Session UO4: PMI & Boundary

Multi-physics modeling of plasma-material interactions

Plasma-material interactions (PMI) can degrade both plasma and material properties. Often, PMI modeling focuses on either the plasma or surface. Here, we present an integrated model with high-fidelity codes coupled within the IPS framework that self-consistently addresses PMI. The model includes, calculation of spatially resolved influx of plasma and impurities to the surface and their implantation; surface erosion and roughening; evolution of implanted species and sub-surface composition; and transport of eroded particles across the plasma and their re-deposition. The model is applied and successfully compared to dedicated PISCES linear device experiments, where a tungsten (W) target was exposed to helium (He) plasma. The present contribution will focus on the analysis of W erosion, He retention and sub-surface gas bubble and surface composition evolution, under the different He plasma conditions across the surface that are calculated by impurity transport modeling. Impact of code coupling, reflected as interplay between surface erosion, fuel / impurity implantation and retention, and evolution of target composition, as well as sensitivity of these processes to plasma exposure conditions is also analyzed in detail.   

GITR Simulation of Helium Exposed Tungsten Erosion and Redistribution in PISCES-A

The extreme heat, charged particle, and neutron flux / fluence to plasma facing materials in magnetically confined fusion devices has motivated research to understand, predict, and mitigate the associated detrimental effects. Of relevance to the ITER divertor is the helium interaction with the tungsten divertor, the resulting erosion and migration of impurities. ~ The linear plasma device PISCES A [1] has performed dedicated experiments for high (4x10\textasciicircum 22 m-2s-1) and low (4x10\textasciicircum 21 m-2s-1) flux,~ 250 eV He exposed tungsten targets to assess the net and gross erosion of tungsten and volumetric transport. The temperature of the target was held between 400 and 600 degrees C. ~ We present results of the erosion / migration / re-deposition of W during the experiment from the GITR (Global Impurity Transport) code coupled to materials response models. In particular, the modeled and experimental W I emission spectroscopy data for the 429.4 nm wavelength and net erosion through target and collector mass difference measurements are compared. Overall, the predictions are in good agreement with experiments. ~ [1] R.P. Doerner Nucl. Fusion \textbf{52 }(2012)   

Fast, Statistical Model of Surface Roughness for Ion-Solid Interaction Simulations and Efficient Code Coupling

Surface roughness greatly impacts material erosion, and thus plays an important role in Plasma-Surface Interactions. Developing strategies for efficiently introducing rough surfaces into ion-solid interaction codes will be an important step towards whole-device modeling of plasma devices and future fusion reactors such as ITER. Fractal TRIDYN (F-TRIDYN) is an upgraded version of the Monte Carlo, BCA program TRIDYN developed for this purpose that includes an explicit fractal model of surface roughness and extended input and output options for file-based code coupling. Code coupling with both plasma and material codes has been achieved and allows for multi-scale, whole-device modeling of plasma experiments. These code coupling results will be presented. F-TRIDYN has been further upgraded with an alternative, statistical model of surface roughness. The statistical model is significantly faster than and compares favorably to the fractal model. Additionally, the statistical model compares well to alternative computational surface roughness models and experiments. Theoretical links between the fractal and statistical models are made, and further connections to experimental measurements of surface roughness are explored.   

Calculations of Helium Bubble Evolution in the PISCES Experiments with Cluster Dynamics

Plasma surface interactions in fusion tokamak reactors involve an inherently multiscale, highly non-equilibrium set of phenomena, for which current models are inadequate to predict the divertor response to and feedback on the plasma. In this presentation, we describe the latest code developments of Xolotl, a spatially-dependent reaction diffusion cluster dynamics code to simulate the divertor surface response to fusion-relevant plasma exposure. Xolotl is part of a code-coupling effort to model both plasma and material simultaneously; the first benchmark for this effort is the series of PISCES linear device experiments. We will discuss the processes leading to surface morphology changes, which further affect erosion, as well as how Xolotl has been updated in order to communicate with other codes. Furthermore, we will show results of the sub-surface evolution of helium bubbles in tungsten as well as the material surface displacement under these conditions.   

Modeling of Fuzz Formation on Helium-Ion-Irradiated Tungsten Surfaces

Experiments have shown that helium (He) from plasma devices is responsible for the formation of a nanostructure with a fuzz-like morphology on the plasma-facing tungsten (W) surface after a few hours of plasma exposure, which can potentially impact fusion reactor performance. We report an atomistically-informed, continuous-domain model capable of describing the surface morphological evolution of He-ion-irradiated W and predicting the initial stage of fuzz formation on W surfaces. Based on this model, a systematic protocol of self-consistent dynamical simulations of the irradiated tungsten surface morphological evolution is conducted to compare the simulation results with experimental studies in the literature. Upon model validation, the simulations are used to identify the critical range of conditions for nanotendril formation on the surface, a precursor to fuzz-like surface growth. We examine a broad range of surface temperature, He ion energy, and He flux values relevant to experimental conditions and present the results of a sensitivity analysis of the key model parameters, such as He concentration and He nanobubble size. Further development of the model, driven by comparisons of its predictions with experimental observations also will be discussed.   

Simulating tokamak PFC performance using simultaneous dual beam particle loading with pulsed heat loading

The performance of plasma facing components (PFCs) in a fusion device are expected to change due to high flux particle loading during operation. Tungsten (W) is a promising PFC candidate material, due to its high melting point, high thermal conductivity, and low tritium retention. However, ion irradiation of D and He have each shown to diminish the thermal strength of W. This work investigates the synergistic effect between ion species, using dual beam irradiation, on the thermal response of W during ELM-like pulsed heat loading. Experiments studied three different loading conditions: laser, laser $+$ He$^{\mathrm{+}}$, and laser $+$ He$^{\mathrm{+}} \quad +$ D$^{\mathrm{+}}$. 100 eV He$^{\mathrm{+}}$ and D$^{\mathrm{+}}$ exposures used a flux of 3.0-3.5 x 10$^{\mathrm{20}}$ m$^{\mathrm{-2}}$ s$^{\mathrm{-1}}$. ELM-like loading was applied using a pulsed Nd:YAG laser at an energy density of 0.38-1.51 MJ m$^{\mathrm{-2}}$ (3600 1 ms pulses at 1 Hz). SEM imaging revealed that laser $+$ He$^{\mathrm{+}}$ loading at 0.76 MJ m$^{\mathrm{-2}}$ caused surface melting, inhibiting fuzz formation. Increasing the laser fluence decreased grain size and increased surface pore density. Thermally-enhanced migration of trapped gases appear to reflect resultant molten morphology.   

3D measurements and simulations of ion and neutral velocity distribution functions in a magnetized plasma boundary

We present progress toward the first paired 3D laser induced fluorescence measurements of ion and neutral velocity distribution functions (I/NVDFs) in a magnetized plasma boundary. These measurements are performed in the presheath region of an absorbing boundary immersed in a background magnetic field that is obliquely incident to the boundary surface ($\psi = 74^{\circ}$). Parallel and perpendicular flow measurements demonstrate that cross-field ion flows occur and that ions within several gyro-radii of the surface are accelerated in the $\vec{E} \times \vec{B}$ direction. We present electrostatic probe measurements of electron temperature, plasma density, and electric potential in the same region. Ion, neutral and electron measurements are compared to Boltzmann simulations, allowing direct comparison between measured and theoretical distribution functions in the boundary region.   

Validation of Boltzmann-Poisson Continuum Code with LIF measurements of Plasma Sheath in an Oblique Magnetic Field

Here we present the first fully three-dimensional validation of a 1D3V Boltzmann-Poisson continuum solver against 3D LIF measurements of ion and neutral velocity distribution functions taken in a magnetized plasma sheath. The multi-species full-f plasma model is solved with finite volumes in the phase-space and computes the velocity distribution functions of plasma species, facilitating a direct comparison to LIF data in the magnetic presheath. LIF measurements were taken near an absorbing boundary with a magnetic field obliquely incident to the surface. The plasma model incorporates ionization and charge exchange through a BGK collision operator, with reaction rates computed directly through convolution with the distribution functions. Results clearly display the 3D structure of the magnetized sheath, including acceleration along the ExB direction. LIF measurement work was supported by U.S. National Science Foundation Grant No. PHY-1360278.   

Alternative power exhaust studies in an advanced upper divertor in ASDEX Upgrade supported by SOLPS and EMC3-EIRENE simulations

In order to study alternative divertor configurations, currently discussed as a possible solution for the power exhaust problem in a fusion reactor, the installation of a pair of in-vessel poloidal field coils in the upper divertor of ASDEX Upgrade was recently decided. Besides the conventional single- and double null configurations, a series of new configurations ranging from an X- divertor, to a low- (LFS SF$^-$) and finally a high field side snowflake minus will be possible with these coils in a machine with a high $P/R$ ratio. The arangement of these coils was based on the pioneering work of TCV as well as simulations with EMC3-EIRENE, which can rather easily handle topologies with two X-points and which identified a series of heat flux mitigation effects. Due to the lack of drifts and volumetric recombination in the code, however, a clear prediction on the detachment degree and threshold is missing as well as a realistic description of the in-out divertor asymmetries. This limit has now been overcome by creating an adequate computational grid for a LFS SF$^-$ configuration for SOLPS. In this contribution we will present the worldwide first simulation on this grid as well as the upgrade plans and discuss the potential different heat flux mitigation mechanisms.   

The role of pumping on particle removal and divertor plasma conditions

The effect of pumping location in a closed detached divertor configuration is examined with SOLPS modeling. A closed divertor can increase neutral pressure and enhance radiative dissipation, thus it is proposed for advanced tokamak operation in order to achieve detachment at as low an upstream plasma density as possible. However, the necessity to pump the closed divertor results in reduction of the high density of neutrals. By changing the recycling rate at the pump, it is confirmed that the pump location has a great impact on the effective pumping speed, which influences the divertor plasma significantly. Higher pumping speed reduces the neutral density and increases Te as well as the heat flux to the target. For a given particle removal rate, however, the plasma conditions are insensitive to the pump location within the divertor. High D2 gas puffing with high pumping could help to achieve detachment only when the upstream density is increased by puffing, in contrast, a deeper detachment can be easily reached in a low pumping and low puffing case.   

Physics of neutral gas jet interaction with magnetized plasmas

It is critical to understand the physics and transport dynamics during the plasma fuelling process. Plasma and neutral interactions involve the transfer of charge, momentum, and energy in ion-neutral and electron-neutral collisions. Thus, a seven field fluid model of neutral gas jet injection (NGJI) is obtained, which couples plasma density, heat, and momentum transport equations together with neutrals density and momentum transport equations of both molecules and atoms. Transport dynamics of plasma and neutrals are simulated for a complete range of discharge times, including steady state before NGJI, transport during NGJI, and relaxation after NGJI. With the \textit{trans-neut} module of BOUT$++$ code, the simulations of mean profile variations and fueling depths during fueling have been benchmarked well with other codes and also validated with HL-2A experiment results. Both fast component (FC) and slow component (SC) of NGJI are simulated and validated with the HL-2A experimental measurements. The plasma blocking effect on the FC penetration is also simulated and validated well with the experiment.   

Impact of the impurity seeding for divertor protection on the performance of fusion reactors

A 0D divertor and scrape-off layer (SOL) model has been coupled to the 1.5D core transport code ASTRA. The resulting numerical tool has been employed for various parameter scans in order to identify the most convenient choices for the operation of electricity producing fusion devices with seeded impurities for the divertor protection. In particular, the repercussions of such radiative species on the main plasma through the fuel dilution have been taken into account. The main result we found is that, when the limits on the maximum tolerable divertor heat flux are enforced, the curves at constant electrical power output are closed on themselves in the R-BT plane, i.e. no improvement would descend from a further increase of R or BT once the maximum has been reached. This occurrence appears as an intrinsic physical limit for all devices where a radiative SOL is needed to deal with the power exhaust. Furthermore, the relative importance of the different power loss channels (e.g. hydrogen radiation, charge exchange, perpendicular transport and impurity radiation), through which the power entering the SOL is dissipated before reaching the target plate, is investigated with our model.   

Single Null Negative Triangularity Tokamak for Power Handling

Power and particle control in fusion reactor is challenge and we proposed the negative triangularity tokamak (NTT) to eliminate ELM by operating L-mode edge with improved core confinement [1-3]. The SN configuration has more flexibility in shaping by adopting rectangular-shaped TF coils. The limiting normalized beta is 3.56 with wall stabilization and 3.14 without wall [3]. The vertical stability is assured under a reasonable control system. The wetted area on the divertor plates becomes wider in proportion to the larger major radius at the divertor strike points due to the NT configuration. In addition to the major-radius effect, the ``Flux Tune Expansion (FTE)'' [4] is adopted to further reduce the heat load on the divertor plate by factor of \textasciitilde 2.6 with a coil current 3 MA. L-mode edge also allows further increase in wetted area. The fusion power of 3 GW is deliverable only at normalized beta 2.1. Therefore this reactor may be operable stably against the serious MHD activities. The CD power for SS operation is \textasciitilde 175 MW at Q $=$ 17. AC operation is also possible option. A required HH factor is relatively modest $H=$1.12. [1] M.~Kikuchi et al., 1st Int. e-Conf. Energies 2014, e002, [2] S. Medvedev,M.Kikuchi,et al.,Nucl. Fus. \textbf{55}, 063013 (2015) [3] S. Medvedev, M. Kikuchi, et al., 26$^{\mathrm{th}}$ IAEA-FEC, ICC/P3-47 (2016), [4] T. Takizuka, et~al., J. Nucl. Mater. \textbf{463}, 1229 (2015).