• Energy Research
• EU close
• Nuclear Materials and Energy close
• Aalto University close

Erosion, transport and deposition of wall impurities are major concerns in future magnetic fusion devices, both from the perspective of the fusion plasma and the machine wall. An extensive study on molybdenum transport and deposition performed in the TEXTOR tokamak yielded a detailed deposition map that is ideal for benchmark deposition studies. A qualitative benchmark is attempted in this article with the ASCOT code. We set up a full 3D model of the TEXTOR tokamak and studied the influence of different physical mechanisms and their strengths on molybdenum deposition patterns on the simulated plasma-facing components: atomic processes, Coulomb collisions, scrape-off layer (SOL) profiles, source distribution, marker starting energy, radial electric field strength, SOL flow and toroidal plasma rotation. The outcome comprises 13 simulations, each with 100,000 markers. The findings are: • Toroidal plasma movement, either within the LCFS or as SOL flow, is negligible. • SOL profile and marker starting energy have modest impact on deposition. • Source distribution has a large impact in combination with radial electric field profiles. • The E⇀×B⇀ drift has the highest impact on the deposition profiles.

Toroidal and poloidal flows of injected N+ ions were measured in the high-field side (HFS) scrape-off layer (SOL) of ASDEX Upgrade by Doppler spectroscopy with different degrees of HFS divertor detachment. In high-recycling conditions, the results suggest reversed parallel N+ flow away from the inner divertor in the near SOL close to the separatrix, while the flow is towards the inner divertor throughout the SOL in detached conditions. The measured poloidal N+ flows were directed away from the HFS divertor in the near SOL for all density cases. Divertor plasma oscillations, characterized by momentary peaking of the HFS target ion flux and decrease of the HFS SOL density, were observed slightly before the roll-over of the ion saturation current to the HFS target and lead to an increase in the N+ flow towards the HFS divertor. SOLPS and ERO simulations of the experiment predict entrainment below 50% between the velocities of N+ and D+ ions, suggesting that N+ ions are quantitatively a limited proxy for measuring D+ flows. ERO simulations show significantly higher entrainment for higher ionization states, e.g., N2+ and N3+.

In this study, beryllium tiles from Joint European Torus (JET) vacuum vessel wall were analysed and compared regarding their position in the vacuum vessel and differences in the exploitation conditions during two campaigns of ITER-Like-Wall (ILW) in 2011–2012 (ILW1) and 2013–2014 (ILW2) Tritium content in beryllium samples were assessed. Two methods were used to measure tritium content in the samples – dissolution under controlled conditions and tritium thermal desorption. Prior to desorption and dissolution experiments, scanning electron microscopy and energy dispersive x-ray spectroscopy were used to study structure and chemical composition of plasma-facing-surfaces of the beryllium samples. Experimental results revealed that tritium content in the samples is in range of 2·10$^{11}$–2·10$^{13}$ tritium atoms per square centimetre of the surface area with its highest content in the samples from the outer wall of the vacuum vessel (up to 1.9·10$^{13}$ atoms/cm$^{2}$ in ILW1 campaign and 2.4·10$^{13}$ atoms/cm$^{2}$ in ILW2). The lowest content of tritium was found in the upper part of the vacuum vessel (2.0·10$^{12}$ atoms/cm$^{2}$ and 2.0·10$^{11}$ atoms/cm$^{2}$ in ILW1 and ILW2, respectively). Results obtained from scanning electron microscopy has shown that surface morphology is different within single tile, however if to compare two campaigns main tendencies remains similar.

In attaining the cold plasma condition at the divertor target via gas puffing and/or nitrogen seeding, a large variation in the pressure-momentum and cooling losses has, for the first time, been demonstrated on JET with the ITER-like wall, leveraging novel target electron temperature measurements and JET's unique capability for disentangling the main-ion-driven net dissipation effects from impurity radiation. These observations provide key insights into the dependence of volumetric losses on the dominant dissipation mechanisms, whether due to neutral-plasma interaction processes, impurity radiation induced detachment, and their interplay. While numerical code studies indicate that both volumetric cooling and pressure-momentum losses in the scrape-off layer (SOL) are essential for detachment to occur, experimental confirmation has been largely lacking, in large part due to diagnostic limitations. In the present assessment of L-mode discharges in the diagnostically optimized outer horizontal target configuration, the onset of pressure-momentum losses in nitrogen seeding scans in low recycling conditions is shown to occur at relatively high target electron temperature of 10 eV compared to pure deuterium pressure scans, leading to a more pronounced reduction in the target heat flux, and demonstrating for the first time the importance of momentum losses in impurity radiation dominated dissipative divertors in the absence of strong neutral-plasma interactions. Conversely, a relatively constant target heat flux response is observed above Te,t = 5 eV in pure deuterium density ramps, where the decrease in Te,t is balanced by a rise in target pressure prior to the ion flux rollover, coinciding with the pressure-momentum loss onset. While in practice the relative strength of each dissipation channel will be driven by integrated scenario requirements and core plasma performance constraints, this work demonstrates that there is no universal form of the fmom-loss and fcooling volumetric loss factors and their correlation. Moreover, the results confirm the theoretical and code predictions that both volumetric loss processes are required for detachment, while also resolving an apparent, but counter-intuitive, finding of the code results that volumetric power loss is insensitive to the presence of impurities, demonstrating that this is not the case experimentally. Further interpretive modelling is required to elucidate the dominant momentum and cooling loss channels in the observed trends.

The edge fluid code EDGE2D-EIRENE was used to compare the calculated low field side mid-plane separatrix temperature (Teu, sep) in the presence of seeding impurities with the two point model. The value of Teu, sep and scalings of Teu, sep with the power crossing the separatrix (Psep) are studied. Two scalings of Teu, sep with Psep can be derived from two point model; (1) Teu,sep∝Psep27 which assumes that the power decay length λq is constant and; (2) Teu,sep∝Psep49 which accounts for the λq dependence on Psep and the presence of prescribed diffusive radial transport. A linear scaling between Teu, sep and Psep is observed in the EDGE2D simulations, although the Teu,sep∝Psep49 scaling captures the variation in Teu, sep (in the simulations) for all but highly seeded simulations. This linear dependence (between Teu, sep and Psep) is due to a stronger than linear dependence on the parallel heat flux entering the divertor (q∥, u, sep) on Psep (for a doubling of Psep a factor six variation in q∥, u, sep is observed). Psep is reduced by a factor of two due to the varying impurity radiation inside the separatrix. However, q∥, u, sep reduces more than a factor two because (i) impurity radiation preferentially removes heat flux above the x-point and near the separatrix and (ii) the variation in λq with Psep due to increasing radial diffusive heat flux (Eq. (5.77) Stangeby 2000). The largely varying q∥, u, sep, interestingly, is successfully captured by the power decay length (λq, Eich) calculated by fitting to the target heat flux density. Accounting for the q∥, u, sep variation by using λq, Eich and Psep (both experimentally measurable quantities), an agreement between the two point model equation and the predicted Teu, sep from EDGE2D-EIRENE was obtained. A variation of Teu, sep from approximately 60 eV to 120 eV, for electron separatrix density range of 2−3×1019m−3, was observed. This separatrix temperature variation from EDGE2D is in contrast to a routinely assumed separatrix temperature of 100 eV used for pedestal stability analysis at JET. Keywords: Separatrix temperature, EDGE2D-EIRENE

In this paper, we report experimental and numerical investigations of gross and net erosion of gold (Au) and molybdenum (Mo), proxies for the common plasma-facing material tungsten (W), during L-mode plasma discharges in deuterium (D) in the outer strike-point region of the ASDEX Upgrade tokamak. To this end, erosion profiles of different marker spots (for Au, dimensions 1 × 1 and 5 × 5 mm2) and marker coatings (for Mo) have been determined and modelled using the ERO code. The smaller marker spots were designed to quantify the gross-erosion rate while on the bigger markers local prompt re-deposition of Au allowed obtaining data on net erosion. The experimental results indicate relatively uniform erosion profiles across the marker spots or coatings, very little re-deposition elsewhere, and the largest erosion taking place close to the strike point. Compared to W, the markers show up to 15 times higher net erosion but no major differences in the poloidal migration lengths of Au and W can be seen. Gold thus appears to be a proper choice for studying migration of W in the divertor region. The ERO simulations with different background plasmas are able to reproduce the main features of the experimental net erosion profile of Au. Of the studied parameters, electron temperature has the strongest impact on erosion: doubling the temperature enhances erosion by a factor of 2.5–3. In contrast, for Mo, the simulated net erosion is ~ 3 times smaller than what experimental data indicate. The discrepancies can be attributed to the deviations of the background plasma profiles from the measured ones as well as to the applied models or approximations for the ion temperature, plasma potential, and sheath characteristics in ERO. In addition, the surrounding areas of the marker samples being covered with impurities and W from previous experiments may have considerably reduced the actual re-deposition of Mo. All the simulations predict a toroidal tail of re-deposited particles, downstream of the markers, but the particle density seems to be below the experimental detection threshold. The comparison between the 1 × 1 mm2 and 5 × 5 mm2 marker spots further reveal that re-deposition drops from >50% to <40% when decreasing the marker size. This indicates that small enough marker samples can be used for accurately determining gross erosion in ASDEX Upgrade.