Assessing the performance of tungsten plasma-facing components under disruption-relevant heat loads re-quires objective and reproducible methods to quantify surface damage in post-exposure microscopy images. In this work, we develop an automated framework for detecting and quantifying damage in scanning electron microscopy images of tungsten samples exposed at the OLMAT facility at CIEMAT, where leading-edge heat loads of up to 1200 MW/m2 are reproduced.The methodology follows three phases. A classical computer vision pipeline (TungstenAnalyzer) was first constructed, incorporating a dynamic topographic edge scanner, a multichannel Watershed engine, and a Frangi-Blackhat crack detector with dual hysteresis thresholding. A genetic multi objective optimisation algorithm (NSGA-II via Optuna) was then applied raising IoU over 261 generations. Recognising the intrinsic ceiling of handcrafted features, the third phase introduced a U-Net with a ResNet34 encoder pre-trained on ImageNet, augmented with scSE attention gates, deep supervision, and a crack-aware combined loss (BCE + Dice + Focal). Despite being trained on only 99 annotated SEM images, the model achieved an IoU of 0.839 for crater-like damage and 0.818 for cracks, corresponding to F1 scores of 0.912 and 0.688, respectively. The resulting multiscale inference workflow extracts physically meaningful quantities, including damaged area, crater severity and crack density, using automatic scale calibration from the SEM images.The initial analysis performed shows, as expected, that damage is systematically more severe at sample corners and leading edges than on equivalent flat regions, consistent with enhanced thermal loading due to local geometry. The complete workflow is integrated into a graphical user interface, providing a reproducible tool for routine quantification of high-heat-flux damage in tungsten plasma-facing materials for future works.

This study reports in-situ laser absorptance values at 1070 nm for solid tungsten samples, as well as liquid tin and lithium targets based on the tungsten capillary porous system technology. The effective absorptance has been estimated by matching transient thermal simulations based on the finite-element method to experimental temperature measurements at the OLMAT (Optimization of Liquid Metal Advanced Targets) facility. The targets have been characterized post-mortem by cross-sectional SEM-EDS, and the resulting microstructural and compositional data has been used to develop the numerical models. The resulting effective absorptance values have been obtained for different laser power levels and pulse lengths. In most cases, the effective absorptance decreases with increasing laser power and temperature, revealing significant discrepancies compared to reference values or pre-/post-exposure measurements, underlining the influence and complexities of the in-situ estimates. The results also indicate that the laser coupling coefficient is not constant under the tested conditions, which may be attributed to physical factors, such as surface state, wetting quality, and oxidation, as well as inherent modelling assumptions. Furthermore, we report the first dedicated effective absorptance values for liquid lithium, highlighting the critical role of the formation of passive oxide films. Quantifying this absorptance is essential to ensure that laboratory-scale energy deposition accurately reflects the thermal impact of real off-normal events. These results offer a consistent dataset for benchmarking the thermal response of liquid metal targets under laser-simulated fusion-relevant heat loads.

First experiments are reported of the simultaneous exposure of a number of Sn-wetted W CPSs and a reference W CPS to 100 ms NBI pulses (divertor steady-state loading conditions) and 2 ms long high-energy laser pulses (divertor ELM like loading conditions) at the High-Heat Flux OLMAT facility. The use of a fast-frame imaging camera allows monitoring the onset of particle ejection from the targets during laser pulses and obtaining the corresponding laser heat fluxes as a measure of the resilience of these targets. Fast camera images are used also to determine ejected particle numbers and to estimate their maximum velocities as laser power is increased in order to compare the influence of W CPS structure on these parameters. In addition, the craters resulting from particle ejection are studied for each target with an optical microscope and a scanning electron microscope. Moreover, in-situ W and Sn particle ejection is followed using visible emission spectroscopy and post-exposure W melting after particle ejection is observed using the energy dispersive X-ray method EDX for all the studied targets. This shows that Sn is unable to protect the underlying W substrate from high-energy laser damage, albeit a subsequent refilling of the formed craters with Sn is visible during NBI-only pulses after laser damage. Thus, it is considered that optimization of surface refilling/replenishment with Sn is needed to improve the W substrate protection. From this work, it is also found that the W CPS reference material has a higher laser heat flux threshold for particle ejection than the Sn-wetted targets. Nevertheless, it is important to take into account that in these experiments with laser pulses, the possible beneficial effects of vapor shielding that can take place during particle irradiation at ELMs or disruptions are not present, thus these experiments represent a worst-case scenario.

The armor for future nuclear fusion reactors is one of the main areas of research due to the harsh conditions it will undergo. Thermal fatigue is one of the most serious damage, as it will cause any material to fail even if the heat and particle loads during the reactor operation are always maintained low. In this work we have compared the actual tungsten armor for ITER tokamak with a new advanced tungsten material: tungsten reinforced by tungsten fibers (Wf/W). ITER-like W has shown small intergranular cracking at heat loads similar to the ones found in other similar devices: heat flux factor of FHF = 5.2 ± 1.6 MW/m2s0.5. But at much lower number of pulses: 641 versus 105. H embrittlement by the high-energy ions of OLMAT has been postulated as one of the main reason of this relatively prompt cracking appearance. Opposed to this, the type of Wf/W studied here, Porous Matrix (PM-Wf/W), has shown no damage at the same conditions and up to 950 pulses. These results show the capabilities of OLMAT for fatigue studies in conditions relevant to a future nuclear fusion reactor. New upgrades of OLMAT will partially solve the issues found in this first phase. Future work to continue with fatigue studies will be addressed.

We provide an overview of activities carried out at the TJ-II stellarator for improving our understanding of- and developing plasma physics models for particle density profiles in stellarators. Namely, we report on recent progress in turbulent particle transport simulation, validation of pellet deposition models, density profile shaping for performance control and new experimental techniques for edge turbulence and plasma-neutral interaction.

The Secondary Electron Emission (SEE) from the Plasma Facing Components, (PFCs) affects the plasma sheath and edge structure of magnetically confined toroidal plasmas with effects on the Scrappe Off Layer (SOL), heat transport and incident plasma heat flux to the divertor targets. It is also behind the performance degradation in aerospace devices as Hall thrusters. In future reactor prototypes using tungsten (W) components, much more demanding and longer term conditions for the exposed PFCs are expected. These scenarios will exacerbate the material degradation of the exposed surfaces that will change its morphology, thus modifying its plasma material interaction behavior. In such scenario, specific studies on the influence of the microstructure in the tungsten SEE yields become necessary. In these laboratory experiments, the SEE emission of different microstructured tungsten coatings exposed to helium Glow Discharge (GD) plasmas has been analyzed by using a previously developed laboratory technique. It enables the characterization of the I-V characteristics of the biased sample and, at the same time, the acquisition of the electron incident flux by using a gridded probe adjacent to the exposed sample. Different microstructured W coated samples differing in coating thickness (2500 nm and 500 nm) and primal substrate surface finishing (translated in final differences within the W coating topologies) were analyzed, also including cold rolled tungsten and original stainless steel (SS) substrate (material in which the microstructured tungsten film was deposited) as benchmarks for comparison. The overall results have shown that the presence of a 2500 nm thick microstructured W coating decreases the SEE yield of tungsten (cold rolled) up to a factor 40 % at electron mean energies of 100–175 eV. Conversely, 500 nm coatings did not reduce the SEE yield even showing an increase at 25–100 eV. The final increase/decrease in the obtained SEE yields seems to be more influenced by the microstructured coating thickness (where the role of intrinsic differences in oxygen content of the surfaces is discussed) rather than other questions possibly derived by the differences in surface topology (ordering, directionality and/or roughness of the microstructure features).

The OLMAT (Optimization of Liquid Metal Advanced Targets) facility has recently undergone the commissioning and start-up phases. Solid Titanium -Zirconium-Molybdenum (TZM) alloy and liquid tin (Sn) metallic targets were exposed to a hydrogen neutral beam injector (NBI) particle flux with power densities up to 58 +/- 14 MW/ m2, pulse duration up to 150 ms, and repetition rates up to 2 pulses/minute. These beam parameters are well above the estimates based on the typical performance of this NBI system when used for heating plasmas in the TJ-II stellarator. The parameters of the plasma generated through the interaction of the fast (32.5 keV) neutrals and ions and the solid were characterized by spectroscopic methods while surface temperature and total absorbed power were followed using pyrometry, infrared (IR) thermography, and calorimetry, respectively. Targets were visually monitored during the exposure and microscopically analyzed ex-situ. Electrical isolation of the target permitted recording the floating voltage during irradiation as well as for active biasing tests. In this work, a description of the facility, its operating parameters, and firsts results are provided and assessed as a new High Heat Flux (HHF) Facility for testing solid and liquid metal divertor targets under reactor-relevant heat load conditions.

The operation of the Optimization of Liquid Metal Advanced Targets (OLMAT) facility began in April 2021 with the scientific objective of exposing liquid-metal plasma facing components (PFCs) to the particle and power fluxes provided by one of the hydrogen neutral beam injectors of the TJ-II stellarator. The system can deliver heat fluxes from 5 to 58 MW m −2 of high energy hydrogen neutral particles (≤ 33 keV) with fluxes up to 10 22 m 2 s −1 (containing an ion fraction ≤ 33% in some instances), pulsed operation of 30–150 ms duration and repetition rates up to 2 min −1 . These characteristics enable OLMAT as a high heat flux (HHF) facility for PFC evaluation in terms of power exhaust capabilities, thermal fatigue and resilience to material damage. Additionally, the facility is equipped with a wide range of diagnostics that includes tools for analyzing the thermal response of the targets as well as for monitoring atomic/plasma physics phenomena. These include spectroscopy, pyrometry, electrical probing and visualization (fast and IR cameras) units. Such particularities make OLMAT a unique installation that can combine pure technological PFC research with the investigation of physical phenomena such as vapor shielding, thermal sputtering, the formation/characterization of plasma plumes with significant content of evaporated metal and the detection of impurities in front of the studied targets. Additionally, a myriad of surface characterization techniques as SEM/EDX for material characterization of the exposed PFC prototypes are available at CIEMAT. In this article, first we provide an overview of the current facility upgrade in which a high-power CW laser, that can be operated in continuous and pulsed modes (0.2–10 ms), dump and electrical (single Langmuir) probe embedded on the target surface have been installed. This laser operation will allow simulating more relevant heat loading scenarios such as nominal steady-state divertor heat fluxes (10–20 MW m −2 in continuous mode) and transients including ELM loading and disruption-like events (ms time scales and power densities up to GW m −2 range). The work later focuses on the more recent experimentation (2022 fall campaign) where a 3D printed Tungsten (W) Capillary Porous System (CPS) target, with approximated 30 μm pore size and a 37% porosity and filled with liquid tin. This porous surface was a mock-up of the PFC investigated in the ASDEX Upgrade divertor manipulator. The target composed with this element was eventually exposed to a sequence of shots with the maximum heat flux that OLMAT provides (58 ± 14 MWm −2 ). Key questions as resilience to dry-out and particle ejection of the liquid metal layer, its refilling, the induced damage/modification of the porous W matrix and the global performance of the component are addressed, attempting to shed light on the issues encountered with the PFC at tokamak scale testing.

Four different tin-wetted, tungsten CPS (Capillary Porous System) targets where exposed to NBI pulses in the OLMAT High Heat flux (HHF) facility. They include two flexible ones placed on a TZM support (W meshes and W felt) and two compact ones (sintered W disk and 3D printed W). A comparative study was performed using a fastframe imaging camera and an infrared pyrometer. Surface temperature increase and homogeneity, particle ejection, CPS damage and overall behaviour were studied for each case. Sn drop/accumulation at the lower part of the targets was observed for all cases when they are heated up to around 400 C except for the 3D printed W target that has its own Sn deposit. The 3D printed W target presented the best results in all aspects, withstanding heat pulses up to 58 MW/m2 in 100 ms without any damage or particle ejection. On the other hand, the W mesh targets displayed damage at 20 MW/m2 due to a bad thermal contact with the deposit, while the sintered W disk developed a crack during a series of 15 MW/m2 NBI pulses. As might be expected, a reduced increase of temperature during pulses is observed for the two compact W targets. The results and their relevance for the design of a Sn wetted W CPS for application as a DEMO-divertor material are discussed in the present work.

OLMAT (Optimization of Liquid Metal Advanced Targets) is a new High Heat Flux (HHF) Facility for testing solid and liquid metallic targets at DEMO-relevant power densities that has been installed, commissioned and first operated at the National Fusion Laboratory in Madrid in 2021 [1] . A Neutral Beam Injector (NBI) is used as a high-power source providing 100 ms pulses at maximum repetition rate of 1p/30s and delivering power densities at the target (Mo, W, liquid Sn) above 50 MW/m 2 [2] . A calibrated survey spectrometer covering the UV-visible-near IR ranges provided information about the impinging hydrogenic species, plasma composition and its microscopic parameters, while a Balmer Ha (656.3 nm) monitor, a 16-Channel PMT array and a fast-frame imaging camera were used for the characterization of particle recycling and the plume’s spatial structure. Moreover, the floating voltage and saturation current developed at the target was monitored on a shot-to-shot basis. In addition to the above, nitrogen injection was performed through a dedicated piezoelectric valve for active spectroscopy. Plasma temperatures between 1 and 2 eV were deduced from the ratios of H Balmer lines, in line with the recorded values of floating potential, but higher, inconsistent values were obtained from the ratio of molecular N 2 and N 2 + bands. The facility and the plasmas therein generated will be described.

TJ-II stellarator results on modelling and validation of plasma flow asymmetries due to on-surface potential variations, plasma fuelling physics, Alfvén eigenmodes (AEs) control and stability, the interplay between turbulence and neoclassical (NC) mechanisms and liquid metals are reported. Regarding the validation of the neoclassically predicted potential asymmetries, its impact on the radial electric field along the flux surface has been successfully validated against Doppler reflectometry measurements. Research on the physics and modelling of plasma core fuelling with pellets and tracer encapsulated solid pellet injection has shown that, although post-injection particle radial redistributions can be understood qualitatively from NC mechanisms, turbulence and fluctuations are strongly affected during the ablation process. Advanced analysis tools based on transfer entropy have shown that radial electric fields do not only affect the radial turbulence correlation length but are also capable of reducing the propagation of turbulence from the edge into the scrape-off layer. Direct experimental observation of long range correlated structures show that zonal flow structures are ubiquitous in the whole plasma cross-section in the TJ-II stellarator. Alfvénic activity control strategies using ECRH and ECCD as well as the relation between zonal structures and AEs are reported. Finally, the behaviour of liquid metals exposed to hot and cold plasmas in a capillary porous system container was investigated.

Electron-induced secondary electron emission (SEE) on a liquid lithium surface confined in a capillary porous system (CPS) and exposed to a plasma is reported for the first time. The liquid lithium surface is exposed to the bombardment of a suprathermal electron flux with energies up 150 eV created in a low pressure He Dc-Glow discharge. Various methods of surface oxidation have been used to reproduce realistic conditions in fusion plasma experiments. Exposure to very low residual gas pressures (around 3 × 10−7 Torr) is enough to increase the maximum of the SEE to values of about 2. The formation and dissolution of a surface oxide film is strongly temperature dependent and plays an important role in the resulting SEE yield and its time evolution: values have been found to range from ~0.8 for clean surfaces to ~2 for samples that have been heavily oxidized. In the case of O2 molecule exposure, a clear difference with temperature was observed. The molecular exposure at 330 °C had a much stronger effect on the increase in SEE than the molecular exposure at 220 °C. These results have a direct impact on the development of lithium-based divertor targets in fusion as well as in the understanding of the SEE characteristics of contaminated liquid surfaces, which has never been reported before.

In the present work, the results of the exposure of in laboratory manufactured SnLi and Sn CPS targets to Deuterium plasmas with a flux of 4 x 10(19) D m(-2)s for different fluences and temperatures are presented. For both SnLi and Sn CPS targets a saturation of the deuterium retention is observed for fluences over 1.2 x 10(22) D m(-2) in liquid state, while no saturation is observed up to 1.9 x 10(23) D m(-2) for the solid state. In both cases a decreasing Deuterium retention with increasing temperature is observed and a 100 times higher retention for SnLi than for Sn is obtained. The SnLi CPS target presents a preferential evaporation of Li, showing an almost complete depletion of the Li content after 100 min heating at 770 degrees C. Bubble formation and bursting was also visually observed and recorded during plasma exposure for the case of the SnLi CPS target but not in the case of the Sn CPS target for the same conditions. This behaviour for Sn is different from what has been observed in previous works but the system geometry and surface type, as well as the difference in the flux to the target, could explain this difference. The results stress the relevance of the design of the containing system in the D retention and Sn behaviour under D exposure.

The use of liquid metal as an alternative to cover the plasma-exposed areas of fusion reactors has called for the development of substrates where refilling and metal spreading occur readily and at reasonably low temperatures. In the search for common materials for this purpose, we show that nanostructured tungsten coatings deposited on stainless steel (SS) by magnetron sputtering at oblique angles (MS-OAD) is a good option, provided that the surface microstructure of substrate is properly engineered. Tungsten thin films with nominal thicknesses of 500 and 2500 nm were deposited onto SS plates subjected to conventional surface finishing treatments (sand blasting, sand paper abrasion and electrochemical polishing) to modify the surface topography and induce the appearance of different groove patterns. In the first part of this work we show how the topographical features of the SS substrates affect the typical nanocolumnar microstructure of OAD thin films of tungsten. Subsequently, we characterize the spreading behavior of liquid lithium onto these tungsten nanocolumnar surfaces and critically discuss whether nanocolumnar tungsten thin films are a suitable option for the wetting and spreading of molten lithium. As a result, we reveal that the features of the tungsten nanocolumnar coating, characterized by a given height and void spaces between nanocolumns in the order of 1–2 μm, is critical for the spreading of molten lithium, while the existence of wider channels affects it very weakly. Moreover, it is shown that tungsten films deposited by MS-OAD on SS substrates subjected to conventional finishing procedures represent a good alternative to other more complex surface engineering procedures utilized for this purpose.

Fusion experiments use lithium and boron getters to capture impurities and control the density of fuel species in the plasma. In the TJ-II Stellarator, the addition of a boron-carbon substrate was found to extend the lifetime of the very reactive lithium wall conditioning. However, the synergies between the lithium layer and the underlying boron-carbon, or the effect of a glow discharge cleaning, on oxygen and deuterium gettering are not fully understood. Laboratory experiments were therefore initiated to gain deeper insight on the getter properties of B(C)/Li walls. The coatings were exposed to oxygen gas and plasmas while the capture and release of O-2 was followed by mass spectrometry. Without the boron substrate, the maximum oxygen uptake is half the number of lithium atoms in the film for both oxygen gas and oxygen plasma. Conversely, the behaviour of B(C)/Li walls depends on the nature of the oxidation process: molecular oxygen (O:Li approximate to 0.15) or plasma (O:Li approximate to 0.50). The oxygen plasma gettering is thus preserved against molecular oxidation, e.g. an overnight exposure to the residual gas. While oxygen is known to promote hydrogen retention in lithium, the oxygen content of a B(C)/Li coating shows little effect on deuterium retention. Unlike simple lithium, the oxygen and deuterium gettering of B(C)/Li recovers after a helium glow discharge treatment. These advantageous features clearly point to a change in chemistry, with complex interactions between film constituents and the oxygen-rich environment of magnetic fusion devices.

In a future fusion reactor like DEMOnstration reactor (DEMO) one of the main concerns is the handling of the power exhaust from the plasma, especially at the divertor. The expected power loads cannot easily be handled by traditional armor solutions based on solid materials like tungsten, especially when the effect of intense neutron bombardment is also considered. Interest in armor concepts based on liquid metals has been subsequently on the rise, as they prove to be more resilient against high, fast power loads and neutron bombardment. However, engineering solutions for those concepts are very complex, and need to be tested. For this purpose, Optimization of Liquid Metal Advanced Targets project (OLMAT) has been envisaged. The project will use the Neutral Beam Injection of the TJ-II stellarator to irradiate liquid metal targets with power densities (neutrals plus occasionally ions) relevant to DEMO steady state operation, in the range of 20 MW/m 2 . OLMAT design will allow a series of experiments that other divertor simulator devices cannot easily perform: in-situ measurements of hydrogen retention, redeposition, vapor shielding, material fatigue, dust and precipitates effects, etc. Moreover, a high-power fiber laser will be used to simulate Edge Localized Modes in a small area, or to simulate the strike point power deposition profile.

The main results obtained in the TJ-II stellarator in the last two years are reported. The most important topics investigated have been modelling and validation of impurity transport, validation of gyrokinetic simulations, turbulence characterisation, effect of magnetic configuration on transport, fuelling with pellet injection, fast particles and liquid metal plasma facing components. As regards impurity transport research, a number of working lines exploring several recently discovered effects have been developed: the effect of tangential drifts on stellarator neoclassical transport, the impurity flux driven by electric fields tangent to magnetic surfaces and attempts of experimental validation with Doppler reflectometry of the variation of the radial electric field on the flux surface. Concerning gyrokinetic simulations, two validation activities have been performed, the comparison with measurements of zonal flow relaxation in pellet-induced fast transients and the comparison with experimental poloidal variation of fluctuations amplitude. The impact of radial electric fields on turbulence spreading in the edge and scrape-off layer has been also experimentally characterized using a 2D Langmuir probe array. Another remarkable piece of work has been the investigation of the radial propagation of small temperature perturbations using transfer entropy. Research on the physics and modelling of plasma core fuelling with pellet and tracer-encapsulated solid-pellet injection has produced also relevant results. Neutral beam injection driven Alfvénic activity and its possible control by electron cyclotron current drive has been examined as well in TJ-II. Finally, recent results on alternative plasma facing components based on liquid metals are also presented.

•The ejection velocities for Li from both liquid metals are basically the same.•There is no significant evolution with temperature of the kinetic energy of the ejected atom over the temperature range recorded by the pyrometer (T > 150 ºC).•Hints of higher energies at T < 150 ºC were found, with values up to 0.5 eV.•The kinetic energy at high temperatures is a factor of 3 to 4 larger that the corresponding thermal energies, but also several times smaller than the expected sputtering energies.•There is no difference in the toroidal dispersal of Li and He ions around their sources, which can be ascribed to classical thermalization with the plasma ions.