer (CESOL) code has been successfully tested against existing DIII-D super-H mode discharges for an optimized pedestal regime to simultaneously improve core performance and control plasma heat and particle exhaust. Recent DIII-D experiments utilized advanced control algorithms to expand the operating space of the super-H regime using combination of plasma shaping, nitrogen impurity seeding, deuterium gas puffing, and 3D magnetic perturbations. The core, edge pedestal and SOL regions are strongly coupled but governed by different physical processes, emphasizing the need for a quantitative understanding of the trade-offs or integration. CESOL can reproduce the experimentally measured plasma density and temperature profiles reasonably well across the regions from the magnetic axis to the scrape-off-layer by integrating three independent, compound IPS workflows: IPS-FASTRAN for core, IPS-EPED1 for edge pedestal, and IPS-SOLPS for SOL. Based on its reasonable interpretive capability in accordance with experimental findings against super-H mode plasmas on DIII-D, CESOL will be employed for predictive modeling to optimize both the core and edge simultaneously for DIII-D shape and volume rise studies.   

Study of the halo current region resistivity on the DIII-D tokamak

In this work we report measurements of the temperature and density of the halo current region on DIII-D during disruptions using the recently upgraded Thomson scattering diagnostic allowing low-temperature measurements down to a few eV with a sub-ms repetition rate. This is done by employing deliberate downward vertical displacement events (VDEs) and relying on the expansion of the halo current region upward, intercepting the core Thomson channels. Both 'hot' and 'cold' VDEs were studied using ohmic and H-mode target plasmas respectively. 'Hot' VDE, having the vertical instability growth rate greater than the current decay rate, characterizes by the electron temperature of the halo expansion region in the range of 1−10 eV. While 'cold' VDE has the opposite time scale dynamics and much lower electron temperature of 1−2 eV. VDE of both types results in the electron density of the halo region comparable with the core plasma density and quickly decreasing to the edge. Poloidal halo currents, measured using the tile current shunts, exhibit values greater by about 20% for the 'hot' VDE cases. Modeled poloidal and toroidal halo currents, as well as modeled halo current width are also presented. Implication of the halo region resistivity on the JxB forces is discussed.   

Divertor dissipation in the DIII-D V-shaped slot divertor with strong external heating

Experiments in the DIII-D V-shaped slot divertor confirms the essential roles of target shaping, B_T and I_p on divertor dissipation with strong external heating (P_NBI~9MW). For ion B×▽B drift into the divertor, upstream density at detachment onset (T_e < 10eV) is lower by 7-12% with the outer strike point at the inner slanted baffle instead of the slot V-end. SOLPS-ITER was able to quantitatively reproduce the measured trends, indicating that the enhanced E×B drift flows lead to better dissipation and a double-peak T_e profile. Upstream density required for complete detachment remains nearly the same, regardless of the strike point position and heating power. The detachment threshold increases almost linearly, by ~35%, when I_p increases from 1MA to 1.25MA, independent of the strike point position. Reversing the B_T(ion B×▽B drift out of the divertor) shows clear benefits of lower detachment threshold, consistent with previous SAS experiments. However, the degree of detachment is less profound with target T_e still around 5eV. In-slot impurity injection of N₂ significantly reduced (~25%) the detachment threshold, accompanied by ~ 15% less confinement degradation, in both B_T directions, and in addition can aid complete detachment with ion B×▽B drift out of the divertor.   

Impact of the Isotope Mass on Divertor Detachment and Pedestal Fueling in DIII-D

Experiments and SOLPS-ITER modeling have been conducted to understand the effect of the isotope mass on the pedestal and divertor region. Comparable discharges were produced by matching the line averaged core density of Hydrogen (H) shots to Deuterium (D) references in both H-mode and L-mode. Access to H-Mode required higher injected power in H and resulted in a lower normalized beta compared to the D references. The pressure pedestal degradation is more severe at higher gas puffing in H versus D despite the higher pedestal electron densities and density gradients in H. Pedestal stability analysis shows reduced peeling-ballooning stability in H with both the H and D discharges being ballooning limited. SOLPS-ITER modeling with full drifts and matched upstream profiles predicts H detaches at a higher upstream density. The particle source from neutral ionization inside the LCFS is also higher in the H simulations than in the D simulations. Direct comparisons are made between LLAMA spectrometer brightness measurements and that from a synthetic Ly-α diagnostic in the simulations. The inferred ionization and neutral density profile from LLAMA will be used to assess the models accuracy in predicting these parameters.   

Experiments on plasma detachment in the DIII-D V-shaped slot divertor

Experiments in DIII-D demonstrate that the upstream plasma density to detach an un-pumped slot divertor is similar for a V-shaped and a flat-end slot, despite significantly higher neutral pressure in the V-shaped slot and in contrast to SOLPS-ITER predictions. However, the upstream density for detachment is reduced by using in-slot instead of main-chamber gas fueling or by placing the strike point on the inner slanted slot baffle instead of the slot end. In 2021, DIII-D modified its flat-end Small Angle Slot (SAS) divertor into a V-shaped slot (SAS-VW). H-mode experiments show that detachment onset occurs at the same upstream plasma density in SAS-VW as SAS, in each B_T direction, despite higher neutral pressure, neutral compression and particle fluxes in SAS-VW, and in contrast to SOLPS-ITER drift modeling. Using in-slot gas fueling instead of main chamber fueling lowers the upstream plasma density to detach by ~22% / ~7% for strike point on the vertex / inner slant, and with ion B×▽B drift into the slot, consistent with SAS data and SOLPS-ITER. Placing the strike point on the inner slanted baffle instead of the slot V-end also reduces the detachment density by ~12% / ~8% for ion B×▽B drift into / out of the slot, an effect consistent with SAS data but not captured by SOLPS-ITER.   

Comparison of divertor performance dependence on geometric configuration in DIII-D and MAST-U

A comparison between DIII-D and MAST-U has been initiated to examine divertor detachment performance dependence on divertor magnetic geometry, leg length, and baffle structure. The ultimate goal of this work is to determine the optimal divertor configuration for operating a detached divertor with complete heat flux dissipation, while maintaining a high temperature X-point for a high-performance core plasma. Initial work has included the successful control of the magnetic geometry for advanced divertor configurations in MAST-U, including strike-point and poloidal flux expansion control in the Super-X configuration. Modeling with the 2D fluid code SOLPS-ITER is examining detached divertor conditions in DIII-D and MAST-U with a long divertor leg and divertor baffling. Comparison of initial experimental results from DIII-D and MAST-U are presented. Primary differences for DIII-D higher magnetic field and power density, but with a vertical divertor leg compared to MAST-U's Super-X configuration. A focus of the initial studies is the poloidal electron temperature gradients in the divertor and its dependence on magnetic configuration, divertor baffling and power density. Results of this work are expected to guide future divertor design, including upgrades to the DIII-D divertor.   

Achieving Low-Collisionalty Small/Grassy ELMs in DIII-D High-Performance Hybrid Scenario Plasmas

Low-collisionality(nu_e*~0.1-0.4) small/grassy ELMs have been achieved in recent DIII-D high-performance hybrid scenario plasmas (beta_N~3.5, beta_p~2.0, H_98~1.6), with ITER similar shape and favorable direction. Stability analysis found that grassy ELMs with high-frequency >400Hz and small ELM energy loss <1.0% are located near the peeling boundary. The high >1.5 and low pedestal top collisionality nu_e*=0.1-0.4 are important in obtaining grassy ELMs on DIII-D. A branch of low-collisionality small ELMs is observed and its ELM size increases with the collisionality increase, which is opposite to the typical type-I ELM scaling. Long stationary phases of grassy ELMs have been achieved by using the n=3 RMP of 1.5kA I-coil current, without occasional large ELMs and without clear density pump out. Additionally, ELM frequency is further increased while the ELM size is further decreased for the RMP-assisted grassy ELMs. Divertor detachment was achieved with divertor N₂ puffing, while high confinement with beta_N>3.0 is maintained. However, the pedestal density increases during detachment, and large ELM occurs. The study sheds light on a promising scenario with high core confinement, good edge impurity exhaust, and acceptant boundary heat load for ITER and future machines.   

Measurement of sawtooth induced fast ion transport on DIII-D with a suite of imaging energetic particle diagnostics

Transport of neutral beam generated fast ions in the presence of sawtooth crashes is investigated experimentally at the DIII-D with a suite of newly developed imaging energetic particle diagnostics, which includes two Imaging Neutral Particle Analyzers (INPAs) and one Imaging Fast-Ion D-Alpha (IFIDA) diagnostic. The INPAs and IFIDA measure fast ions that charge exchange with injected beam neutrals and thus provide the local fast-ion distribution. The two INPAs are sensitive to passing and trapped fast ions with energy > 20 keV. The IFIDA integrates fast-ion density in the energy range of 40-80 keV and provides a spatial profile of passing fast-ion density with high spatial resolution. The INPA and IFIDA images before and after sawteeth show that high energy (E>40 KeV) passing fast ions are strongly redistributed from the core to the region outside of the q=1 surface, and the central fast ion density can be reduced by ~30%. Lower energy (E<40 keV) passing fast ions have a similar trend, but the relative change is <10%. The trapped particles in all energies are weakly affected, and a few percent increase of trapped particles in the region just outside the q=1 surface is observed. Tomographic inversion of the fast-ion distribution as well as the comparison with different sawtooth models will be presented.   

Measurement of bursting behavior at sub-ion cyclotron frequencies in DIII-D using Radial Interferometer Polarimeter (RIP) diagnostic

The study of Global/Compressional Alfv´en Eigenmodes (GAEs/CAEs) is relevant for future reactors which would have a large concentration of highly energetic particles. In present day tokamaks, these modes are driven by the energetic particles from Neutral Beam Injection (NBI). The magnetic field of these modes displays sharp periodic change in its amplitude (or bursts), the mechanism of which remains unclear. Recent upgrade in the Radial Interferometer Polarimeter (RIP) diagnostic at DIII-D enables the simultaneous measurement of density and radial magnetic field at frequencies close to the ion cyclotron frequency (f_ci). Fluctuations with frequencies 0.25 < f/f_ci < 0.65, typically associated with GAEs/CAEs, are observed to display bursts in time signals, resulting in their Probability Density Functions (PDFs) deviating from a Gaussian distribution to one with exponential tails. The characteristic density  amplitude of these bursts (n_e^c) normalized by the equilibrium plasma density (n_e) is nce/ne ∼ O(10⁻⁵ − 10⁻⁴). We find the timescales of growth (τ_g ), decay (τ_d) and waiting time (τ_w) between bursts respectively to be τ_g ∼ τ_d ∼ O(0.1) ms and τ_w ∼ O(1) ms. These observations are found to hold true for both positive and negative triangularity configurations of plasma.   

X-ray Fluorescence analysis of global tungsten transport inside DIII-D tokamak during tungsten-coated, V-shaped Small Angle Slot divertor campaign

After the DIII-D Small Angle Slot, V-shaped, W-clad divertor (SAS-VW) campaign, tungsten (W) deposition (<10 nm) was measured on tiles outside of the divertor. Handheld X-ray fluorescence (XRF) analyzers were used to quantify the tungsten and other metallic species, that sputter or melt off of surface tiles and structural materials. XRF has been used to identify alloys, detect trace impurities (ppm), determine coating thicknesses, and other applications requiring bulk analysis. Post-campaign measurements were taken at various different toroidal and poloidal locations (including those just outside of SAS-VW) and compared with pre-campaign calibrations. Although low W deposition outside of the divertor was observed, larger W deposition ratios in relation to iron (Fe) and nickel (Ni) were evaluated at the uppermost and bottommost center post tiles. Grinding paper proved successful in removing all measurable tungsten and the majority of Fe and Ni from exposed tiles. Laser-based coordinate measuring machine (CMM) data was used to assess tile-to-tile misalignment based on post-mortem observations of large W erosion on tile edges. CMM data was also used to investigate tungsten deposition from the W-clad divertor tile to nearby V-tiles.   

Spatial Heterodyne Spectroscopy for Motional Stark Effect measurements at DIII-D

One technique to explore the magnetic field structure of fusion experiments is Motional Stark Effect (MSE) spectroscopy, wherein the Stark splitting of the Balmer alpha emission of neutral beams is analyzed. A recent technique to spectrally resolve this radiation is Spatial Heterodyne Spectroscopy (SHS), offering high photon throughput and high spectral resolution. While previous iterations of the SHS system [1] at DIII-D focused on high-speed measurements, the spectrometer has been revised for improved optical performance and the ability to resolve both the splitting and the sigma-to-pi line ratio of the Stark manifold. The revision of the spectrometer entails a reduction in optical components, and a novel technique to compensate for geometric Doppler broadening of the beam emission. Moreover, a calibration technique using a HeNe laser ($lambda$ = 632.8 nm) has been developed that allows one to precisely align the two gratings of the SHS system. To that end, comparisons of the calibration with a newly developed SHS forward modeling code and one-to-one performance-comparisons with conventional Czerny Turner spectrometers will be presented.

[1] M.G. Burke et al.: Rev. Sci. Instrum. 94 (2023)   

Overview of Research Results from the SciDAC Partnership for Simulation of Fusion Relevant RF Actuators

An overview is given of research related to the interaction of ion cyclotron range of frequency (ICRF) power with the tokamak scrape-off layer (SOL) and core. We discuss the development and application of the RF full-wave solver Stix and the more general FEM analysis platform Petra-M, both based on the Modular Finite Element Framework (MFEM), highlighting work on implementation of high-fidelity geometric descriptions of ICRF antennas in the NSTX-U and WEST Tokamaks, incorporation of nonlinear RF sheath boundary conditions, matrix pre-conditioners, and progress on implementation of our RF codes on GPU's. We will review progress on the development of a far-SOL fluid transport solver for equilibrium (the MAPS MFEM mini-app) and the study of RF ponderomotive force effects on the edge / SOL using a combined edge transport and RF solver model (UEDGE and Vorpal). Results for the interaction of RF power with metallic surfaces will be presented, including PIC simulations of the ion energy-angle distributions at ICRH antennas in RF sheaths as well as a description of an integrated workflow for RF impurity generation and transport. Finally, we will report progress on coupled Fokker Planck / full-wave simulations for ICRF using both continuum and particle-based approaches.   

Benchmarking 1- and 2-D fluid plasma transport solutions in MAPS (MFEM Anisotropic Transport Solver)

Coupled radio frequency (RF) / plasma transport interactions in the far scrape-off layer (SOL) of a magnetized fusion plasma are thought to drive a number of nonlinear effects close to actuators, including RF enhanced sheath driven EXB convection and ponderomotive forces. Self-consistent modeling is a necessity for robust description of these effects and their impact on heating performance. However, reliably meshing both the plasma and actuator in the far-SOL region is difficult due to the requirements of each: field aligned for the anisotropic plasma transport, and geometry conforming for the RF antenna. In this work we present recent updates and improved benchmarking of the MAPS code. Results from additional 2D, anisotropic heat conduction benchmarks with [Umansky 2005, 2008] are shown, along with 1D comparison with the SD1D code. We will discuss recent updates to the MAPS algorithm, along with planned verification and validation exercises that will include coupling with the MFEM Stix2d electromagnetics solver.

[Umansky 2005] M. V. Umansky, M. S. Day, and T. D. Rognlien, “On numerical solution of strongly anisotropic diffusion equation on misaligned grids”, Numerical Heat Transfer, Part B: Fundamentals 47, 533 (2005).

[Umansky 2008] M. V. Umansky, R. H. Cohen, L. L. LoDestro and X. Q. Xu, “Suite of verification test problems for edge turbulence simulations”, Contributions to Plasma Physics 48, 27 (2008).   

RF rectified sheath calculations in 3D RF full wave simulations with the Petra-M framework

During ICRF operation a high DC potential due to RF sheath is considered a root cause of the deleterious accumulation of high-Z impurities in the core plasma. Quantitative predictions of the RF sheath induced potential that develops on the plasma facing components (PFCs) are limited with the existing codes due to the use of a 2D antenna geometry. In this work, a new form of RF sheath physics is developed using magnetic potentials allowing us to predict for the first time the RF sheath potential in 3D RF full wave simulations [S. Shiraiwa et. al, Nucl. Fusion 63 026024 (2023)]. Our result using a 3D antenna geometry predicts that the sheath potential of 100 - 200 V is generated for 1 MW of RF power injection. Application to different devices and antennas are in progress and will be discussed.   

GPU-Accelerated Modeling of RF Propagation in a Full Tokamak Plasma

We have successfully ported the time-domain plasma model [1] in the VSim software [2,3] to run on large scale parallel GPU clusters.  We present characteristics and benchmarks of the new algorithm’s operations.  We demonstrate use of the algorithm on the ITER ICRF antenna and wall geometry, including diverter, with 2^nd order (cut-cell) accurate CAD defined geometries. The time-domain algorithm is locally implicit, spatially explicit, so that it can step over electron time scales without the need for global matrix inversion, e.g., it is ideally suited for GPU application.  An important algorithmic improvement has also been made, and is demonstrated, which uses a sub-grid octant-averaging technique to eliminate noise on the order of 2Δx that was inherent in the original algorithm.  This eliminates the need for costly filtering operations, and stencil enlargement, which further slowed down the CPU-based calculation. Continuing upgrades to the GPU based algorithm is also reported, with addition of self-consistent RF sheath physics being a priority.

[1] David N. Smithe, Finite-difference time-domain simulation of fusion plasmas at radiofrequency time scales, Physics of Plasmas 14, 056104 (2007)

[2] C. Nieter and J. Cary. 2004. VORPAL: a versatile plasma simulation code. J. Comput. Phys. 196, 2, 448-473.

[3] www.txcorp.com   

Beyond Maxwell-Boltzmann electrons for RF microscale sheath modeling

Radio frequency (RF) sheaths are of fundamental importance in understanding boundary ICRF interactions in fusion devices. A microscale (Debye scale) RF sheath model, considered here, provides an RF sheath admittance boundary condition for global wave codes, and the sheath potential for sputtering calculations. The commonly employed Maxwell-Boltzmann (MB) electron model can be invalid at high RF frequencies relative to the electron sheath transit time, and the RF-induced jitter motion of bulk electrons in the sheath can cause direct wall impact. In this work a 1D RF sheath microscale model that generalizes the MB model for electrons is presented using a fluid framework, taking into account the electron current lost to the wall from both electron thermal and RF jitter motion. MB electrons are recovered in an appropriate limit. Challenges and progress in numerical implementation will be discussed.   

Application of MCGO-TORIC coupled codes to modeling of HHFW heating in NSTX

The 6D (optionally 5D) Monte-Carlo code MCGO [1] with RF kick operator has been further developed to allow coupling to full wave fields computed with AORSA or TORIC. The RF operator uses local quasilinear diffusion coefficients with all components of RF field and all-orders terms in Larmor radius. The MCGO-TORIC coupled modeling is applied for the case of High Harmonic Fast wave (HHFW) heating in NSTX, where fast ions (FI) are produced by NBI. Although MCGO uses a realistic NBI-to-FI deposition module, while TORIC calculations are based on equivalent-Maxwellian representation for FI, the agreement in RF power density profiles between TORIC and MCGO is very good. To facilitate a better verification between the codes, MCGO now includes an option for two-Maxwellian distributions. Furthermore, the MCGO runs are made in both zero-orbit-width (ZOW) and finite-orbit-width (FOW) configurations. In the latter case, the radial profiles of RF power density, fusion neutron rates become quite different from the ZOW runs. Also, the MCGO FOW runs allow computing of the increase in heat load to the chamber wall when RF is applied, that we show in our presentation.

[1]   www.compxco.com/mcgo.html   

Analytical, numerical, and experimental investigation of nonlinear radio-frequency sheaths

Recent progress made toward the development, implementation, and benchmarking of a self-consistent radio-frequency (RF) sheath model [1] is presented. RF sheaths form on plasma facing components during the injection of RF power in the ion-cyclotron range of frequencies (ICRF). Plasma ions are accelerated across the sheath potential where they impact the material surface causing several deleterious effects such as erosion of the wall material and the sputtering of impurities which contaminate the plasma. Sheath rectification in the WEST tokamak due to ICRF is modeled using the three-dimensional finite element code Petra-M, which includes realistic antenna, limiter, and wall geometry [2]. In simplified geometries, analytical results are leveraged to demonstrate that self-consistent RF sheath—plasma interactions host a rich set of nonlinear features such as multivaluedness, instability, hysteresis and spontaneous symmetry breaking [3]. Finally, a basic RF sheath-plasma experiment is performed on the Large Plasma Device (LAPD) which qualitatively demonstrates the existence of a nonlinear hysteresis associated with sheath-plasma interactions.   

Development of surrogate models for the TORIC ICRF spectrum solver using ML algorithms

Machine learning (ML) approaches could enable real-time capable high-fidelity radio-frequency actuator modeling. Recently Wallace et al. [1] demonstrated ML-based surrogate models applied to the RF lower hybrid current drive providing 1D current and power predictions. In this work we present power absorption predictions in the ion cyclotron range of frequencies for two heating schemes: high harmonic fast wave (HHFW) at NSTX and hydrogen minority at WEST. Accurate (i.e. R²=0.93-0.96) and fast (i.e. O(50μs)) predictions are obtained for the HHFW 1D power absorption profiles. Datasets are generated using the full-wave code TORIC [2,3]. Even though minority dataset variance is higher than that of HHFW scenario, application of principal component analysis allowed us to also reach fast predictions of deuterium, hydrogen and electron power absorption profiles, of R²=0.71, 0.85, and 0.85, respectively. Preliminary results of surrogate predictions for the 2D are also presented. Overall, these surrogates show promising results for their incorporation in integrated modeling and control algorithms.

[1] Wallace et al. JPP 88, 895880401 (2022).

[2] M. Brambilla, PPCF 41, 1, (1999).

[3] M Brambilla, PPCF 44, 2423 (2002).   

2nd order differential operator for plasma dielectric to include all-order finite Larmor radius effects

It is well known that the dielectric response to the RF fields in hot plasma is non-local, and the Maxwell wave problem is an integro-differential equation. A differential form of dielectric operator, based on the small k⊥ρ expansion, typically includes up-to the second order terms, and thus the use of such an operator is limited to the waves that satisfy k⊥ρ < 1. We propose an alternative approach to construct a dielectric operator, which includes all-order finite Larmor radius effects without explicitly containing higher order derivatives. We use a rational approximation of the plasma dielectric tensor in the wave number space, in order to yield a differential operator acting on the dielectric current (J). To demonstrate this approach, we use the Petra-M framework and solve the 1D O-X-B mode-conversion of the electron Bernstein wave in the non-relativistic Maxwellian plasma. An agreement with analytic calculation and the conservation of wave energy carried by the Poynting flux and electron thermal motion (“sloshing”) is found. The connection between our construction method and superposition of Green’s function for these screened Poisson’s equations is presented. An approach to extend the operator in a multi-dimensional setting will also be discussed.   

Multi-Pronged Applications of RF Sheath and Ponderomotive Force Modeling with the Time-Domain Plasma Model in Vorpal/VSim

We review results from the study of plasma equilibrium modifications by RF Ponderomotive forces, using coupled Vorpal (FDTD EM+plasma) and UEDGE (2D edge transport) codes.  A bifurcation was observed in the behavior, where low density scenarios push plasma out, exacerbating the low density, while high density scenarios pull plasma in, raising the density.  Also, new work involving RF rectified sheaths, for two different applications, has been done.  Coupling with Univ. of Ill’s plasma-material interaction codes [hPIC2/F-TriDYN/RustBCA /GITR] has been improved, including creation of tools to reduce the size of the surface triangulations from the time-domain code.  We are also using RF rectified sheaths for the study of convective cell behavior in the vicinity of the RF antenna.  Here the same rectified sheath potentials provide boundary conditions for the plasma potential solve, in the equilibrium and turbulence codes [UEDGE/HERMES].  Also, we continue to make improvements to our existing RF sheath model, particularly addressing the modeling of surfaces aligned with the magnetic field, with ion gyro-radius sheaths instead of Debye sheaths, especially in high field regimes.   

Coupling between the Fokker-Planck code CQL3D and the Monte-Carlo neutral particle code pyFIDASIM

As part of the RF-SciDAC effort, in this contribution we describe recent progress towards the goal of two-way (velocity-space resolved) coupling between the continuum Fokker-Planck (FP) code CQL3D [1] and the Monte-Carlo (MC) neutral particle code pyFIDASIM. CQL3D is a continuum bounce-averaged FP code with a fully non-linear collision operator. CQL3D is typically used in tokamaks to calculate the ion/electron distribution functions in presence of RF heating and neutral beam (NB) injection. pyFIDASIM, the Python implementation of FIDASIM [2], is a 3D MC code used for NB deposition modelling which tracks neutrals/excited states as they cross the plasma and provides means for resolving fast ion (FI) birth points and halo neutral dynamics.

 

This coupling effort has been motivated by the need to include velocity-space charge-exchange (CX) sources and sinks in FP calculations where non-thermal effects on both ions and neutrals are expected to be significant such as in HHFW heating of beam-produced FIs in tokamaks and CX-dominated NB deposition in mirror devices. In this contribution, we describe the use of pyFIDASIM to provide FI birth points for the CQL3D FP calculation. Results are compared with those provided by CQL3D’s own NBI deposition model NFREYA. Finally, we discuss the impact of the results, propose strategies to implement velocity-space resolved source/sink data exchanges and outline next steps for the project.

 

[1] R. W. Harvey et al., Nucl. Fusion 59 106046 (2019)

[2] B. Geiger et al., Plasma Phys. Control. Fusion 62 105008 (2020)   

Investigation of materials for radio frequency antenna plasma facing components

The interaction of radio frequency (RF) sheaths with plasma facing antenna materials is being studied to determine their erosion characteristics.  Experiments are being conducted on the Radio Frequency Plasma Interaction Experiment (RF PIE), which consists of an electron cyclotron resonance plasma source (2.45 GHz, 5 kW) with a biased and heated RF electrode.  Helium and/or deuterium plasmas (density of ~1e18/m³, electron temperature of 4-5 eV) are being used to explore both DC and RF sheath formation on materials, with average bias voltages up to 500 V.  Optical emission from the sputtered material is being used to infer sputtering coefficients as a function of bias conditions.  When compared to DC sheaths, it has been shown that the sputtering of tungsten due to RF sheaths has the effect of enhancing the erosion.  RF biasing causes a broadening of the ion energy distribution function, as calculated using the hPIC2 code, and this broadening causes enhanced erosion due to the high energy tail of the distribution.  Optical emission lines from other candidate materials, including SiC and TiB₂, are being identified, with the goal of predicting the erosion of these materials under similar conditions. Experimental details will be presented.    

Development of an integrated modeling framework for plasma-material interaction and its application on tungsten erosion and transport from the RF antenna structures in the WEST tokamak

An integrated modeling framework has been developed to study plasma-material interactions (PMI) at the radio-frequency (RF) antenna structures in various magnetic confinement fusion devices. This  requires the integration of several computational tools at various levels of physics fidelity- (1) SOLEDGE2D/SOLPS-ITER for far-SOL plasma profiles (2) Petra-M/COMSOL for RF rectified voltages at the antenna structures, (3)CAD defeaturing of the geometry, (4)GITR for ion energy angle distributions (IEADs) at the material surface, and (5)F-TRIDYN/RustBCA for material erosion yield and surface interaction physics. A 3D Monte-Carlo, particle tracker code- GITR can then predict amount of sputtering-net erosion and deposition- at the antenna structures. Finally, a synthetic diagnostic using S/XB coefficients from collisional-radiative models is developed to compare simulation data with the spectroscopic measurements from experiments. In present case, this entire workflow is applied to an interpretive modeling of ohmic, lower-hybrid and ion cyclotron (IC) RF heated discharges in the WEST tokamak.  The neutral tungsten (W-I) spectroscopic measurements at the WEST RF antenna are compared with results obtained from the modeling. The amount of W-I erosion estimated from the modeling at the WEST RF antenna limiters increases linearly with increase in the power crossing the separatrix (P_sep). The higher charged states of oxygen- O⁶⁺ and beyond- are found to be the dominant species for W-sputtering at the antenna limiters. The results from this modeling matches reasonably well with spectroscopic measurements at the WEST antenna limiters. Currently this framework is also being applied to understand PMI in various other RF-heated linear and toroidal devices.   

Adaptive Simulation of Cold Plasma in RF Heaters

Accurate RF simulations of fusion energy systems require the definition of high-fidelity analysis geometries. The steps in the RF simulation workflow include defeaturing of antenna CAD models; combining antenna, reactor wall and physics components into an analysis model geometry; applying physics attributes to the model; automatically generating a graded mesh; and executing an adaptive finite element analysis that includes the application of a-posteriori error estimation, and mesh adaptation. The simulation workflow employs modeling and meshing tools from Simmetrix, the high-order PetraM+MFEM analysis platform from PPPL+LLNL, and curved mesh adaptation procedures from RPI. The poster will overview the simulation workflow components. This will include discussion of technical developments required to support the coarse curved mesh generation and curved mesh adaptation needed to support the effective use of high-order adaptive RF simulations. The advantages of using high-order finite element methods for resolving wave fields for geometrically complicated strap type antenna in tokamak fusion systems will be discussed. This will include a study of the right combination of mesh adaptation and element-order that is most effective to obtain results to the desired level of accuracy.   

ICRF Antenna – Plasma Edge Interaction in Alcator C-Mod: Comparison Between Experimental Results and 3D Full Wave Simulations

We show that an optimal ratio of the power coupled by the center and outer straps of the Alcator C-Mod four-strap field-aligned antenna can facilitate H-Mode access and minimize unwanted near-field and far-field RF-enhanced heat fluxes, potentials, and impurities. The interaction between ICRF-induced ExB flows and turbulent filaments in the scrape-off layer (SOL) is studied, and we show that the sheared flows can slow down and destroy filaments, eventually suppressing convective transport in the far SOL. This interaction is eliminated by reducing RF-induced ExB flows via power ratio modulation. A 3D finite element cold plasma model of the C-Mod ICRF system is implemented in Petra-M using realistic antenna and vacuum vessel geometries. The simulations are compared with experimental results involving power ratio modulation. 3D Fourier analysis is used to study the excitation and propagation of different toroidal modes into the plasma. ICRF coupling simulations for field-aligned and toroidally aligned antennas during ELMs and confinement transitions are performed and compared with experimental trends. Preliminary far-field sheath simulations with realistic 3D limiter geometry will also be presented.   

Two-dimensional kinetic full wave analysis of ICRF waves using integral form of dielectric tensor

In order to describe the two-dimensional wave structure and power deposition profile of ion cyclotron waves in magnetized plasmas, full wave analysis using the integral form of dielectric tensor has been extended to two dimensional configurations. The integral form of the dielectric tensor includes three types of kernel function, such as Plasma Dispersion Kernel Function (PDKF), Plasma Gyro Kinetic Function (PGKF), and Parabolic Plasma Dispersion Kernel Function (PPDKF). Inhomogeneity of magnetic field strength along the field line as well as the finite gyro radius effect are consistently take into account. Finite element method taking account of the coupling of elements along particle orbits is utilized. In a tokamak configuration, cases of minority-ion heating and second-order harmonic heating comparison are compared with those of conventional differential operator approaches. Coupling with the short-wave-length ion Bernstein waves and the interaction with energetic ions are discussed. Application to axisymmetric mirror configurations will be also presented.   

Physics-based modeling of RF-assisted start-up in tokamaks: effect of magnetic field on impact-ionization

A tokamak relevant physics-based model has been developed for the energization of electrons undergoing elastic and impact ionization collisions with atoms and molecules of a gas. In the presence of a background electrostatic potential, the energization is due to electron cyclotron (EC) heating by a spatially localized Gaussian beam. In contrast to Monte-Carlo based techniques in which impact ionization is predominantly based on the motion of an electron parallel to the magnetic field, we follow the complete orbit in a magnetic field. Electrons that are predominantly energized by EC transverse to the magnetic field while being slow in their toroidal motion, have the same probability for impact ionization as the toroidally fast ones. The dominant energization appears in the gyro motion and does not affect the cross-section for impact ionization. From our simulations we can estimate the time needed to ionize a neutral gas for plasma startup in a tokamak. The ionization time depends on power in the EC beam, and on the loop voltage and neutral gas pressure. We will present results from our simulations, in particular, for ITER relevant parameters. We will also discuss an analytical formalism that gives a measure of the startup time and agrees with the simulations.   

Comparing Current Drive Efficiencies from Empirical Models and Higher‐Fidelity Ray‐Tracing Codes for Various Heating Sources

The performance of confined plasmas in tokamaks is controlled by several factors including the total

current profile which is comprised of several current drives that result from plasma heating

mechanisms. These heating mechanisms include electron‐cyclotron (EC), ion‐cyclotron (IC), lower‐hybrid

(LH), helicon (HC), and neutral‐beam injection (NBI) heating [1]. Ray‐tracing codes, such as TORAY,

GENRAY, NFREYA, NUBEAM, and TORIC, have been used to accurately calculate the current drive

efficiency and deposition location for each heating mechanism. Considering the large computation

resources required for simulating many heating and current drive schemes using ray‐tracing codes while

searching for an optimum plasma scenario of high performance, the need for reduced, surrogate, or

empirical models for the current drive efficiency is necessary during the design phase [2]. In this work

we compare the current efficiencies using empirical models proposed by Tonon [3] and those calculated

in ray‐tracing codes for radio‐frequency heating mechanisms such as ECH, ICH, LHH, and HCH. The

difference between the empirical models and ray‐tracing codes varies with the heating mechanism. A

further study searches for possible adjustments needed for the empirical models to provide comparable

results to more accurate higher‐fidelity the ray‐tracing codes.   

Improved local helicity injection efficiency by the equilibrium field predicted from 0-D power balance model

Local Helicity Injection(LHI) technique has been demonstrated to overcome the difficulties at plasma startup and current-drive in spherical torus [1]. Versatile Experiment Spherical Torus (VEST) has studied LHI technique using arc plasma guns[2]. Recently, it is suggested that the performance of LHI in VEST can be increased by adjusting the equilibrium field determining the coupling between flux ropes and main plasma in closed flux surface, the efficiency of helicity injection. As the flux ropes discharged by the plasma guns inject helicity into the main plasma, the helicity gradient between the flux ropes and the main plasma determines the flow of the helicity. The helicity gradient strongly depends on the equilibrium field structure and strength [3]. As the new controllable power system for equilibrium field coils and new helicity injector are applied, the available field structure regime is extended to make field structures to maintain high helicity injection flow. Considering the relation between helicity flux and equilibrium field, a 0-D power balance model is used to find efficient helicity injection equilibrium field conditions and analyze LHI experiments in VEST. The 0-D model describes experiment results well and suggests the favorable equilibrium field structure condition for improved LHI efficiency in VEST.

References

[1] Bongard, Michael W., et al. "Advancing local helicity injection for non-solenoidal tokamak startup." Nuclear Fusion 59.7 (2019): 076003.

[2] JongYoon Park, An, Y., Jung, B., Lee, J., Lee, H., Chung, K. J., ... & Y.S. Hwang. Fusion Engineering and Design, 96, 269-273 (2015).

[3] JongYoon Park, Kim, T., Eom, T., & Y.S. Hwang, 64th Annual Meeting of the APS Division of Plasma Physics, Oct. 19. Spokane, Washington, the United States, (2022).   

Helicon Wave Experiments on the LArge Plasma Device (LAPD)

Helicon waves, also known as fast waves in the lower hybrid range of frequencies or whistler waves, are a proposed means for non-inductive current drive in reactor-grade fusion devices. Work at DIII-D to evaluate the prospect of helicon current drive has been done in the low power (<0.5 kW) regime [R.I. Pinsker et al 2018 Nucl. Fusion 58 106007], and high power (~0.5-1 MW) experiments are currently ongoing [B. Van Compernolle et al 2021 Nucl. Fusion 61 116034]. Through the use of two antenna structures, a low-power traveling wave antenna and a four strap antenna, the LAPD can be configured to be representative of the DIII-D scrape-off-layer that these waves must couple and propagate through. Experiments have been conducted to characterize the wave coupling and propagation from the traveling wave antenna over a large range of plasma parameters. Progress of current experiments that are investigating the role that edge turbulence plays in wave propagation in the helicon regime will be reported. Here we will look specifically at how density filaments can cause wave scattering [A.K. Ram 2016 Phys. Plasmas 23, 022504] and ‘stimulated mode-conversion’ [P.L Andrews 1985 Phys. Rev. Lett. 54] leading to wave power in undesirable locations.   

Modelling Microwave Start-up in Spherical Tokamaks

We present the development of a time dependent, self-consistent model for microwave start-up in spherical tokamaks. Microwave start-up is a solenoid-free method of driving the plasma current during the early stages of tokamak operation, from ionisation to the formation of closed flux surfaces. It is a critical issue for tokamak power plants with limited space for a neutron-shielded solenoid. Experiments on MAST achieved 73 kA of current with 100 kW of input microwave power via OXB mode conversion and a mirror polarizer on the centre column [1]. However, there has been limited modelling on this topic so far as most ray-tracing and Fokker-Planck codes assume closed flux surfaces. Including the open field line configuration is important as modelling has shown that the asymmetric confinement of electrons plays a significant role in the current drive [2]. Additionally, the current drive mechanisms and microwave-plasma interactions change over time due to the increasing density and changing poloidal magnetic field, introducing the need for time-dependent simulations.

 

[1] Shevchenko V.F. et al 2015 EPJ Web Conf. 87 02007

[2] Maekawa T. et al 2012 Nucl. Fusion 52 083008   

Plasma startup by electron cyclotron waves in magnetic mirrors

The creation of a plasma from neutral hydrogen gas in a fusion device is primarily through electron impact ionization. Of the various techniques available for the startup phase, the utility of electron cyclotron waves in magnetized plasma has garnered substantial interest. In transitioning from a few electrons, generated by events such as cosmic rays, to about 10% ionization of the gas in a device, we consider two necessary aspects related to the underlying physics. The first is the mean free path length an electron has to traverse prior to an impact ionization event. This physics is well understood, both experimentally and theoretically, and sets the minimum time required for the startup phase. The second aspect is the time needed to energize an electron to energies meaningful for impact ionization. If this is greater than the time needed to negotiate a mean free path length, then the startup time is determined by the energization process. We examine these two aspects within the context of a magnetic mirror, in which the electrons interact with a model Gaussian beam in the electron cyclotron range of frequencies. The decorrelation induced by the mirror magnetic field is effective in heating the electrons and the startup time is determined by the physics of the mean free path.   

RF absorption in presence of fast particles using AORSA coupled with MCGO

Modeling fast magnetosonic wave propagation and absorption in the presence of non-Maxwellian ions at high harmonic mode numbers (>2 fci) is challenging due to absorption being highly sensitive to the non-Maxiwellian distribution and ion finite orbit width. The full wave code AORSA [1] coupled to a Monte-Carlo orbit code MCGO [2]  is being studied as a candidate to address this modeling need. Two examples of cases where non Maxwellian effects are important to incorporate are for a proposed harmonic fast wave (HFW) traveling wave antenna for fast particle generation on DIII-D, and in high harmonic fast wave heating in the WHAM [3]  mirror device. For the HFW DIII-D case, a fast wave tailored to 4th or above harmonics is launched from the high field side and damps on the moderately fast (tens of keV’s) neutral beam deuterium ions, which are accelerated to hundreds of keV’s. These fast particles mimic reactor parameters and can go on to excite Alfven Eigenmode (AE) instabilities, allowing for a D-D tokamak like DIII-D to replicate reactor-relevant fast particle conditions experimentally. The fast ion tail pulled from the neutral beam distribution is highly non-Maxwellian. Therefore, to confidently model a HFW fast particle source, the AORSA/MCGO framework is required. Here we present an initial scoping study with the raytracing/Fokker-Planck codes GENRAY/CQL3D [2] (zero banana width) to identify DIII-D discharges to analyze for the AORSA/MCGO framework. In the second case, full-wave modeling of fast wave heating in the WHAM mirror device is desired. Mirror plasmas exhibit non-Maxwellian ion distributions due to the loss cone and a parallel electric potential. WHAM uses neutral beams, onto which harmonic fast waves damp and accelerate beam ions.  Due to the high mirror ratios in the WHAM device, many cyclotron harmonics are present at once, with most damping occurring on the 2nd harmonic. These many harmonics add further importance to a full-wave treatment. While AORSA/MCGO is well suited for tokamaks, both codes require modification for the mirror geometry. The latest modeling and analysis for AORSA/MCGO’s framework will be presented.

Citations 

1 Jaeger E.F. et al. 2001 Phys. Plasmas 8 1573–83

2 Harvey, B. https://www.compxco.com/index.html  

3 J. Egedal et al 2022 Nucl. Fusion 62 126053   

Investigation of Density Fluctuation Influence on ECH Microwave Propagation in LHD

Electron cyclotron heating (ECH) is an essential electron heating and current drive method for magnetically confined fusion plasmas and this method uses a high-power focused microwave beam, which is absorbed in plasma via electron cyclotron resonance. Recent studies suggested the ECH microwave beam is scattered by turbulent density fluctuations in plasmas and the beam width evaluated by a Gaussian profile is broadened. The beam broadening caused by the density fluctuations will deteriorate the heating locality and the current drive efficiency by ECH in plasmas. Therefore, it is important to evaluate the influence of the density fluctuations on the ECH beam. The Large Helical Device (LHD) is equipped with 1 MW class ECH systems of 77 GHz and 154 GHz. A two-dimensional phase contrast imaging (PCI) system is operational in LHD and can measure the absolute value of electron density fluctuations caused by ion-scale turbulence. This PCI system provides the time evolution of the radial profile of the density fluctuations including the wavenumber spectrum. We carry out the electromagnetic simulation to analyze the effect of the density fluctuations on the 77 GHz microwave beam based on the ECH system in LHD. We will present the influence of the spatial structure of the density fluctuations on the ECH beam and the estimation of the 77 GHz ECH beam broadening caused by density fluctuations measured by the PCI in the LHD.   

Simulation of lower hybrid wave in fusion plasma using FullWave

FAR-TECH, Inc. has been developing the FullWave code,  a hybrid-iterative full wave solver, to model rf waves in hot plasmas including the nonlocal hot plasma responses. The code uses configuration space to discretize wave equations and calculate the nonlocal hot plasma dielectric response.

 

To extend the capability to lower frequencies such as Lower-Hybrid and Helicon waves, an implicit time-stepping algorithm has been implemented. The hybrid iterative algorithm significantly increases spatial resolution of the RF fields but requires time steps to be comparable to Courant condition in explicit methods to fully resolve electron inertia.

 

 We have explored ways to speed up the FullWave simulation by implementing direct solver and local implicit scheme. Direct solver solves the full wave equations in the frequency domain by inverting the conductivity kernal tensor by utilizing SuperLU-Dist library. We have tested direct solver on 50M computation points for cold plasma model on NERSC.  The locally implicit method solves the Ampere’s law implicitly on each computation point while the Faraday’s law is solved explicitly. The convergence and calculation speed are improved by several times using this scheme than the implicit time-stepping scheme. We will present recent 1D and 2D FullWave simulations in cold model using DIII-D parameters.

 

In addition, hot conductivity kernel calculations in the range of LH will be presented.   

Fast particle trajectories, quadratic flux-minimizing maps, and integrability in quasiaxisymmetric and quasihelical stellarators

Energetic particles can be damaging to device walls in stellarator reactors, posing a threat to device longevity. These particles can be lost from stellarator equilibria via mechanisms like diffusive and convective transport. Following fast particle trajectories and understanding the relationship between loss mechanisms and field geometry can further guide improvements in stellarator optimization. We investigate the motion of trapped and passing fast particles in quasiaxisymmetric and quasihelical stellarator equilibria through the application of guiding-center trajectory mappings for trapped and passing particles in phase-space. We apply quadratic flux-minimization methods to produce periodic pseudo-orbits that resolve the nearby integrable Hamiltonian system in the presence of island overlap and phase-space chaos. We demonstrate application of the maps for both trapped and passing particles in quasiaxisymmetric and quasihelical equilibria in vacuum and non-zero β regimes. Analytic results for trapped particle trajectories in the near-axis model are compared to numerical results, along with theoretically predicted and numerically determined characteristic frequencies.   

Single Particle Modeling of Electron Diffusion in Magnetized Plasmas with Magnetic Islands

Recent DIII-D experiments demonstrated that magnetic islands can trap 10 MeV energetic electrons (EEs). Synchrotron emission camera data suggests that during the trapping, the emission from the EE cloud becomes brighter, which can result from increased density or acceleration of the EEs inside the island. To investigate these processes, a random kick algorithm is implemented in the TRIP3D-GPU field line tracing code. In this model, the motion of a test electron is assumed to be driven by the magnetic field line topology corresponding to the DIII-D experiments including the effects of radial magnetic perturbations, error fields, and error field corrections from the various DIII-D coil sets. To model a collision between the electron and a background ion, a random number is sampled from a Poisson distribution. To study the role of island topology, tracer electrons are launched from different starting points near island O-points and X-points. Then, the distribution of end locations (originating from the same starting point) is examined to assess the relative probability of the electron residence location at the end of the simulation. Initial comparison of the results against the experimental observations confirms increased probability for electron trapping near island O-points.   

Orbits of energetic particles near rational flux surfaces in stellarators

We calculate the drift orbits of resonant energetic particles near rational flux surfaces in stellarators. In the vicinity of a sufficiently low-order rational surface, there is a thin boundary layer in which particle orbits exhibit several unusual features. Passing particles make large radial excursions (of order sqrt rho-star), trapped particles do not necessarily conserve the longitudinal adiabatic invariant, and islands exist [1,2] which grow with energy and which may be deleterious for energetic particle confinement. Using a new adiabatic invariant, we determine the island shape: it is sensitive to the shear and deviation from omnigeneity of the stellarator magnetic field. We also study semi-trapped particles that make many toroidal transits before bouncing. They have an enhanced probability of orbit-class transitions near the rational surface, which could facilitate losses [3].

[1]  R. White, A. Bierwage & S. Ethier, Phys. Plasmas 29(5) (2022).

[2]  R. White, Phys. Plasmas 29(9) (2022).

[3]  E. J. Paul, A. Bhattacharjee, M. Landreman, D. Alex, J. L. Velasco, & R. Nies, Nucl. Fusion 62(12), 126054 (2022).   

Alfvén Eigenmodes mediate benign fast ion transport in ITER steady state scenario without microturbulence

We address the fusion product and beam injected ions confinement in the ITER steady-state (SS) scenario in the presence of low frequency Alfvén eigenmodes (AEs). A self-consistent Resonance Broadened Quasi-linear code RBQ is applied [1,2]. AEs are destabilized by MeV neutral beam ions and alphas, but the fast ion transport remains modest if only classical particle slowing down is assumed. A stability study of low-frequency AEs includes NOVA/NOVA-C, RBQ and the guiding-center code NUBEAM packages. AE linear stability addresses the effects of ion's finite orbit width and Larmor radius, thermal ion and electron Landau damping, trapped electron collisional, and radiative dampings. We have found around 40 unstable or marginally unstable AEs spanning toroidal mode numbers from 1 to 39. The subsequent QL simulations are supported by NOVA-C modeling which prepares the wave-particle interaction matrices of unstable AEs. The RBQ code evaluates the constants of motion diffusion coefficients of both fast ion species. Then the NUBEAM package is applied to evolve the fast ion distribution functions in the constant of motion space. We found that the classical EP confinement does not lead to strong fast ion losses. The strong dependence of fast ion relaxation on microturbulence, however, does not allow us to reliably predict the fast ion confinement in ITER advanced scenarios without quantitative predictions of turbulence levels.   

Optimization of Monte-Carlo NBI NUBEAM code for GPU: success and challenges

NUBEAM is a comprehensive software package designed to model fast ion heating in tokamaks. Widely trusted and utilized in major tokamaks worldwide, NUBEAM serves as a reliable tool for plasma beam and heating modeling. This code efficiently calculates various parameters, including heating profiles, torque, current driven profiles, particle sources, and neutron emission resulting from neutral beam injection and fusion products. Its validation is robust, encompassing comprehensive physics that accounts for phenomena like Alfvénic and radio-frequency wave-particle interactions.

Driven by the need for rapid analysis and experimental planning in control rooms, this work draws inspiration from the present-day requirements of quick turnaround between shots. The Monte Carlo beam and fusion product code, NUBEAM, has undergone modifications to leverage the power of GPU architecture, thereby accelerating the simulation of fast ion population deposition, orbiting, and slowing down in tokamak plasmas. The newly developed code has been rigorously tested on two supercomputers, Traverse at PPPL and Perlmutter at NERSC, which utilize NVIDIA V100 SXM2 GPUs and NVIDIA A100 GPUs, respectively.

The results obtained demonstrate a notable speed improvement of approximately 4.5 times on Traverse, employing four GPUs and 32 MPI processes for the NUBEAM simulation with 100,000 Monte Carlo particles. Furthermore, Perlmutter showcases even better performance. Throughout this process, various challenges have been encountered and overcome, resulting in a compelling success story. These challenges and accomplishments will be highlighted and discussed in further detail.   

Investigating Energetic Particle Transport in LHD using Gyrokinetic Simulations

Understanding EP transport is crucial for optimizing stellarator performance and enhancing fusion efficiency. In this research, we utilize the gyrokinetic toroidal code (GTC) to perform neoclassical simulations of EP transport in the Large Helical Device (LHD) to eventually investigate the influence of the ambipolar electric field on EP dynamics. GTC is being benchmarked with GNET which has been extensively used in EP analysis in LHD. As a first step, GTC uses as input the slowing-down distribution function of fast ions obtained from GNET simulations. Utilizing the Fokker-Planck collision operator, GTC includes the effects of collisions between fast ions and thermal plasma. Results show that GTC can successfully preserve the fast ion phase space distribution from GNET.     

Gyrokinetic simulation of Alfvén Wave-Energetic Particle Interplay in Spherical Tokamaks MAST-U and ST40

The Gyrokinetic Toroidal Code (GTC) is utilized to investigate the interplay between Alfvén waves and Energetic Particles (EPs) in spherical tokamaks (STs). In MAST-U discharge #26887, Toroidal Alfvén Eigenmode (TAE) is simulated using antenna and EPs excitation, matching the ideal MHD simulation results and qualitatively agree with experimental measurements. In the ST40 discharge #09781, electromagnetic modes near the core region in the sub-TAE frequency range are observed. These modes, which are accompanied by the transitions between the low-confinement mode (L-mode) and the high-confinement mode (H-mode), are identified as beta-induced Alfvén–acoustic eigenmodes (BAAE) by MHD simulations using the NOVA code. The MHD simulation using GTC successfully reproduces this low-frequency mode with similar frequency and mode structure as predicted by the NOVA. However, in GTC gyrokinetic simulations, higher EP density is needed to drive such modes. The agreement of these low-frequency instabilities has also been observed in two additional shots, namely #09831 and #09894. Ongoing efforts focus on detailed verifications of frequency, growth rate, and mode structure, aiming for more realistic and comprehensive results. 

These simulation results represent the potential of GTC for studying the EPs effect and various phenomena in STs. Validating gyrokinetic simulations with experimental measurements represent the first step toward a better understanding and interpretation of AE phenomena in STs.   

Application of FIDA as a charge exchange loss measurement for NB-produced fast ions in small or medium-size devices

n NBI heating experiments in small or medium-size toroidal devices, the charge exchange (CX) loss of NB-produced fast ions is not negligible in the determination of the fast ions’ slowing down velocity distribution. Although it may be possible for many numerical simulation methods for the fast ions to include this loss mechanism, experimental measurements of the neutral particle density in the 3-dimmensional real space as the input to such calculations are almost impossible. On the other hand, the FIDA (Fast Ion D-alpha) is widely used in various devices for investigating the local fast ion velocity distribution. For the studies of beam-driven phenomena, and the anisotropic pressure MHD equilibrium, the requirement on the velocity distribution is not in the detailed understanding on the slowing down process including the CX loss but in the reduction factor of the lower Legendre order structures of the velocity distribution. In both the direct solving using the eigenfunction and the indirect solving based on the adjoint equation, it can be shown that for this purpose that the effect of the charge exchange loss on the velocity distribution will not appear in the pitch-angle space structure but will appear only in the energy space structure. When we find the substantial neutral particle density by comparison of the FIDASIM calculation including this energy space reduction factor and the experimentally observed H/D-alpha spectrum, we should include this reduction factor also in calculations of the beam-driven effects. The result in the Heliotron-J will be presented.   

Fast ion transport in quasisymmetric equilibria in the presence of a resonant Alfv'{e}nic perturbation

Significant progress has been made in designing magnetic fields that provide excellent confinement of the guiding enter trajectories of alpha particles using quasisymmetry (QS). Given the reduction in this transport channel, we assess the impact of resonant Alfv'{e}n eigenmodes (AEs) on the guiding center motion. The AE amplitudes are chosen to be consistent with experimental measurements and large-scale simulations. We evaluate the drift resonance condition, phase-space island width, and island overlap criterion for quasisymmetric configurations. Kinetic Poincar'{e} plots elucidate features of the transport, including stiff transport above a critical perturbation amplitude. Our analysis highlights key departures from the AE-driven transport in tokamaks, such as the avoidance of phase-space island overlap in quasihelical configurations and the enhanced transport due to wide phase-space islands in low magnetic shear configurations. In configurations that are closer to QS, with QS deviations $delta B/B_0 lesssim 10^{-3}$, the transport is primarily driven by the AE, while configurations that are further from QS, $delta B/B_0 sim 10^{-2}$, experience significant transport due to the QS-breaking fields in addition to the AE.   

Experimental Evidence of Ion Acceleration at Internal Reconnection Events in the MAST-U Spherical Tokamak

Spherical tokamaks often have  internal reconnection events (IRE) and ion acceleration was observed in ohmic plasmas previously[1].  These events correspond with a sudden relaxation of the safety factor at the IRE.  In this work, a solid state neutral particle analyzer (SSNPA) [2] diagnostic oriented to measure mainly trapped and co-passing beam ions in Mega Amp Spherical Tokamak Upgrade (MAST-U) shows significant signal rise following internal reconnection events (IRE).  Initial analysis suggests that fast-ions are accelerated in ~200 micro seconds to > 60 keV energies. Weight function analysis for the SSNPA also indicates that these higher energy fast ions are redistributed from orbits with very low pitch to orbits with slightly negative pitch.  Additional analysis from other fast ion diagnostics[3,4] is also presented, further narrowing the region of fast-ion phase space accelerated by the IREs.     

First principles kinetic simulations of observed ion cyclotron emission driven by energetic ion-beams injected into LAPD

Ion cyclotron emission (ICE) is widely observed from minority energetic ion populations in toroidal magnetically confined fusion plasmas, both tokamaks and stellarators, and from cylindrical plasmas in the LAPD (Large Plasma Device). Comparing ICE spectra from LAPD and toroidal plasmas offers a unique pathway to identifying the influence, if any, of toroidal effects on the observed signal and on the underlying physics. Here we report first principles kinetic simulation of an ICE spectrum from a spiraling ion beam in the LAPD plasma, using the fully gyro-orbit-resolved particle-in-cell code EPOCH in 1D3V slab geometry mode. A simulated proton beam is found to relax collectively under the magnetoacoustic cyclotron instability, in a deeply sub-Alfvenic regime which is predominantly electrostatic. The simulated ICE power spectrum, obtained from the spatiotemporal Fourier transform of the excited field, quantitively captures the key features of the observed spectrum. These include downshifting of sequential narrow spectral peaks with respect to lower integer cyclotron harmonics below the flattening of the fast Alfven wave (FAW) dispersion relation, a broad continuum feature around the lower hybrid frequency, and exact higher-integer cyclotron harmonic peaks.   

Verification of electromagnetic fully-kinetic-ion simulations of high-frequency waves in toroidal geometry

In recent fusion experiments, several high-frequency modes driven by energetic particles (EPs) have been observed on the order of the ion cyclotron frequency, such as compressional Alfvén eigenmodes (CAEs), global Alfvén eigenmodes (GAEs) and ion cyclotron emission (ICE). To study these waves, an electromagnetic simulation model, in which the ion dynamics is described by a six-dimensional Vlasov equation is formulated and implemented in the global gyrokinetic toroidal code (GTC). Linear simulations of generalized ion Bernstein waves (propagation perpendicular to equilibrium magnetic field) and shear Alfvén waves (propagation parallel to equilibrium magnetic field) are verified by comparing simulation results with analytic dispersion relation in cylindrical geometry. Simulation of the ICE excitation by EPs has been verified by comparing with magnetoacoustic cyclotron instability theory. Using a set of typical plasma parameters from the DIII-D tokamak experiment, the electromagnetic fully-kinetic-ion simulations find the ICE frequency spectra qualitatively consistent with the experimental results. Kinetic simulation of the ICE global structure is needed for the development of the ICE as a diagnostic tool for alpha particles in future burning plasma experiments.   

[PLACEHOLDER] ITER, HBT-EP, AND TOKAMAK CONTROL

.   

Divertor-safe Nonlinear Burn Control Using a Reference Governor Based on a Core-edge Model with ITER-SOLPS Scalings

For tokamaks such as ITER, the highly nonlinear and coupled dynamics of burning plasmas necessitates the design and application of sophisticated burn-control strategies that regulate the plasma’s density and temperature. Furthermore, the coupling between the core- and edge-plasma regions will render integrated burn and divertor control in ITER even more challenging. For example, conditions in the scrape-off-layer (SOL) and divertor regions determine the strength of the fuel recycling from plasma-wall interactions and the severity of the impurity pollution into the core plasma. Important burn- and divertor-control objectives include, respectively, producing a desirable amount of fusion power and avoiding catastrophic melting of the divertor targets. Because increasing the fusion power amplifies the power flowing through the SOL and onto the divertor targets, burn- and divertor-control objectives may conflict at times. The presented burn- and divertor-control strategy addresses this issue by augmenting a nonlinear burn controller with a reference governor that is designed to protect the divertor targets. This control design is based on a core-SOL-divertor model that couples the nonlinear density and energy response models of the core plasma with edge-plasma scalings that were generated from SOLPS4.3 simulation results [1].

[1] H.D. Pacher, et. al., J. Nucl. Mater. 463 (2015) 591–595.   

Tungsten neoclassical transport in ITER pedestal for non-axisymmetric perturbations

Tungsten transport in the pedestal region of ITER poses a serious threat to the device fusion performance due to radiation losses. The consequences of breaking the magnetic field symmetry in tungsten transport in the pedestal region of ITER are examined. The scenario considered is an inductive operation ITER H-mode plasma with Q ≈ 10, I_p = 15 MA and B_t = 5.3 T obtained from ITER integrated plasma simulations. Neoclassical transport calculations are performed with the SFINCS code for 3D MHD, free boundary equilibria, calculated with the VMEC code, considering: i) an axisymmetric case ignoring the discrete nature of the TF coils, ii) a set of equilibria with toroidal ripples ranging from 0.25% to 2% and iii) an equilibrium with the ELM control coil angle offsets maximizing the Resonant Magnetic field Perturbation (RMP) close to the LCFS.

The results indicate a negligible ripple effect on the radial electric field with respect to the theoretical axisymmetric case and a reduced E_r region for RMP fields. However, for the region with large electric fields, it is not possible to extract robust conclusions on tungsten flow direction due to, first, the possible radially non-local character of the dynamics and, second, the high poloidal Mach number for tungsten that puts into question the accuracy neoclassical theory ordering. Further advancements in modeling techniques are needed to address this issue to provide more accurate predictions, which, needless to say are of paramount importance for ITER.   

Fueling modeling and control for ITER first plasma operation

The initial phases of operation on ITER aim to generate plasmas up to ~10 MA for several seconds. The ability to control the fueling gas via the Gas Injection System (GIS) and eventually the Pellet Injection System (PIS) is essential for the prefill & density control during the entire plasma discharge. The GIS control, as part of the plasma control system, is designed to handle the ITER specifications and plasma scenario requirements.

The GIS consists of 10 Gas Valve Boxes (GVB) each containing six Mass Flow Control (MFC) valves. Each GVB is connected to the Vacuum Vessel (VV) by a separate pipe (~17 m to ~32 m long). The pipes are the main source of delay of the fueling response, which affects the system controllability.

To evaluate the GIS design with respect to the fueling control requirements, a model of the system based on a 1-D description of the gas flow diffusion in each pipe has been developed. The minimum gas flow of the MFCs is too large to reach the prefill pressure target. A PWM solution for the control is proposed.

The model also computes the gas balance in the VV, and thus, computes the required prefill gas to reach the breakdown pressure and initial plasma density. The control effort to fuel the plasma at low density can be predicted by coupling the 1-D model with the particle balance equation that describes the plasma-neutrals interactions.

A GIS failure event handling strategy based on MFC equivalence is also proposed. The simulation results and future work to optimize the ITER prefill & fueling phases are reported here.   

Progress in extending the DIII-D high-βP scenario to long pulse on EAST

Recent experiments on EAST have made progress in extending performance and duration of the high-β_P scenario, with world record duration of 403 s at β_P~2.5, β_N~1.5. Unlike the high β_P scenario on DIII-D with large radius internal transport barriers (ITBs) in all channels, on EAST the ITB is only visible in the electron temperature profile, at small radius. Transport analysis suggests that, at large radius, these plasmas are limited by ITG turbulence, despite low ion temperature gradients. Significant turbulence reduction is predicted with higher q_min and reduced magnetic shear. Experiments using EC waves for early heating, produced higher q_min plasmas with lower internal inductance, but lacked the power to sustain those conditions via sufficient bootstrap current. New experiments have been planned using TRANSP time-dependent modeling to optimize early EC heating timing, power, and deposition profile. It is predicted that strong off-axis EC heating will also cause the LHCD profile to peak off-axis. This could help sustain the high q_min profile, together with additional heating power.   

Autoencoding and noise reduction of plasma diagnostics in Tokamak

Accurate measurement of plasma parameters is crucial for understanding and optimizing plasma behavior in Tokamaks. However, noise and disturbances often degrade the reliability of diagnostic data. This poster provides an overview of applying autoencoding and noise reduction techniques to enhance plasma diagnostics in Tokamaks. Autoencoders, a type of artificial neural network, effectively reduce noise and extract meaningful features from complex data. By training an autoencoder on a large dataset of measured plasma diagnostics, it learns to denoise signals and preserve essential plasma information. Specialized variants like variational autoencoders and denoising autoencoders address specific challenges in Tokamak plasma diagnostics. They reconstruct high-quality plasma parameters from noisy or incomplete measurements, enhancing accuracy and reliability. Integrating autoencoding techniques with noise reduction algorithms offers a comprehensive approach to address plasma diagnostics challenges. By leveraging machine learning and signal processing, these methods improve the quality of plasma parameter measurements, enabling precise analysis and modeling of plasma behavior. This research highlights the potential of autoencoding and noise reduction techniques to revolutionize plasma diagnostics in Tokamaks, advancing fusion research and sustainable energy sources. The poster serves as a foundation for further exploration and development of innovative methods to enhance plasma diagnostics in fusion devices.   

Vertical instability growth rate studies with rigid and deformable plasma models and proximity controller development in the TCV tokamak

The control of plasma position and shape is of fundamental importance to achieve high performances in fusion tokamaks. In particular, the control of the vertical position is essential for elongated plasma to avoid vertical displacement events (VDE), which cause the disruption of the plasma column against the tokamak facing components. A proximity controller based on the real-time growth rate estimation has been implemented on the TCV tokamak. The controller tracks the plasma growth-rate to avoid overcoming stability limits. The growth rate estimation is based on the RZIp model (in a similar way as in [Olofsson 2022 PPCF 64 072001]), which proves to be fast enough for an online implementation. The calculations of the RZIp model has been compared to a more accurate one: the deformable free boundary FGE (and its linearised version FGElin) [Carpanese et al. 2020 EPFL PhD thesis n. 7914]. The two models are implemented in the MEQ (Matlab EQuilibrium) suite, a flexible matlab toolbox for tokamak equilibrium calculations. A validation of the FGElin growth rate calculations against the KINX code [Degtyarev et al. 1997 CPC 103 10], a linear ideal MHD growth rates and eigenvectors solver for axysimmetric plasmas with resistive wall and plasma mass included, has been carried out. The codes have been tested for different plasma shapes to inspect the sensitivity of the growth rate calculation on the model adopted and on the details of the equilibrium reconstruction.   

Incorporating Gas Puffing Delays into Density Control Synthesis in Tokamak Reactors By Combining Optimal Control and Reinforcement Learning Techniques

Gas puffing is one of the main actuation systems available for regulating particle density in tokamaks. In next-generation tokamaks like ITER, the distance between the gas valves and the vacuum vessel is expected to delay the response of gas-puffing actuation systems. If not appropriately accounted for during control synthesis, such delayed response from the actuators can lead to performance degradation or destabilization of the density control system. This work presents a systematic approach to designing a model-based density controller that inherently accounts for input delays in the gas puffing systems. The proposed control synthesis relies on formulating an infinite-horizon optimal control problem and solving it using reinforcement learning algorithms with an actor-critic architecture. The resulting solution is then reformulated to account for input delays. In addition to control synthesis, systematic methods to handle changes in plasma operating conditions, which can lead to substantial uncertainties in the density model, are also proposed in this work. Finally, the effectiveness of the proposed density controller is illustrated using nonlinear numerical simulations.   

HBT-EP Program: MHD Dynamics and Active Control through 3D Fields and Currents

The HBT-EP research program aims to: (i) understand the physics of scrape-off layer currents (SOLC) and interactions between the helical plasma edge and conducting boundary structures, (ii) test new methods for measurement and mode control that integrate optical and magnetic detector arrays with both magnetic and SOLC feedback, and (iii) understand MHD issues from disruptions, resonant magnetic perturbations, SOLC, and disruption mitigation using an in-vessel passive coil. A two-color multi-energy EUV/SXR tangential array has been used to observe the suppression of sawtooth MHD activity correlated with the coupling of an internal 2/1 mode and an external 3/1 kink mode. This is consistent with observations of flux pumping which broadens the current profile as has been seen in hybrid plasmas in DIII-D and JET. A deep-learning-based MHD mode tracking algorithm to process video frames from the upgraded HBT-EP high-speed videography system has been used to determine the n=1 mode amplitude and phase [1]. A real-time application of this algorithm using an FPGA has achieved a latency less than 17 μs, on par with the magnetic sensor GPU-based control system. Measurements from a new set of high-field side halo current diagnostic and control tiles before and during disruptions will be presented, along with plans for runaway electron mitigation coil (REMC) studies.

[1] Y. Wei, et al, PPCF 65 (2023) 074002   

FPGA-based microsecond-latency MHD mode tracking using high-speed cameras and deep learning on HBT-EP

A deep-learning-based MHD mode tracking algorithm using high-speed imaging cameras has been developed for real-time feedback control application on the High Beta Tokamak – Extended Pulse (HBT-EP) device. Our algorithm [1] uses the convolutional neural network (CNN) to process video frames taken during plasma shots by one or multiple cameras and predict the n=1 mode amplitude and phase over time. The model is able to accurately track the n=1 modes consistently over the testing shots and demonstrated significant improvement over the previous SVD-based method [2].

For real-time application we utilize the hls4ml (High Level Synthesis for Machine Learning) [3] framework to optimize the deep learning models for deployment onto Xilinx FPGA devices. Using hls4ml, our model is implemented directly on the Euresys Coaxlink Octo framegrabber board in the existing camera diagnostics system and achieves an input-to-output latency below 17 μs, on par with the current GPU-based control system using the magnetic sensors. The proposed controller will later be integrated into the feedback control system on HBT-EP to perform real-time mode control.

[1] Wei, Y. et al (2023) PPCF 65 074002

[2] Angelini, S. et al (2015) PPCF 57 045008

[3] Fahim, F. et al (2021), arXiv:2103.05579   

Observation of sawtooth suppression by flux pumping on HBT-EP

The tangential EUV/SXR diagnostic system has been installed on HBT-EP tokamak for measurements of electron temperature and study of mode dynamics. Analysis of the temperature profile from the system's signals reveals clear sawtooth events. It is observed that sawtooth existence correlates with low-amplitude edge modes, while suppression links to larger and saturated edge mode amplitudes. To further investigate this correlation, the plasma-wall interaction was manipulated by adjusting the positions of conducting walls in HBT-EP, influencing the stabilization of the edge modes. It was found that sawtooth events only occur when the walls and the plasma are in close proximity, facilitating effective edge mode stabilization. Even slight differences in major radius result in divergent shot styles: sawtoothing shots or sawtooth-free shots. Through various mode structure analyses, the coupling between the m/n=3/1 resistive wall mode (RWM) and the m/n=2/1 tearing mode was identified. This coupling induces anomalous current broadening and indicates the occurrence of flux pumping. Based on these findings, we can conclude that sawtooth suppression in the HBT-EP tokamak is achieved through the process of flux pumping. Analysis of internal MHD structures that induce the dynamo effect will be explored using various diagnostic systems on HBT-EP. This research offers valuable insights into the underlying physics of sawtooth dynamics and contributes to the comprehension and control of instabilities in tokamak plasmas.   

Preparations and Plans for Runaway Electron Mitigation Coil (REMC) Studies on HBT-EP

A Runaway Electron Mitigation Coil (REMC) is an optimally shaped 3D conducting element inside of the vacuum vessel that is driven by the disruption voltage to provide a "fail-safe" prevention of high-energy runaway electrons [1]. This poster presents preparations and plans to install a REMC in HBT-EP. Scoping studies show a wide range of coil positions yield significant coupling of the pre-disruption plasma current to the passive coil, exceeding the coupling expected in DIII-D [2] and SPARC [3]. The flexibility of the HBT-EP device simplifies installation and will enable extensive model validation, including coupling changes from modifying the plasma major radius. REMC studies will explore interactions between the passive coil system and plasma operation. Disruption forces, measured with piezoelectric strain gauges and load cells, and the role of halo currents will be documented. Preparations for REMC studies also include scoping experiments conducted on both regular and "slide-away" discharges. Large non-axisymmetric fields from existing control coils will also be configured to resemble the REMC fields and triggered to study mode locking, disruption effects, and plasma motion monitored by fast videography, hard X-ray detectors, EUV sensors, halo current detectors, and Mirnov coils.

[1] Allen H Boozer 2011 Plasma Phys. Control. Fusion 53 084002.

[2] D.B. Weisberg et al 2021 Nucl. Fusion 61 106033.

[3] R.A. Tinguely et al 2021 Nucl. Fusion 61 124003.   

Observing Coupled Mode Dynamics Across Independent Diagnostics on HBT-EP

The core and edge structures of kink, tearing, and sawtooth crash instabilities have been studied across independent physical signatures at the HBT-EP tokamak. The Singular Value Decomposition is used to temporally correlate normalized spatial structures from independent diagnostics in selected time windows. This method is sufficiently robust to choices of signal filtering and composition, and can assist in analysis of nonlinear or coupled behaviors such as pre/postcursors to sawtooth crashes and rapidly evolving modes. Particular use is made of HBT-EP's spatially calibrated 4-array poloidal tomographic and two-color tangential EUV systems. These diagnostics permit us to observe rotating modes in many conditions, including 2/1 tearing modes coupled to edge 3/1 kink modes. New observations of sawteeth (see Boting Li this session) are supplemented by describing the MHD structures occurring before and after sawtooth crashes, and still unexplained observations include "bursting" coupled modes and fast MHD phenomena with very rapid amplitudes changes.   

Development of non-axisymmetric resistive wall models for MHD simulations of HBT-EP and other tokamaks

The NIMROD [1] code is used to validate multiphysics models (MHD + resistive wall) for the prediction of mode structures and scrape-off-layer (SOL) currents in tokamaks using high-resolution current, magnetic, and optical diagnostics of HBT-EP [2]. NIMROD’s existing thin resistive wall model is extended to include non-axisymmetric wall resistivity, capturing effects of ports and other wall structures. Simulations with a resistive wall observe non-disruptive, saturated, mode activity, consistent with experimental data on HBT-EP. Effects of varying wall resistivity with toroidal mode number are investigated in the context of a saturated resistive wall mode complete with magnetic islands and disruptions initiated with an artificial thermal quench. Work on improving boundary conditions in the resistive wall model to capture large-scale n=0 equilibrium evolution during disruptions will be discussed. Applications toward better understanding the 3D structure of wall-connected currents and the effects of runaway electron mitigation coil (REMC) fields will be presented. Plans to validate numerical models of wall-connected currents and quantify phase differences between diagnostic signals on HBT-EP, including current-sensing tiles, will be presented with the goal of improving SOL and wall models for ITER and next-step devices.

[1] C. Sovinec et al., J. Comput. Phys. (2004)

[2] J. Levesque et al., Nucl. Fusion (2017)   

The role of blob-filament turbulence on biased H-mode transition thresholds in HBT-EP

Blob-filament structures are a type of turbulence known to dominate the scrape-off layer (SOL) region of plasmas by radially ejecting turbulent structures with well-defined probability-distribution functions [1]. These structures have been postulated to affect L- to H-mode transitions through enhancing plasma-wall interactions and acting as a momentum, energy, and particle sink [2]. It was recently shown in the HBT-EP tokamak that the probability distribution functions of ion saturation and SOL currents closely resemble one another and indicate the presence of blob-filament turbulence [2]. This poster further characterizes blob-filament turbulence via a new set of biasable high-field side SOL current detectors to compare turbulent structures in regions of good and bad curvature. This greater understanding of blob-filament turbulence is applied to analyzing the critical voltages and currents necessary to achieve a biased H-mode on HBT-EP, which is controlled by using a high-power amplifier to vary the voltage on a bias probe inserted into the plasma. Measurements of the timescales and biasing requirements of L- to H-mode transitions with respect to the production of blob-filaments are presented through controlling the ramping rate of the bias voltage.

[1] G. Grenfell et al 2018 Nuclear Fusion 59, 016018.

[2] I. Stewart et al 2021 Phys. Plasmas 28, 052506.   

Scrape-Off-Layer Currents in HBT-EP Investigated Using High-Field-Side Tiles and Magnetic Sensors

A new set of high-field-side scrape-off-layer (HFS SOL) tiles was recently installed on the High Beta Tokamak-Extended Pulse (HBT-EP) device, providing data on SOL currents (SOLC). The tiles provide particularly important data on disruptions (See Levesque, this session). This study focuses on currents seen by the tiles during the main discharge period of plasmas, focusing on the correlation and phase relationships between current and magnetic perturbations seen through the tile arrays and magnetic sensor arrays. An extension of work previously done using low-field-side (LFS) tiles, this work characterizes SOLC using the HFS array in two toroidal locations separated by ~90°, comparing these results with those found using LFS tiles in six toroidal locations. We then use the observed properties of SOLC in HBT-EP to distinguish between proposed theoretical mechanisms, such as those presented in [1], for asymmetric SOLC.

[1] J.P. Levesque et al 2017 Nucl. Fusion 57 086035.   

Scrape-off-layer current measurements during disruptions in HBT-EP

Non-axisymmetric halo currents that occur during tokamak disruptions can result in significant asymmetric forces on the vessel, potentially causing damage to components. Understanding the generation and rotation of these currents is important for establishing design requirements for high-current devices to tolerate disruptions. A new set of halo current diagnostic tiles has been installed on the high-field-side of the HBT-EP tokamak, in regions where there is strong plasma-wall contact and current flow during disruptions. Two tile arrays are built as electrically-isolated blade limiters with poloidal resolution and separate toroidally-facing collection areas. Arrays are located at 2 toroidal locations separated by Δφ=96°, integrating over 6 poloidal regions with Δθ~12° resolution to measure narrow poloidal contact areas. We present the first disruption current measurements from these tiles. Strong localized electron currents to tiles correlate with magnetic perturbations, and are observed to rotate poloidally from tile to tile. Cathode spots form on tiles in some cases, leading to strong local electron emission. Plans for biasing the tiles and for expanding the system to a set of four equally-spaced arrays are also described.   

Zeeman Polarization Spectroscopy on Gas Puff Z-Pinches

Zeeman Polarization Spectroscopy, or ZPS, on 1 MA gas-puff z-pinches in CO2 doped Argon is being used to determine the magnetic field distribution in the plasma during implosion. Light is collected parallel to the azimuthal magnetic field tangential to the gas puff implosion sheath. The light is split into left and right hand circularly polarized components and then focused into two linear fiber bundles and delivered to a 750 mm spectrometer. By assuming cylindrical symmetry and from the sharp thermal gradient localized Zeeman measurements are made. The peak wavelengths of the two polarizations can be resolved in our 1 MA experiments, and the magnetic field determined from their separation, despite Stark Broadening. Introducing dopants into the gas puff allows the use of additional emission lines, such as a Carbon IV or Oxygen VI doublet, to increase spatial resolution. This method was developed for z-pinch experiments on a 500 kA, 500 ns rise time generator as part of the Cornell/NNSA pulsed power center.     

Radiometry and Phase Contrast Imaging Diagnostics on COBRA’s Gas Puff Z-Pinch Plasmas*

Understanding the energy partitions in a gas puff z-pinch plasma allows for further exploration of the behavior at stagnation. Energy can be partitioned into the magnetic field, turbulent flow, radiation, and kinetic energy. A radiometer and Phase Contrast Imaging (PCI) system are implemented on COrnell Beam Research Accelerator (COBRA), a ~ 1 MA driver at Cornell University, to determine the heat radiation, kinetic energy or time resolved radiated power, and turbulent flow of the plasma. Millimeter, Infrared (IR), and Ultraviolet (UV) radiometers are implemented to determine temperature measurements and time resolved radiated power. The millimeter wave radiometer has frequencies of 10 and 94 GHz, and the IR and UV radiometers are operated at 214, 1100, 1310, and 1550 nm. The PCI system uses a Nd:YLF laser at 1053 nm to determine the turbulence of the plasma as it transverses across COBRA’s vacuum chamber. Highlighted are the designs for the radiometer and initial measurements from Ne and Ar gas puff z-pinches, as well as designs for the PCI system and future goals for implementing it on COBRA.   

Electron probing of electromagnetic fields generated by intense laser pulses interacting with a liquid jet target

Recent results have demonstrated the feasibility of using electron beams from a Laser Wakefield Accelerator (LWFA) to measure electric and magnetic fields on ultrafast timescales. In this work, we use the ultrafast LWFA electron beam from the BELLA HTW system to probe the interaction of a longer-pulse laser (t ~ 200 ps, E ~ 1 J, I ~ 10¹⁴ W/cm²) with a cylindrical water jet at 1 Hz repetition rate. We also investigate the implementation of beam optics to manipulate the probe divergence while preserving good imaging properties. By varying the relative delay, the electron probe images reveal the dynamics of electromagnetic fields generated during the laser-water interaction. In particular, the images initially capture an ionization channel through the water vapor along the laser trajectory, followed by evidence of rapid heating and ablation of the liquid target. The electron probing results help explain some of the discrepancies observed between hydrodynamics simulations and experimental measurements. Additionally, these results show that electron probing can be used to help diagnose kinetic effects and non-equilibrium conditions in high-energy-density plasmas.   

Developing an experimental platform to benchmark x-ray fluorescence spectroscopy as a temperature diagnostic for high-energy-density plasmas

Accurate temperature models are critical in modeling planetary interiors and for inertial confinement fusion designs, as temperature affects many important physical properties such as ionization, heat transport, and compressibility. When constructing equation of state (EOS) models for high-energy-density plasmas, temperature measurements add an additional constraint to the more readily available pressure and density data. In this study, we present the development of the experimental platform and initial results for x-ray fluorescence spectroscopy (XFS) at temperatures of 10s of eV to be used as a temperature diagnostic in EOS experiments. The experiments conducted at the OMEGA laser facility use a buried layer of copper (Cu) tamped by plastic (CH) on both sides, the two sides are then irradiated by a symmetric laser drive. Simultaneous measurements using XFS, and x-ray absorption spectroscopy yield independent measures of the temperature. In addition, shock timing experiments on single-sided equivalent dimensioned targets measured with VISAR and SOP on the rear side with quartz window, to inform backlighting of XFS and x-ray absorption spectroscopy measurements. 

This work was performed under the auspices of the U.S. Department of Energy (DOE) by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344 and supported by Laboratory Directed Research and Development (LDRD) Grant No. 22-ERD-005. This material is based upon work supported by Sandia National Laboratory, managed, and operated by NTESS under DOE National Nuclear Security Administration contract DE-NA0003525.   

Optical Thomson Scattering and the Motivation for a 5ω Beamline at the NIF

The National Ignition Facility has an extensive suite of almost 120 diagnostics [1], designed to study many aspects of the capsule implosion and the exotic states of matter created in the hohlraum. However, measurements of the plasma created at the laser entrance hole (LEH) and within the hohlraum gas cannot be made. Knowledge of parameters such as the electron and ion temperatures, mean ionization state and flow velocity would allow for the benchmarking of rad-hydro codes. These can all be provided by Optical Thomson Scattering (OTS) operating in the correct regime. Developing predictive capabilities in hohlraum design will be particularly important as the NIF aims to increase laser energy and obtain ever higher yields, requiring larger capsules and new hohlraums.  This poster reviews the basic theory of OTS, the fitting process to extract key plasma parameters, and the challenges involved.   

Implementation of a Faraday Rotation Diagnostic on Gas-Puff Z-Pinch Implosions at 1-MA

Gas-puff Z-pinch implosions are magnetically driven implosions of an annular plasma sheath which is compressed onto the z-axis. Understanding the current distribution which generates the driving magnetic field within the imploding plasma sheath is key to understanding the implosion dynamics. A diagnostic which is non-perturbative and offers good spatial resolution is key to determining the exact nature of this distribution. This poster presents progress in developing a Faraday Rotation magnetic field diagnostic and current results using it to measure the magnetic field distribution. The gas-puff Z-pinches under study are generated on the 1-MA COBRA generator at Cornell University using a triple gas puff nozzle, where outer and inner annular plasma sheaths collapse onto a central target jet, compressing it. These plasmas are generated with a current rise time of approximately 100ns using argon gas. The Faraday Rotation diagnostic is combined with interferometry measurements of the electron density in order to calculate the magnetic field. These are supplemented with gated visible-UV light self-emission images, XUV (extreme ultraviolet) quadrant camera images, and PCD (photo-conducting diodes) signals to diagnose the implosion dynamics.   

Single Hit CCD Spectrometers for X-ray conversion efficiencies

X-ray conversion efficiency is often difficult to obtain given uncertainties of the detector sensitivity.  CCD cameras with known distance and attenuation, for which less than one photon is detected per pixel, can provide absolute spectra via the image histogram, but noise, fluorescence, sky-shine, and charge splitting between pixels (“charge bleed”) are challenges.  We present new results and explain features that were previously unnoticed.  We found that most hits suffer from charge bleed which may be reversed by summing over neighboring pixels, but this process becomes increasingly inefficient at exposures for which charge-bled events overlap, resulting in a continuous background instead of well reconstructed spectral lines.  We show compromises for signal strength versus quantitative fidelity, in which conversion efficiencies can be determined despite incomplete charge-bleed reversal. To demostrate the power of the diagnostic, we present the study of conversion efficincies for a laser-driven Cu-K_a source in dependence of laser-incidence angle and detector observation angle.   

Phase Contrast Imaging and Radiometry for Power Balance Studies of Magnetized Pinches

The behavior at stagnation of magnetized pinch plasmas, including gas puff Z-pinches, is thought to be strongly influenced by the partition of energy during the early time implosion phase. In particular, the partition of energy in thermal, magnetic field, radiation, and turbulent kinetic energy channels may be of direct consequence to plasma performance at stagnation – e.g. in fusion and X-ray source applications. In order to further elucidate early time energy partition, a set of millimeter wave (mmW) and infrared (IR) radiometers have been fielded both on gas puff plasmas on the 1 MA COBRA generator at Cornell University, as well as on X-pinch plasmas on the ~ 100 kA LoboLTD generator at the University of New Mexico (UNM). Additionally, a Phase Contrast Imaging (PCI) system to characterize the k-spectrum of turbulent density fluctuations on COBRA is in development. The design of the PCI and radiometer systems, their calibration, and initial experimental results will be discussed   

High-repetition-rate diagnostics and analysis techniques for HED experiments

As the number and capability of high-repetition-rate (HRR), high-intensity lasers generating data for the high-energy-density (HED) science community increases, so too must our diagnostic and analytic capabilities. Demonstrating the robustness of HRR-capable diagnostics and fast, robust analysis tools which can deliver high-quality outputs in real time, initially during burst mode operations but eventually continuously, will enable the development of a fully integrated system which can run at the maximum repetition rate of the host facility. We present progress in converting single-shot optical, neutron, charged particle and x-ray diagnostics to HRR operation, and the application of machine learning and physics-based system models to accelerate data analysis. These are important steps towards realizing autonomous active feedback control of the widest possible range of HRR HED experiments.   

Ultrafast Pixel Array Camera: Initial measurement results for a Compact 96 channel 10GSa/s data recorder for HEDS applications

We describe the design and initial characterization measurements for the UPAC-96: Ultrafast Pixel Array Camera. Upac-96 is a compact 96- channel waveform digitizing data acquisition system with a large buffer length (4096 samples per channel) and high timing performance (100ps sample time, <10ps resolution), suitable for applications in HEDP and ICF diagnostics. It is designed to work with a variety of detector arrays such as high-speed photo-diodes and fast X-ray detectors. The system is being integrated with silicon photomultiplier (SiPM) and scintillator arrays to record single-hot neutron pulses for neutron time of flight (nTOF) spectroscopy. We have measured relevant UPAC-96 performance metrics such as timing resolution, trigger rate and its ability to discern various neutron energies as needed for fast imaging and nTOF diagnostics.   

Creating observable QED collective plasma effects

QED cascades can generate electron-positron pairs when the field exceeds the Schwinger limit. This condition is expected to become available with the rapid development of ultra-strong laser sources and particle accelerators. Recent studies [1-3] have shown that the formed pairs can exhibit collective plasma effects through, e.g., plasma-induced frequency upshifts in the laser spectrum. This poster elaborates on the details of the laser-pair-plasma interaction process. It shows how creation of QED plasma also induces strong coherent laser reflection at high reflection coefficient and how plasma instabilities cause substantial laser energy scattering. 

    

An implicit particle code with exact energy and charge conservation for electromagnetic studies of dense plasmas

A collisional particle code based on implicit energy- and charge-conserving methods is presented. The particle-suppressed Jacobian-Free Newton-Krylov method is implemented as a fixed-point iteration method for the particle positions. The model can exactly conserve global energy and local charge and can efficiently use time steps larger than the plasma period. The algorithm's ability to simulate dense plasmas accurately and efficiently is quantified by simulating the dynamic compression of a plasma slab via a magnetic piston in a 1D planar geometry. Analogous results using A) 2D planar and B) 1D cylindrical geometries are presented.   

Particle-in-cell simulations of shockwaves in plasma

Plasma shocks are a ubiquitous phenomenon that occur in dynamically driven inertial confinement fusion and dense Z-pinch experiments, as well as in many areas of astrophysics. The structure of a shockwave in a plasma is more complicated than that in a neutral gas because of the disparity between the electron and ion masses and their mean free paths, the latter of which is sensitive to the charge state Z. Here, we present fully kinetic results of shockwaves in plasmas ranging from Z=1 (hydrogen) to Z=10 (neon). The simulations are performed using the energy- and charge-conserving particle-in-cell algorithm in PICNIC. In contrast to some theories for shocks in plasmas with Z>5, where the classical electron viscosity can exceed that of the ions, the electron heating in the viscous sublayer is insignificant compared to that for the ions for all values of Z considered. This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344 and was supported by the LLNL-LDRD Program under Project No. 23-ERD-007.   

SPECT3D, Imaging and Spectral Analysis Package

SPECT3D is a collisional-radiative spectral analysis package designed to compute detailed emission, absorption, or x-ray scattering spectra, filtered images, XRD signals, and other synthetic diagnostics. The spectra and images are computed for virtual detectors by post-processing the results of hydrodynamics simulations in 1D, 2D, and 3D geometries. SPECT3D can account for a variety of instrumental response effects so that direct comparisons between simulations and experimental measurements can be made. We will present new features of SPECT3D and highlight their application to the analysis of HEDP experiments. We will discuss a newly implemented capability to simulate scattering signatures from realistic experimental configurations, which include the influence of plasma non-uniformities and collecting scattered x-rays from a range of angles. Other improvements include support for a wider range of hydrodynamics codes, utilization of FAC atomic data, and improved lineshape models.   

2-D Kinetic Simulations of Biermann-battery Magnetic Field Generation and Current Sheet Formation in Laser-Solid Interactions

Magnetic reconnection is a ubiquitous process in astrophysical plasmas, and is thought to play an important role in particle acceleration and  energy conversion. Recent experiments at NIF and OMEGA provide a platform for laboratory studies of magnetic reconnection between high-energy-density plasma plumes.

Using the fully kinetic particle-in-cell (PIC) code PSC, we seek to simulate these conditions and study the plasma’s behavior. Previous 3-D studies have found the Biermann-battery effect, where magnetic fields are generated due to non-parallel density and temperature gradients, to play a crucial role in fast reconnection rates. To further investigate this effect and better match experimental conditions, a first-principles model of the laser energy deposition has been implemented. We model Biermann battery magnetic field generation in plasmas expanding from a thin target, studying the evolution of the magnetic fields and the formation of a current sheet when two plumes collide with each other.   

Self-Consistent Non-LTE Radiation Transport in High Energy Density Plasmas

Radiation transport in the high energy density regime produces changes in material conditions through the transport of significant energy in the form of radiation.  Strong coupling between radiation and matter requires a self-consistent computational approach, usually obtained with an implicit treatment.  Effective and efficient methods for providing an solution implicit in temperature and radiation field are available for physical systems under the assumption of local thermodynamic equilibrium (LTE), where the material response (at fixed density) can be fully characterized by an evolving temperature.  Many high energy density applications, particularly those with with high-Z materials, require a non-LTE treatment in which the material properties also depend on the radiation field.  Obtaining implicit solutions for these systems remains problematic.

 

We discuss methods for achieving full self-consistency for these non-LTE systems and present a radiation transport treatment which provides an implicit solution for material properties, temperature and radiation field.  The information needed to achieve consistency comes from the Linear Response Method¹, which uses coefficients describing the material response to radiation to provide a tabular approach to non-LTE atomic kinetics in the presence of a radiation field.  Using this information for radiation transport requires incorporating response coefficients over the full spectral regime into all aspects of the solution algorithm.  We discuss the resulting formulation and present progress towards an efficient implementation of this approach to non-LTE radiation transport.

 

[1] H.A. Scott, J.A. Harte, M.E. Foord, and D.T. Woods, Phys. Plasmas 29, 082703 (2022)   

[PLACEHOLDER] FUNDAMENTAL PROCESSES IN PLASMAS

.   

Observation of High Frequency Emission from the MST Tokamak Plasma in the Presence of Runaway Electrons

Local magnetic fluctuations of high-frequency emissions ranging from 37 MHz up to 5.5 GHz have been measured in a runaway-dominated MST tokamak plasma with a magnetic probe inserted into the plasma, up to <!--[if gte msEquation 12]> m:val="lin"/>r style='mso-bidi-font-style:normal'>a style='mso-bidi-font-style:normal'>~0.8 <!--[if gte msEquation 12]> m:val="lin"/>r style='mso-bidi-font-style:normal'>a style='mso-bidi-font-style:normal'>~0.8 r/a = 0.8, and digitized on a high-speed oscilloscope (up to 12.5 GHz). The emissions are observed in several frequencies that correspond to low and high Whistler bands; frequencies which would otherwise be hidden to probes outside the plasma. The magnetic emission and radial x-ray emissions bursts are correlated, consistent with pitch-angle scattering of energetic electrons. Hard x-ray measurement, using CdZnTe detector show high energy photons up to about 1 MeV. Runaway electrons are generated throughout the discharges at plasma density less than 0.03x10¹⁹ m³, toroidal magnetic about 0.14 T, and plasma current in the range of 50-60 kA. Whistler mode have been observed with toroidal magnetic field as low as 900 gauss. The emission frequencies scale with the magnitude of the toroidal field. The parallel and perpendiculars wave number have been measured as a function of frequency. The polarization of different frequency modes have also been measured. Modes at frequency 37 MHz and the 3.2 GHz are elliptically polarized. The high frequency branch of the Whistler is a backward propagating wave, consistent with anomalous Doppler. 

    

X-ray Diagnosis of Runaway Electron Distributions During Whistler-Wave Scattering Events in MST Tokamak Plasmas

 One promising route for mitigation of runaway electrons (RE) in tokamaks stems from the excitation of whistler waves due to a RE-driven kinetic instability which scatters the fast particles and as a result limits their energy. Here, we study RE dynamics during bursts of whistler-wave activity in low-density (n_e ~ 10¹⁷ - 10¹⁸ m^-3), quiescent tokamak plasmas at low toroidal field, B_T = 0.13 T, in the Madison Symmetric Torus (MST). A soft-x-ray detector operating in pulse-height mode is used to measure emission spectra due to pitch-angle scattering of the RE. The x-ray distribution is tracked through whistler burst cycles which occur at a rate of 2 kHz and each last approximately 0.3 ms. Emission spectra show a drop in the higher-energy range, 10-15 keV, aligned with the onset of whistler waves and persisting to the end of heightened activity. Correlations between waves and x-ray emission are studied at both the individual burst scale and over the full I_P flat-top in an effort to understand the energy budget of the RE. The results are compared to Fokker-Planck simulations using the CQL3D code with a synthetic x-ray diagnostic. These results improve understanding of the evolution of RE radial and velocity-space distributions during wave-scattering events, which is potentially important for avoiding tokamak damage due to high-energy RE beams.   

Development of fast ionization gauge for studying neutral flow dynamics in the BRB.

Most of plasma sources on the BRB at the Wisconsin Plasma Physics Laboratory (WiPPL) require various types of gas injection mechanisms. The breakdown of the gas, and therefore the efficacy and reliability of the plasma source itself, is heavily dependent on the neutral gas density and flow characteristics; therefore, it is critical to understand the density profiles on the time timescales of the gas dynamics (1-10 kHz). We have planned to implement a fast ionization gauge similar to that used for other experiments ^[1,2] to get local density measurements on the timescales required. The fast ionization gauge is based on a modified Bayard-Alpert gauge ^[2] designed to be compact and mobile to allow measurement over a wide range of positions to acquire a more complete and accurate density profile of the neutral gas and have a better understanding of the plasma dynamics being studied.  

[1] You, Setthivoine, et al. "The Mochi LabJet experiment for measurements of canonical helicity injection in a laboratory astrophysical jet." The Astrophysical Journal Supplement Series 236.2 (2018): 29.

[3] Yun, Gunsu S., and Paul M. Bellan. "Plasma tubes becoming collimated as a result of magnetohydrodynamic pumping." Physics of Plasmas 17.6 (2010).

[2] T. E. Weber, T. P. Intrator; A compact fast ionization gauge for in situ measurement of high-density neutral flow dynamics. Rev Sci Instrum 1 April 2014; 85 (4): 043501   

Magnetically driven plasma jet experiment on the Big Red Ball

The complex morphology and formation of astrophysical jets on all scales remains obscure due to observational limitations. The MHD scalability of numerical simulations and laboratory plasma jets to astrophysical jets offers to bridge this gap. We are generating a magnetic-tower model of nonrelativistic plasma jets driven by magnetic fields having both toroidal and poloidal components, on the Big Red Ball (BRB), a 3m diameter spherical chamber, at the Wisconsin Plasma Physics Laboratory (WiPPL). We aim to study the propagation dynamics, collimation, and stability, mainly the kink and Rayleigh-Taylor instabilities, of these magnetically driven plasma jets. Prior experiments developed jets in vacuum backgrounds and smaller chambers. Our goal is to launch plasma jets in pressurized plasma background and investigate the formation of shocks and their precursors between the magnetized jet and the unmagnetized background plasma and how the development of the kink instability could depend on the background pressure. Our experimental configuration comprises bias field coils generating poloidal magnetic fields and a plasma gun consisting of neutral gas-puff drives and a disc-shaped cathode with a coplanar annulus-shaped anode with high voltage difference across them. We will present the preliminary results of the characterization of plasma jets from time and space resolved measurements of plasma parameters, including magnetic field, electron and ion densities, temperatures, and velocities.   

Equilibrium reconstructions of tokamak and RFP plasmas above the Greenwald density limit in MST

Recent experiments in the Madison Symmetric Torus (MST) have demonstrated the capability of sustaining tokamak and reversed-field pinch (RFP) plasmas with density above the Greenwald limit, n_G. Here, we present toroidal equilibrium reconstructions of these plasmas using the MSTFit code, constrained by line-integrated electron density measurements from an 11-chord far-infrared interferometer and, in some cases, insertable probe measurements of the magnetic field. The capability of sustaining plasmas with n > n_G is likely due in part to a high-voltage feedback power supply system driving the plasma current and a close-fitting conductive shell with resistive wall time 0.8 s. In the tokamak configuration with toroidal field B_T = 0.13 T, variations in density and toroidal current profiles are studied for central density ranging up to 10 n_G, with a focus on sudden broadening of the profiles near 2 n_G, which persists to higher densities. In the RFP configuration with plasma current I_p ≤ 150 kA, the maximum density is not precisely known but is thought to be around 2 n_G. Efforts to resolve this uncertainty are discussed, as well as plans for further diagnosis of high-density plasmas in MST.   

Rotating magnetic field system to emulate a pulsar magnetosphere in BRB

We are building a dipolar rotating magnetic field (RMF) system with the intention to emulate the magnetosphere of an obliquely rotating pulsar and demonstrate production of an outgoing plasma wind in the Big Red Ball (BRB) device at the Wisconsin Plasma Physics Laboratory (WiPPL). The driver will consist of an orthogonal pair of Helmholtz-like RMF drive coils in the center of BRB powered by two tube-based amplifiers. The coils will be placed at atmospheric pressure in an alumina-coated fiberglass pressure vessel within BRB. A power of 150 kW per channel will be fed to the drive coils via transmission lines running into the pressure vessel. Fuel gas will be puffed from outlets on the surface of the vessel, and the RMF will ionize the gas. The experimental hardware is in development. The first of two coupling transformers has been constructed, installed, and tested. Transmission line hardware design is finalized, and assembly of the transmission lines and RMF coils is pending. The pressure vessel has been painted with alumina to protect the fiberglass surface from the plasma. Key next steps are to test the RMF drivers at full power prior to installation on BRB, implementing the gas puff system, and initial plasma tests. Once the driving system is operational, we expect to form a striped plasma wind similar to that in the neighborhood of a neutron star and potentially to observe magnetic reconnection within the stripes.   

Compact toroid injection for the MST experiment

We are developing experiments to inject compact toroid (CT) plasma into the Madison Symmetric Torus (MST) device and measure the plasma response with an emphasis on possible ion energization due to magnetic reconnection. CT injection has previously been experimentally tested for plasma fueling, rotation, and improved confinement. To understand the dynamics of a CT in a tokamak magnetic field, we adapt models that treat it as a moving spheromak plasmoid with an intrinsic magnetic moment. After injection into the tokamak, the modeled CT tilts and decelerates due to MHD wave drag. We expect magnetic reconnection to occur as it loses its kinetic and magnetic energy and dissipates in the background plasma. A high-order Runge-Kutta method is used to calculate CT trajectories in MST for different parametric cases and initial conditions. Using an injector developed for the Big Red Ball, CTs will be injected into MST at speeds of order 100 km/s. During tokamak operation, the MST device has a toroidal magnetic field strength of 0.13 T on the axis and electron densities of several times 10¹⁸ m^-3. Changes in ion temperature in the background plasma due to injection will be measured with passive Ion Doppler Spectroscopy. We also plan for experiments to inject CTs into RFP plasmas in the future.   

Relaxed-state Modeling of RFP and Low-q Tokamak Plasma Formation with Programmable Power Supplies

The plasma current in a reversed-field pinch (RFP) is sustained by a combination of inductive drive and magnetic self-organization. The current profile exhibits rigidity due to the onset and nonlinear saturation of tearing instabilities, as the tendency for profile-peaking in the core is arrested by a dynamo-like emf, which simultaneously drives large poloidal current that sustains a reversed toroidal field. In the Madison Symmetric Torus (MST), the RFP equilibrium forms spontaneously (self-reversal) if the initial toroidal field is not too large. New programmable power supplies (PPS) are installed on MST that provide high-bandwidth arbitrary waveform control of the poloidal and toroidal magnetic field circuits. They greatly expand the possible ways to form and manipulate RFP plasmas, and also enable access to relatively-unexplored low-q and ultra-low-q tokamak plasmas. We provide an overview of the PPS and show comparisons with relaxed-state modeling that assumes a rigid current profile shape, with plasma resistance projected from empirical trends for the electron temperature (related to confinement scaling). For example, the modeling predicts the inductive volt-seconds required to reach maximum current depends on the initial toroidal field. A practical outcome is that MST might be capable of larger plasma current by simply optimizing startup using PPS. The study will include low-q and ultra-low-q tokamak plasmas, which exhibit some similarities to RFP self-organization.   

Overview of Wisconsin Plasma Physics Laboratory research

The Wisconsin Plasma Physics Laboratory (WiPPL) is a multi-device collaborative research facility supporting experiments in basic, astrophysical, and fusion plasma science. WiPPL is a founding member of the MagNetUS experimental plasma network, and outside collaborator run time is allocated via a US DOE joint proposal review process for the participating MagNetUS facilities. We present an overview of WiPPL capabilities, recent and ongoing projects, and key WiPPL-led results. In the BRB device, collisionless reconnection is examined with the use of a new drive system upgrade that achieves Lundquist numbers reaching into the kinetic regime. First experiments utilizing a planar spheromak injector to mimic astrophysical jets are underway, and work to generate a rotating magnetic dipole to emulate a pulsar wind is ongoing. In the MST device, tokamak plasmas are produced for a variety of studies: dynamics, transport, and acceleration of runaway electrons, whistler-range wave correlation with runaway electrons, self-organization of low-q tokamaks, and high density exceeding the Greenwald limit. RFP plasmas in MST are used for studies of plasma self-organization, with programmable power supplies expanding the Lundquist-number overlap with nonlinear MHD simulations. MST's high-temperature plasmas and high availability make it a useful source for the development of advanced x-ray diagnostics.   

Measurements and modeling of runaway electron radial transport in quiescent, low-density MST tokamak plasmas

Understanding the creation and radial transport of runaway electrons (RE) in tokamak plasmas is vital for safety and preventing machine damage during plasma disruptions. We use a Multi-energy Soft X-Ray (ME-SXR) camera on the Madison Symmetric Torus (MST) to probe the radial and energy distribution of photon emission using a 450 um silicon detector. Low-density (10^17 – 10^18 m^-3) quiescent plasmas, with B_T = 0.13 T and T_e ~ 100 eV are tailored such that a RE population with energy ~ 10-100 keV grows during the plasma current flat-top and produces SXR radiation via pitch-angle scattering. Ongoing work is being completed to model RE acceleration and diffusion rates in MST using the Fokker-Planck code CQL3D with a synthetic ME-SXR diagnostic. Also discussed are temporal variations in the SXR emission; shot-to-shot reproducibility and ensemble-averaging efforts; and plans for further RE studies using the ME-SXR diagnostic.   

Withdrawn

Nonlinear MHD fluctuations appear in both natural and magnetic confinement settings, such as the solar wind, self-organization dynamics in the RFP and spheromak, and current disruptions in tokamak plasmas. Here we describe parameter scaling experiments oriented toward nonlinear MHD dynamics in RFP plasmas. Experimental data are gathered spanning a wide range of Lundquist number, S ∼ 10⁴ −10⁷ by varying the plasma current and density, n_e/n_G , where n_G is the empirical density limit. Magnetic fluctuation amplitudes, b_n, for toroidal modes n= 7-10 resonant at mid-radius are scaled as b_n ∝ S^-α, with α=0.3  for <!--[if gte msEquation 12]> style='font-family:"Cambria Math",serif;mso-ascii-font-family:"Cambria Math";
mso-fareast-font-family:"Times New Roman";mso-hansi-font-family:"Cambria Math";
mso-bidi-font-family:"Times New Roman";font-style:italic;mso-bidi-font-style:
normal'>n _{style='mso-bidi-font-style:normal'>e} n_e/n_G =0.23 and α= 0.4<!--[if gte msEquation 12]> style='font-size:12.0pt;font-family:"Cambria Math",serif;mso-fareast-font-family:
"Times New Roman";mso-bidi-font-family:"Times New Roman";mso-ansi-language:
EN-US;mso-fareast-language:EN-US;mso-bidi-language:AR-SA'> m:val="roman"/>α= 0.4 for n_e/n_G =0.34. A new programmable power supply enabled low-current, low-S operation, which overlaps parameters available in numerical MHD modeling. Experimental S scalings for b_n agree well with those from the DEBS and NIMROD codes. A transition from bursty relaxation with discrete sawteeth at high current to quasi-continuous relaxation at low current is observed at threshold S ∼ 10⁵. The spectral properties of the magnetic fluctuations change, including a reduction in phase velocity that suggests plasma flow and/or flow profile changes. Work supported by the U.S. DOE and an M. Hildred Blewett Fellowship.   

Data driven discovery of system equilibration

The field of dynamical systems is undergoing a revolution. In the last 5 years, many tools have emerged which discover simple governing equations from timeseries data [arXiv:2102.12086 (2021)]. The discovered model is often simpler than the original data generating model! Recently, data driven discovery allows us to observe a transient system’s dimension reduction as it evolved to attractor [J Stat Phys 179, 1028–1045 (2020)]. This capability allows physics to gain traction on long-standing problems. One such problem is Hilbert’s 6^th problem, which challenged physics to formulate the limiting processes that leads from the unequilibrated atomistic systems to equilibrated continuum systems. It has been shown that transient non-equilibrium dynamics explore a larger dimensional space and that some dimensions are removed by fast relaxation towards the equilibrium attractor [Bulletin of the AMS 51.2, 187-246 (2014)]. In this work, we formulate Hilbert’s 6^th problem in an operator framework and use data driven discovery to observe the aforementioned fast relaxation towards the hydrodynamic attractor.   

High-Resolution Three-Dimensional Particle-In-Cell Simulations of Two-Dimensional Bernstein-Greene-Kruskal Modes

We will report the latest three-dimensional (3D) Particle-In-Cell (PIC) simulations to study the stability of two-dimensional (2D) Bernstein-Greene-Kruskal (BGK) modes [Ng, Phys. Plasmas, 27, 022301 (2020)] in a magnetized plasma with a finite background uniform magnetic field. The simulations were performed using the Plasma Simulation Code (PSC) [Germaschewski et al., J. Comp. Phys., 318, 305 (2016)]. These modes are exact nonlinear solutions of the steady-state Vlasov equation with an electric potential localized in both spatial dimensions perpendicular to the axial magnetic field that satisfies the Poisson equation self-consistently. These solutions have cylindrical symmetry and are invariant along the axial direction, with distribution functions depending on the particle energy, the axial component of the canonical angular momentum, and the axial component of the canonical momentum. We will present simulation results using higher resolutions to study whether solutions stable in 2D simulations can remain stable as the box size along the axial direction increases. We have also utilized the exact form of the solution for the initial conditions of the simulations. We will present our most recent results using this new method of initialization to study kinetic instabilities as seen in our latest 2D simulations.   

Particle-In-Cell Simulations of Two-Dimensional Bernstein-Greene-Kruskal Modes using Exact Analytic Distributions as Initial Conditions

We will report the latest two-dimensional (2D) Particle-In-Cell (PIC) simulations to study the stability of 2D Bernstein-Greene-Kruskal (BGK) modes [Ng, Phys. Plasmas, 27, 022301 (2020)] in a magnetized plasma with a finite background uniform magnetic field. The simulations were performed using the Plasma Simulation Code (PSC) [Germaschewski et al., J. Comp. Phys., 318, 305 (2016)]. These modes are exact nonlinear solutions of the steady-state Vlasov equation with an electric potential localized in both spatial dimensions perpendicular to the axial magnetic field that satisfies the Poisson equation self-consistently. These solutions have cylindrical symmetry and are invariant along the axial direction, with distribution functions depending on the particle energy, the axial component of the canonical angular momentum, and the axial component of the canonical momentum. Our previous runs used local Maxwellian approximation of the analytic electron distribution functions by matching the first three moments of density, flow velocity and temperature at each spatial location. Now, we present runs using the exact analytic distributions as initial conditions. We found that the modes are significantly steadier in time for cases with large background magnetic fields, but exhibit unique wave patterns apparently from kinetic instability for cases with weak background magnetic fields, as compared to the local Maxwellian runs. Such wave patterns begin as co-rotating azimuthal electrostatic perturbations, and evolve into spiral shapes later in the instability.   

Local Kinetic Analysis for the Instability Observed in Particle-In-Cell Simulations of Two-Dimensional Bernstein-Greene-Kruskal Modes

In the latest two-dimensional (2D) Particle-In-Cell (PIC) simulations to study the stability of 2D Bernstein-Greene-Kruskal (BGK) modes [Ng, Phys. Plasmas, 27, 022301 (2020)] using the Plasma Simulation Code (PSC) [Germaschewski et al., J. Comp. Phys., 318, 305 (2016)], an instability was observed in runs with weak background magnetic fields which exhibits a unique wave pattern, apparently from a kinetic instability. Such wave patterns begin as co-rotating azimuthal electrostatic perturbations, and evolve into spiral shapes later in the instability. We will present local analysis along the azimuthal direction based on kinetic theory to try to understand the simulation results. Growth rates and phase velocities of different azimuthal mode numbers will be compared with results from the PIC simulations.   

Simulated annealing of reduced magnetohydrodynamics by mixed Hamiltonian and artificial dynamics

Simulated annealing (SA)[1], a kind of relaxation method, has been applied to equilibrium calculations and stability analyses of reduced magnetohydrodynamics (RMHD) systems[2, 3].  The artificial dynamics of SA is derived utilizing Hamiltonian nature of the system so that the energy of the system monotonically changes while the Casimirs are preserved.  An energy minimum is an equilibrium of RMHD, which can be reached by SA.  Also, the stability of a given equilibrium can be analyzed by SA from the view point of energy.  Since SA solves an initial-value problem for the relaxation, the simulation time tends to be longer, especially near the equilibrium.  Therefore, accelerated relaxation is necessary.  We have developed such a method in [3].  In this presentation, we examine whether the relaxation can be accelerated if we include the original Hamiltonian dynamics in addition to the SA dynamics.  Numerical results show that the inclusion of the Hamiltonian dynamics rather slows down the relaxation to the equilibrium in the cases tested until now.

[1] G. R. Flierl, P. J. Morrison, Physica D 240, 212 (2011). 

[2] Y. Chikasue and M. Furukawa, Phys. Plasmas 22, 022511 (2015),  M. Furukawa and P. J. Morrison, Plasma Phys. Control. Fusion 59, 054001 (2017),  M. Furukawa, Takahiro Watanabe, P. J. Morrison, K. Ichiguchi, Phys. Plasmas 25, 082506 (2018).

[3] M. Furukawa and P. J. Morrison, Phys. Plasmas 29, 102504 (2022).   

Freestanding Crossed-Field Plasma Rotation with Suppressed Dissipation

The most common way to produce a rotating magnetized plasma in the laboratory is to impose perpendicular magnetic and electric fields oriented so as to drive rotation through drift motion. In a high-temperature linear device, the field lines are often isopotentials, so that the same voltage drop imposed across the core of the plasma will also appear at the edge, and impinge on the plasma-facing components. This limits the maximum possible voltage drops that can be sustained; above some threshold, the plasma-facing components cannot tolerate the resulting electric fields [1]. Here we consider the possibilities -- and the challenges -- associated with alternative configurations that maintain voltage differences along field lines, balancing longitudinal and perpendicular voltage drops. These configurations could in principle avoid material limitations on the maximum possible voltage drop, but there are nontrivial constraints that must be taken into account in order to avoid unacceptably large dissipation. Nonetheless, such configurations suggest the possibility of greatly increasing the DC voltage drops maintainable in laboratory and industrial settings.

[1] J.-M. Rax, R. Gueroult, and N. J. Fisch, Phys. Plasmas 30, 072509 (2023).   

A canonical vorticity probe formed by four tetrahedral clusters of Mach and magnetic probes to measure the evolution of ion canonical vorticity during RFP relaxation

Conversions between twisted magnetic and ion flow vorticity flux tubes provide an alternative viewpoint of self-organized processes in plasmas. We have built a canonical vorticity probe capable of simultaneously measuring the magnetic field and ion flow vorticity during sawtooth relaxation in RFP plasmas on MST. The probe consists of tetrahedral clusters of 3x magnetic (B-dot) and 4x Mach probes. The three-dimensional 3D Mach vector is determined from the geometric addition of the logarithms of ion saturation currents collected at the vertices of a tetrahedron, not 180 degrees apart. The clusters are arranged in a tetrahedral finite difference stencil so that the curl of these vectors can be determined. The full probe an two prototypes of a single cluster have been constructed [1] and fielded on MST to examine self-organization and sawtooth relaxation. In the MST RFP, tearing instabilities drive a reorganization of current and momentum. Recent two-fluid simulations indicate that there can be a change in cross helicity, the interlinking of magnetic and flow vorticity flux tubes, during sawtooth events. This work will help generalize our understanding of plasma relaxation to a canonical-helicity-constrained relaxation explaining multi-scale dynamics.   

Relaxation of current-carrying flux tubes and flux rope formation

Flux ropes are ubiquitous structures in space and laboratory plasma environments. Although their equilibrium states have been extensively studied, how they relax from non-equilibrium states is an important process that has been relatively less explored. By analogy to previous studies on current sheets, particle orbits are analyzed and their phase-space trajectories are studied. Analytical calculations show that orbital transitions must occur in a non-equilibrium current sheet and leave distinct footprints in the phase-space distributions. These footprints are verified by kinetic simulations of disequilibrated current sheets, confirming the relaxation process. It is shown that any small parallel current will relax to a flux rope configuration.   

Relative Rydberg Atom Formation Rates as a Function of Electron Magnetization in Ultracold Neutral Plasmas

Rydberg atom formation via three-body recombination in ultracold neutral plasmas is interesting both because the main limitation on the coldest achievable electron temperatures comes from three-body recombination heating and simulating Rydberg formation numerically has challenges such that experimental tests are useful in evaluating theoretical predictions. Different plasma systems such as those associated with antihydrogen formation are strongly impacted by three-body recombination, too. Recombination rates are predicted to decrease substantially as a function of magnetic field in the plasma. Ultracold neutral plasma electrons can become extremely magnetized at comparatively low laboratory magnetic fields and so are excellent systems to conduct studies of magnetization effects on Rydberg atom formation rates.  We describe our techniques for measuring these rates experimentally along with explorations of a new method for calibrating the strength of electric fields in the region where the ultracold neutral plasmas are formed in our apparatus.   

Numerical Investigation of Ionization Dynamics in Intense Laser-heated Argon Plasmas using the NLTE Model

Ionization dynamics of laser-produced plasma play a pivotal role in comprehending phenomena in larger macroscopic systems, including thermal transport and plasma instabilities. This study investigates the ionization dynamics of Argon gas plasma (N_e: 10¹⁹ ~ 2.5x10¹⁹ cm^-3) irradiated by intense laser beams using the widely used NLTE model, CRETIN [1, 2]. Our objective is to accurately determine the ionization process. Under normal circumstances, the photon energy of an optical laser (~3 eV) is typically insufficient to overcome the binding energy of bound electrons, resulting in a low probability of photoionization. However, when a high-energy laser with an intensity of approximately 660 kJ (60 kJ × 11 beams) is focused on the plasma, the abundance of photons leads to multi-photon ionization. This process generates numerous highly excited states through dielectric capture (or inverse process of Auger ionization) of the free electrons as well as various excitation processes. Consequently, even with low photon energy, the result demonstrates the photoionization process can be dominant throughout the experimental time range and dynamics may not reach a steady state. This research enhances our understanding of ionization phenomena in laser-produced plasma and provides valuable insights for various applications.   

Experimental Measurement of Refilling Rates in the High-Energy Tail of the Thermal Electron Distribution of Ultracold Neutral Plasmas

Plasma electrons and ions in the high-energy tails of velocity distributions are of interest because of their importance in fusion, Tokamak, runaway electron, and fundamental physics research. Ultracold neutral plasmas (UNPs) can be studied in controlled table-top conditions, providing an experimental setting for studying the behavior of the rate at which electrons populate the high-energy tail of their velocity distribution. Measured rates can be used to test theoretical descriptions that are applicable not only to electrons but other particles in the plasma. We have developed an applied electric field sequence that depletes and then measures the refill rate of electrons in the high-energy tail of the velocity distribution without experimental distortions that were previously present. We can model the UNP electron extraction and fill rates using a Molecular Dynamics (MD) code to convert our measured data to quantities that can be compared to theory predictions. Measured refill rates along with descriptions of the experimental challenges will be presented.    

Opacity calculations: including more and more states

One of the main challenges of opacity calculations is the balance between accuracy and completeness, i.e. the search for the best compromise between the use of elaborate and precise physical models, and the number of states (levels, configurations, superconfigurations) included in the computation. Statistical methods have the advantage of representing a large number of states, but the corresponding spectra are often too crude for some applications. In this work, we discuss the strengths and drawbacks of different approaches aiming at reaching the best compromise between completeness and accuracy.

Firstly, we propose to start from global approaches and refine them, either by combining statistical approaches such as Unresolved Transition Arrays, Spin-Orbit Split Arrays or Super Transition Arrays (STA), with Detailed Line Accouting (DLA) calculations [1], or by revealing the underlying structure of the spectra as in the Configurationally-Resolved Super Transition Array (CRSTA) [2] or Partially-Resolved (PR)-CRSTA [3] approaches.

In a second step, we try to go the other way, starting with detailed calculations and including (even in an approximate manner) more spectral features.  In multiply-charged ion plasmas, a significant number of electrons may occupy high-energy orbitals. These Rydberg electrons, when they act as spectators, are responsible for a huge number of satellites of X-ray absorption or emission lines, yielding a broadening of the red wing of the resonance lines. We suggest to model the perturbation induced by the Rydberg spectators as a partial DLA calculation, their effect being included through a shift and width, expressed in terms of the canonical partition functions, which are key-ingredients of the STA model [4].

 

[1] G. Hazak and Y. Kurzweil, High Energy Density Phys. 8, 280 (2012).

[2] Y. Kurzweil and G. Hazak, Phys. Rev. E 94, 053210 (2016).

[3] J.-C. Pain, F. Gilleron and T. Blenski, Laser Part. Beams 33, 201 (2015).

[4] J.-C. Pain and F. Gilleron, High Energy Density Phys. 15, 30 (2015).   

Unification of Gravity and Electromagnetism

Gravity and Electromagnetism are two sides same coin, which is clue of this unification. Gravity and Electromagnetism are represented by two mathematical structures, symmetric and antisymmetric respectively. Einstein gravitational field equation is the symmetric mathematical structure. Electrodynamics Lagrangian is three parts,for Electrodynamic field,Dirac field and interaction term. The definition of canonical energy momentum tensor was used for each term in Electrodynamics Lagrangian to construct the antisymmetric mathematical structure; symmetric and antisymmetric gravitational field equations are two sides of the same Lagrangian.   

Ubiquitous Plasma Motions in Magnetized Plasma Arcades with Different Experimental Configurations

Magnetized plasma arcades and flux ropes appear in space and astrophysical contexts and have been the subject of recent laboratory, observational, and computational studies. We generate arcade-shaped plasmas in the lab using two parallel copper electrodes and a magnetic coil surrounding the electrodes. We compare and contrast behavior with short (5 cm) and long (20 cm) cathodes and a variety of hydrogen gas fueling methods with the aim of more reproducible plasmas amenable to combining measurements across multiple shots. ICCD cameras, photodiode arrays, and fast current and voltage measurements are used to characterize the discharges. We observe that all configurations share the common features of (i) shifts in discharge location along the electrodes within a shot and (ii) repetitive, fast-moving current filaments that propagate through the arcade. Both types of plasma motion persist even when fueling through a single hole in the cathode. The most reproducible operations are obtained with high vacuum magnetic fields, localized fueling through the cathode, and diffuse fueling near the anode. However, even in this case, significant shot-to-shot variations remain.   

Withdrawn

The highly stressed, Earth-ward propagating plasma resulting from a reconnection event is known as a dipolarization front because of its strong dipole (B_z) field. At its Earth-facing edge, there are sharp discontinuities in plasma flow, density, temperature, and electromagnetic fields. This region of space contains large releases of energy driven by sheared flows and fields. Our work is based on a non-local theory for the generation of shear-driven electrostatic Electron-Ion Hybrid (EIH) waves with frequencies near the lower hybrid frequency. We compare analytical and numerical dispersion relations to find their range of agreement. We find that the wave number of the waves varies inversely to velocity shears.  We then compare numerical simulations and experimental data to establish a threshold for wave growth. We expect that the likely length scale of the threshold for wave growth is at the electron Larmor radius.   

Acceleration of relativistic particles and gamma-ray emission in standing Alfvén waves

In a strong magnetic field where the cyclotron frequency exceeds the frequency of electromagnetic waves, the scattering rate of the relativistic intensity wave will be high, and it is not evident that transmission is possible. Such considerations can constrain the theoretical modeling of the magnetosphere of planets and neutron stars. Therefore, we have been analyzing wave-particle interactions under such extreme plasma conditions using PIC simulations. In this presentation, we will primarily focus on the collision process of Alfvén waves. We theoretically show that all electrons with non-relativistic speeds are accelerated to relativistic speeds in standing waves created by counter-propagating circularly polarized Alfvén waves. The condition to realize such acceleration is when the amplitude of the electromagnetic wave is larger than the background magnetic field. Furthermore, most of the energy of the original electromagnetic wave can transfer to the radiation energy of gamma rays, which may lead to pair production. We will also discuss the possibility of future proof-of-principle laser experiments of this mechanism.   

Determining wave fields from perturbed particle orbits.

We test a procedure for determining electric fields over a finite volume based on measurements of a space, time, and velocity resolved measurement of a velocity distribution function at a single location. The technique is based on an exploitation of the non-local dielectric response of weakly-collisional plasma. As an example, we consider electron plasma waves in a 1-D simulation with periodic boundary conditions. Previously, it was found that ion-acoustic fields over a finite volume can be determined from a local measurement of the ion distribution function directly by means of an integral transform introduced by Morrison [1]. That technique cannot straightforwardly be applied in circumstances where one needs to perform a Fourier transform in space in order to construct the Morrison transform. Since most wave modes require Fourier transformation, it is of interest to generalize the procedure so that it can be applied to the more general case. The new procedure is tested for the electron plasma wave is with numerically generated data from a Vlasov code. 

[1] F. Skiff, H. Gunell, A. Bhattacharjee, C. S. Ng, and W. A. Noonan, Phys. Plasmas 9 pp 1931-1937 (2002).    

Two-fluid Gkeyll Simulations of Alfven Wave Reflection From an Alfven Speed Gradient in LAPD

The heating of the solar corona and acceleration of the solar wind is likely driven by Alfvenic turbulence, which requires counter-propagating Alfvenic fluctuations. Alfven waves are observed to be driven from the base of the corona, but the source of inward propagating waves is not yet established. Based on Magnetohydrodynamic (MHD) theories, the leading candidate is reflection from an Alfven speed gradient in the corona. However, prior experimental tests of Alfven wave reflection from a magnetic field gradient in the LArge Plasma Device (LAPD) at UCLA do not agree with the MHD reflection predictions, possibly due to physics beyond MHD. In this talk, we present the Gkeyll simulation framework as general use tool to model LAPD. In this case, we use the Gkeyll two-fluid solvers to explore the role physics beyond MHD may play in the reflection of Alfven waves. We compare Gkeyll simulations to previous LAPD experiments for which reflection was not observed, and we present recent LAPD experiments that do exhibit reflection. In both cases, we find that two-fluid physics well models Alfven wave reflectance in LAPD experiments.   

Helicity dependence in interactions between pairs of arched, laboratory plasma loops

Arched, plasma-filled flux tubes -- or "loops" --- can be generated "on demand" in laboratory vacuum chambers with pulsed-power, magnetized plasma guns.  These plasmas exhibit rich dynamics during their brief lifetimes (which are measured in microseconds), ranging from behavior that is well described by ideal MHD --- and thereby more scalable to phenomena observed in the solar atmosphere -- to phenomena that access non-ideal-MHD regimes, e.g., involving kinetic instabilities, Hall MHD, or middling Lundquist numbers [1].  Over the years, a number of different gun designs have been employed to create a wide variety of single- and multiple-loop plasma structures.  Among these is a "quad" gun capable of producing two adjacent loops, each of which can be set up to have left-handed (L) or right-handed (R) helicity.  There are thus two different co-helicity configurations (LL and RR) and two different counter-helicity configurations (LR and RL); whether the loops have the same or opposite helicity was seen to affect how they interact with each another as they evolve [2].  Here, we describe experiments in which this gun was operated with dual-gas techniques (so that the two loops could initially contain different ion species and thereby be distinguished in fast camera images with optical filters), and B-dot probe measurements were taken of the loop pairs as they evolved and "erupted".  Particularly striking are the differences between the two "counterhelicity" configurations (LR and RL); one configuration yields bright, fast, dynamic merged structures, whereas the other stagnates.  These results extend previous studies on two-loop interactions and have potential to shed new light on the non-ideal-MHD effects at play.

[1] Bellan, P. M. (2020). Caltech lab experiments and the insights they provide into solar corona phenomena. Journal of Geophysical Research: Space Physics, 125, e2020JA028139.

[2] Hansen, J. F. and Tripathi, S. K. P. and Bellan, P. M. (2004) Co- and counter-helicity interaction between two adjacent laboratory prominences. Physics of Plasmas, 11 (6). pp. 3177-3185.   

Tokamak Energy selected for U.S. DOE Milestone-Based Fusion Development Program

Tokamak Energy Ltd is delighted to have Tokamak Energy Inc, its West Virginia-based U.S. subsidiary, selected by the U.S.DOE for an award as part of DOE's bold decadal vision for delivering commercial fusion energy (DE-FOA-0002809). It's a great opportunity and endorsement for the company-team's people and technology. As part of the project, Tokamak Energy will work with U.S. national laboratories and organisations to address major technical and commercialisation milestones for the successful design of a fusion pilot plant. In addition, since the INFUSE program was established in 2019, the company has received multiple INFUSE awards that enable the acceleration of fusion energy development through public-private research partnerships in the US. This talk will outline some of the key technical milestones for the initial 18-month phase of the five-year Milestone-Based Fusion Development Program.   

Optimization of a Diagnostic Suite for Magnetized Target Fusion Experiments

Future Magnetized Target Fusion (MTF) experiments at General Fusion (GF) will attempt to compress a plasma with an imploding liner to 10 keV and approach the Lawson criterion. A suite of diagnostic sensors will measure plasma density n, temperature Te, and energy confinement time at various compression stages. The suite will maximize accuracy on these parameters or their source quantities (magnetic energy, Ohmic power). Other plasma parameters (q and lambda profile) will also be quantified to monitor plasma instabilities and characterize the compression target.

Machine learning algorithms are applied to libraries of Grad-Shafranov (GS) equilibria calculated with GF's custom solver, and to scaling laws from simulated compression cases. The data are processed by a dimensionality reduction algorithm paired with a clustering algorithm, which identifies key sensor positions. The expected error on each parameter is back-calculated with a reconstruction algorithm. A Monte Carlo error propagation algorithm is finally applied to each case, in order to estimate the impact of measurement accuracy on the reconstructed values. Preliminary results show that reliable measurements of magnetic parameters can be achieved with 10 surface probes, with errors on magnetic energy as low as 10%. The same analysis will be expanded to include more diagnostics, and to several stages of compression.

A custom Monte Carlo algorithm paired with physics-based constraints will predict the expected error on the triple product, quantify the accuracy required on each diagnostic system, and provide a reliable confirmation of the main scientific milestones.   

Characterizing the Performance of a Cylindrical Electrostatic Analyzer for Measurements of Ion Beam Energy

The direct measurement of electric potential in the plasma interior remains a need on many magnetic confinement devices. Such measurements can advance understanding of the electric field and its role in particle transport. Traditional beam probes can fulfill the need by measuring the energy difference between injected and detected beams, yet the diagnostic can be large and expensive. We are developing a new energy analyzer that uses cylindrical annulus sector electrodes, rather than flat electrodes, which lowers the maximum electrode voltage required, thereby reducing the overall cost of the diagnostic. It combines two subsystems, one that measures the angle of the beam ions entering the analyzer, and one that measures their energy via deflection through an electric field. Determination of the detected beam energy using this measurement combination achieves the energy resolution, 0.01% of the injected beam energy, of traditional analyzers. We have designed and built a prototype of the analyzer, installed it on a beam teststand, and assessed its performance by injecting a singly charged ion beam with precisely known energy while varying the cylindrical electrode voltages and beam angle. The data is then analyzed by performing a multiple linear regression of the beam energy in terms of the deflection and angle measurements. The resulting empirical analyzer response will be used to infer the beam energy and plasma potential when the system is operated on a plasma device. Measured signals and results of our characterization will be presented.   

A novel spectral filtering technique to improve incoherent Thomson scattering diagnostics

Incoherent Thomson scattering, or simply TS, diagnostics are ubiquitous in the plasma physics community and have had unparalleled success in providing direct measurements of plasma electron temperature (T_e) and density (n_e). Many plasma experiments use polychromators with three to eight wideband spectral channels to spectrally resolve measurements of Thomson scattered radiation. Theoretically, increasing the number of spectral channels in a polychromator can increase the T_e measurement range and decrease overall measurement uncertainties. However, in practice, there is a direct trade-off between the number of spectral channels and the signal-to-noise ratio of the measurement.

 A novel filtering technique called Thomson Scattering Spectral Multiplexing (TSSM) has been developed at the Helically Symmetric eXperiment (HSX) in order to circumvent the limitations imposed by existing polychromator systems. TSSM implements an optimized set of multiple-band dichroic filters with two spectral passbands per filter instead of only one passband. Comparisons of the expected measurement uncertainty between a standard TS approach and TSSM show that spectral multiplexing not only increases the measurement range of T_e but also reduces the average measurement uncertainty and the deviation in the uncertainty over the increased range of T_e. Calculations, calibration data, and experimental results for the TSSM system on HSX are presented.   

Observation of Supersonic FRC Merging Process using Wide-View Visible-Light Tomographic Camera

At Nihon University, experiments on the collisional merging experiments of Field-Reversed Configuration (FRC) have been conducted using the FAT-CM device. The time evolution of the internal structure during this collisional merging process, which is presumed to involve dynamic structural transitions, has not yet been observed. The tomographic camera (T-cam) has been developed to visualize FRCs through the imaging of visible light. While it has successfully reconstructed the internal structure of FRCs and determined the center of gravity of plasmoids during the translation process, it has been incapable of doing so during and after the collisional merging process due to its narrower field of view than the emission area. The observable range of the newly developed T-cam encompasses most of the area within the confinement chamber. This advancement is expected to enable the reconstruction of the collisional merging process and the subsequent state. This work reports the insights from the observations made using the newly built T-cam.   

Development and Test Results of a Modified Crystal Cathode Pressure Gauge for KSTAR

Neutral pressure measurements in the tokamak divertor region can support to control the heat flux and help to understand the plasma behavior in the divertor region. In addition, as KSTAR has recently replaced carbon-based divertors by tungsten-based ones, similar to ITER, understanding how neutral pressure in the divertor region of KSTAR affects the plasmas can provide valuable information to develop ITER operation scenarios. For these reasons, six crystal cathode pressure gauges (CCPG) [Wenzel, U., et al., Rev. Sci. Instrum. 89 033503 (2018)] are planned to be installed in the KSTAR divertor region. Since accessibility to these CCPGs will be limited after installation of the new divertors, we have modified the conventional LaB₆ crystal cathode with the aim of achieving better mechanical stability. Due to its higher heat capacity compared to the conventional one, the modified cathode has a slower response time for controlling the electron emission rate, which adversely affect the measurements. We are developing an improved feedback control algorithm to compensate for the slower response time and will present the test results together with the modified cathode design.   

Absolute electron density fluctuation reconstruction for two-dimensional Hydrogen-beam diagnostics

Scrape-off layer (SOL) and edge plasma turbulence contribute significantly to the radial particle and heat transport tampering plasma confinement. SOL turbulence is predominantly intermittent which manifest in the occurrence of isolated density filaments or blobs. Filaments propagate radially outwards towards plasma facing components limiting their lifetime by erosion and sputtering. To characterize this phenomenon in detail few diagnostic techniques are available. Beam emission spectroscopy is a diagnostic capable of measuring plasma turbulence in both SOL and edge plasmas. Due to the finite lifetime of the excitation states during the beam - plasma interaction, and the misalignment between the optics and the magnetic field, spatial smearing is introduced in the measurement. We introduce a novel method to overcome this hindering effect by inverting the fluctuation response matrix on an optimally smoothed signal. We show that this method is fast and provides significantly more accurate absolute density fluctuation reconstruction than the direct inversion technique. The presented method is utilizable for all types of beam emission diagnostics where the spatial resolution is higher than the combined smearing of the atomic physics and the observation.   

Overview and Benchtop Testing of 2D Fiber Optic Bolometer Array

A 2D fiber-optic bolometer (FOB) array is being developed for DIII-D as a plasma radiated power diagnostic for use in challenging spatial and electromagnetic (EM) noise situations. A prototype single-channel system also installed in DIII-D showed that the FOB is comparable to resistive bolometer performance and avoids EM interference by using a Fabry-Pérot resonator system to encode small temperature changes related to the incoming power. The 2D FOB array is designed to have high spatial resolution near the X-point and divertor to resolve highly localized emissions. The array consists of 8-by-8 sensors (64 sensors in total) with a tangential view into the DIII-D tokamak. Performance of the array design is evaluated using two forward-modeling tests using synthetic radiation profiles: sectional radiated power and radiation structure analyses. These tests show that the design will have spatial resolution of ~4 cm. In addition, a new technique, which adds a copper thermal sink to the sensor, is implemented to reduce the time-constant of the raw FOB measurements from ~6 seconds to ~150 ms. The diagnostic design and benchtop test results including calibrations of the FOB array will be presented.   

Pop-up divertor Langmuir Probe diagnostic in the High Heat Flux (HHF) divertor of Wendelstein 7-X (W7‑X)

The water-cooled HHF divertor, recently installed in the W7-X stellarator is equipped with 36 pop-up Langmuir probes (LP) for plasma parameter measurement starting from Operational Phase (OP) 2.1. These tungsten probes present oblique faces to the magnetic field, serving as an upgrade over the flush-mounted graphite probes used previously in W7-X. This paper focuses on the design and development of the pop-up LP diagnostic tailored for the water-cooled divertor in W7-X. Each pair of probes is connected to a "drive-coil" actuator. An upper and a lower divertor module each house 9 drive coils. By passing an appropriate current (j) through a drive coil in the magnetic field (B) of W7-X, a j×B force is applied on the coil, causing the probes to move. Such a pop-up LP concept was previously used in JET and Alcator C-Mod. Each drive coil includes a co/counterweight for passive retraction of the probes when j = 0. Before installing the drive coils in W7-X, extensive durability tests were conducted on prototypes under relevant experimental conditions (B ∼ 2.5T, ultra-high vacuum). The probe design is customized for each of the 36 probes to prevent the probe tips from posing leading edges to the flux tubes carrying high heat flux. An electronic bridge circuit is used for measurement to compensate for the effects of signal propagation time on the long cable lengths. The diagnostic system is seamlessly integrated in W7-X segment control system for an automated operation and control of the diagnostic. The system was successfully put into operation in the recent campaign and successfully measured plasma density, temperature and electric potential.   

Withdrawn

The SAMI-2 diagnostic is a synthetic aperture microwave imaging diagnostic using phased array technology to probe the edge plasma of the MAST-U tokamak based at Culham in the UK. SAMI-2 is currently in a commissioning phase, with the project aims being to create a reliable working diagnostic package, producing magnetic pitch angle measurements at the edge of the MAST-U plasma.

SAMI-2 uses a monostatic antenna array and measures the doppler shift of back-scattered radiation from the edge of the plasma. Turbulence elongated along the magnetic field lines causes the back-scattering, and as the plasma is rotating, there will be separate doppler red and blue-shifted regions. The bisection of these two regions is orthogonal to the magnetic field lines and will produce pitch angle measurements. It may be possible to use these pitch angle measurements to infer the edge current density, making SAMI-2 one of the first reliable diagnostics for edge current density measurements. This would help for defining the parameter regimes that lead to ELM events, and may provide groundwork for future control systems.

SAMI-2 uses a phased array that does not include optical focusing components. Phased array systems are used in the military for radar applications and would provide a unique solution to maintenance issues of optical components undergoing neutron irradiation. Work will be presented on the calibration methods of the diagnostic including the phased antenna array and potential initial results from the MAST-U experiment.   

Conceptual design for real-time measurement of isotope ratio and RF electric field and in the scrape-off layer

We present a conceptual design for a real-time feedback control loop utilizing experimentally measured parameters from the Dynamic Stark Effect Low Hybrid Field (DSELF) diagnostic. DSELF is a high-resolution spectroscopic diagnostic that measures the polarized D_β/H_β line profile using optical emission spectroscopy. By properly modeling the line profile, we can extract information about the plasma isotope ratio and the lower hybrid (LH) wave RF electric field vector. Least squares analysis of the spectral data is computationally expensive and takes tenths of minutes for convergence due to the large number of free parameters required in the model. To overcome this limitation, we employ neural networks that provide the necessary throughput for real-time control with enough accuracy (uncertainty of 0.5% for the isotopic ratio and 0.09 kV/cm and 0.4 kV/cm for the field radial electric and RF poloidal, respectively). We investigate with DSELF the isotope ratio (D-H), which is an important parameter for ion cyclotron resonance heating (ICRH), and discuss diagnostic applicability to D-T ratio estimation. The other quantity inferred directly from the DSELF spectrum is the lower hybrid RF electric field, which is essential to quantify the propagation of LH waves through the edge of the plasma. That information can be used to optimize and control LH launchers via actuators (i.e., impurity powder dropper, supersonic molecular beam injection, pellets, etc.) in a feedback control loop.   

real-time analysis of signals from an array of Langmuir probes in a tokamak divertor

In this study, we are supporting divertor detachment control via a spatial analysis of Langmuir probe data for implementation into tokamaks such as DIII-D and KSTAR. A reliable heat exhaust control system is needed to deduce the degree to which impurity seeding must be introduced into the system. There is a balancing act as complete detachment of plasma can cause problems including radiative collapse but too little detachment may lead to the plasma melting the walls. Langmuir probes may be a good control sensor for degree of detachment as they provide information about particle flux which is directly related to ion saturation current density (Jsat). Inspecting data from past experiments, Jsat profiles are fit with the use of Eich fits and spline fits. Then the full width at 80% maximum of the profiles are accessed and compared to probe spacing to better constrain future control diagnostics. In the fully attached regime, Jsat profiles may be very narrow, and the peak may be missed leading to measurement errors. In other cases, such as where the plasma is detaching, the Jsat profile may be much wider. A survey of Jsat profiles from past DIII-D and KSTAR experiments may guide future solutions to real time data processing to form a useful control variable.   

Plasma Density Imaging using a 2-Dimensional, Second-Harmonic, Dispersion Interferometer

The Second-Harmonic Dispersion Interferometer (SHDI) has a single, common-axis, beam path, where the primary laser is frequency doubled before, and after, the plasma, allowing the dispersive-phase change between the two-color beams to be accurately measured. The system design is robust, low-cost, and low-maintenance, providing performance that is comparable to a microwave interferometer and second-harmonic CO₂ interferometer.[1] The 2D-SHDI is constructed using a pulsed, Nd:YAG laser, reflective-beam expanders, two-digital cameras, and image-processing software, providing: >10 mRad phase change, 100-μm spatial resolution, 1 ns sampling time,100 Hz frame rate.[2] The density is calculated from two phase image measurements, one for the background (BG) and a second one for the sample; digital subtraction removes the BG even in the presence of a high noise environment.[3] This paper presents a summary of the performance measured using a pulsed-plasma jet and pulsed-gas jet. The minimum instrument sensitivity is for a line-integrated plasma density of, ∫n·dl > 10¹⁴ cm^-2 , thus a low density measurement requires longer pathlenghs. The measured-density profile has a 3.5-cm diameter.    

Quasi-optical design for the cross-polarization scattering diagnostic on the HL-2M tokamak

As the plasma beta (β) increases in high-performance tokamaks, electromagnetic turbulence becomes more significant, potentially constraining their operational range. To investigate this turbulence, a cross-polarization scattering (CPS) diagnostic system is being developed for the HL-2M tokamak, which is a new advanced tokamak with the capability of high-performance plasma. To optimize properties of the Q-band CPS diagnostic such as wavenumber resolution and signal-to-noise ratio, a quasi-optical system has been designed. The system includes a lens group for beam waist size optimization, a rotatable wire-grid polarizer for polarization adjustment, and a reflector group for matching the emission angle of the launched microwave. Laboratory tests demonstrated a beam spot size of around 2.5 cm at the target measurement location (near the plasma pedestal), cross-polarization isolation exceeding 50 dB, and poloidal and toroidal angle adjustment ranges of ±40 degrees and ±15 degrees, respectively. These results verify the system's feasibility through laboratory evaluations. The CPS diagnostic with the quasi-optical system has been installed on the HL-2M and is currently being tested in this experimental campaign.   

Surface composition of advanced W materials during high-flux plasma exposure

Increasingly, advanced tungsten materials containing dispersoid particles (e.g. TiO₂, TaC, Ni) or dopants (Re, K) are recieving consideration as plasma-facing materials for MFE. The effect of shallowly implanted D, He, and N on sputtering and surface-to-bulk transport of impurity species is not yet well understood, but nevertheless can have a significant influence on the composition of the outer-most atomic layers presented to the plasma. In addition, these new materials can have much more complex surface chemistry than pure polycrystalline tungsten. In this study, we rely on a combination surface sensitivie diagnostics, including in-situ low energy ion beam analysis, XPS, and temperature-programmed desorption to decipher these processes. We exposed W-Re, W-Ni, and W-TiO₂ specimens to beams of atomic hydrogen at thermal energies for studies of hydrogen chemisorption, as well as high-flux D₂⁺ plasmas seeded with He and N. Our measurements, obtained at temperatures ranging between 100 - 850 °C, show considerable surface-to-bulk transport mediated by He and N plasma exposure. The concentration and distribution of the implanted N was also found to depend strongly on temperature up to a depth 50 nm depth. As part of this work, we also demonstrate a novel ion time-of-flight (TOF) spectrometer for high-resolution depth profiling that allows for surface composition measurements (including D surface concentration) at elevated pressures (up to 5×10^-2 Pa) during low-flux particle bombardment.