Search

In high tech equipment involving high voltages, such as switchgear, surface charging of solid dielectrics can significantly affect the performance. A Pockels-effect based technique is proposed as a promising diagnostic for its capability to create a 2D surface charge map in real time. However, a characterization and comparison of the performance of the most commonly used Pockels setup configurations is missing in existing literature, which makes it difficult to choose the best configuration for a specific task. To overcome this limitation, current knowledge on the diagnostic is extended by simulations, mimicking the Pockels setup. The transfer function between the measurable quantities and the electric field in a sample is derived by Mueller calculus. Finally, experiments are performed to test its correctness. A homogenous electric field is applied on an electro-optic material (Bismuth Silicon Oxide) using transparent electrodes (Indium Tin Oxide). A pre-compensated transmissive Pockels setup has been found to have a less extensive measurement scheme and larger robustness against errors in the setup than a post-compensated transmissive setup. A reflective setup offers a factor two higher sensitivity and allows a non-transparent surface on the sample. However, it has a higher sensitivity to errors and a smaller measurement range than the transmissive setups. With these results, the Pockels-effect based diagnostic becomes more accessible to future users, as it is faster to implement, and the interpretation of results is improved.

To develop realistic liquid lithium divertors for future fusion reactors, this paper aims to improve the understanding of their power handling capabilities. A liquid lithium divertor target prototype, designed to facilitate liquid metal experiments in tokamaks, was tested in Magnum-PSI. The target has an internal reservoir pre-filled with lithium and aims to passively re-supply the textured plasma facing surface during operation. To assess the power handling capability the target was exposed to helium plasmas with increasing power flux density in the linear plasma device Magnum-PSI. The temperature response of lithium targets was recorded via an infrared camera, and compared to finite element method modeling taking into account dissipation via lithium in the plasma. It was found that the target works as intended and can take up to 9 1 MW m−2 for 10 s before the mesh layer was damaged, and could continue operating at higher power densities even after being damaged. The total lifetime of the targets was up to 100 s. Overall the targets are found suitable for use in tokamak experiments. Additionally, a central surface temperature evolution indicative of vapor shielding was observed on intact targets. Predicting the target temperature (and consequently the evaporation rates and thermal stresses) is considered very relevant for the design of lithium divertor targets for DEMO. The observed temperature response could indeed be replicated through modeling, which showed that a significant power fraction was dissipated by the lithium in the plasma.

The InfraRed Video Bolometer (IRVB) is a powerful tool to measure radiated power in magnetically confined plasmas due to its ability to obtain 2D images of plasma emission using a technique that is compatible with the fusion nuclear environment. A prototype IRVB has been developed and installed on NSTX-U to view the lower divertor. The IRVB is a pinhole camera which images radiation from the plasma onto a 2.5 μm thick, 9 × 7 cm2 Pt foil and monitors the resulting spatio-temporal temperature evolution using an IR camera. The power flux incident on the foil is calculated by solving the 2D+time heat diffusion equation, using the foil’s calibrated thermal properties. An optimized, high frame rate IRVB, is quantitatively compared to results from a resistive bolometer on the bench using a modulated 405 nm laser beam with variable power density and square wave modulation from 0.2 Hz to 250 Hz. The design of the NSTX-U system and benchtop characterization are presented where signal-to-noise ratios are assessed using three different IR cameras: FLIR A655sc, FLIR A6751sc, and SBF-161. The sensitivity of the IRVB equipped with the SBF-161 camera is found to be high enough to measure radiation features in the NSTX-U lower divertor as estimated using SOLPS modeling. The optimized IRVB has a frame rate up to 50 Hz, high enough to distinguish radiation during edge-localized-modes (ELMs) from that between ELMs.

Recovery of a flat tungsten surface from a nano-tendril surface is attempted through three techniques; a mechanical wipe, a 1673 K annealing, and laser-induced thermal transients. Results were determined through SEM imaging and elastic recoil detection to assess the helium content in the surface. The mechanical wipe leaves a ∼0.5 μm deep layer of nano-tendrils on the surface post-wipe regardless of the initial nano-tendril layer depth. Laser-induced thermal transients only significantly impact the surface morphology at heat loads of 35.2 MJ/m2 s1/2 or above, however a fully flat or recovered surface was not achieved for 100 transients at this heat load despite reducing the helium content by a factor of ∼7. A 1673 K annealing removes all detectable levels of helium but sub-surface voids/bubbles remain intact. The surface is recovered to a nearly flat state with only some remnants of nano-tendrils re-integrating into the surface remaining.

A prototype of an infrared imaging bolometer (IRVB) was successfully tested on the Alcator C-Mod tokamak at the end of its 2016 campaign. The IRVB method interprets the power radiated from the plasma by measuring the temperature rise of a thin, ∼2 μm, Pt absorber that is placed in the torus vacuum and exposed, using a pinhole camera, to the full-spectrum of plasma’s photon emission. The IRVB installed on C-Mod viewed the poloidal cross section of the core plasma and observed Ohmic and ion cyclotron range of frequency (ICRF)-heated plasmas. Analysis of total radiated power and on-axis emissivity from IRVB is summarized, and quantitative comparisons made to data from both resistive bolometers and AXUV diodes. IRVB results are clearly within a factor of two, but additional effort is needed for it to be used to fully support power exhaust research. The IRVB is shown to be immune to electromagnetic interference from ICRF which strongly impacts C-Mod’s resistive bolometers. Results of the bench-top calibration are summarized, including a novel temperature calibration method useful for IRVBs.

Both the physics of divertor detachment and vapour shielding are characterized by a relatively large amount of radiation produced in the divertor. The linear plasma generator Magnum-PSI is well-suited to study such processes due to its ITER-divertor relevant plasma conditions, simplified geometry and diagnostic accessibility. The need the quantify the plasma radiated power close to the target surface motivated the development of a 4-channel resistive bolometer for Magnum-PSI, and marks the first deployment of such a diagnostic on a linear device. An axially resolved measurement of plasma emission at arbitrary distances from the target surface is now possible. The radial position of the detector can be varied, hereby viewing the full diameter of the plasma column or down to a central region. The overall system design is discussed alongside a comparison of the spectral absorbance of carbon-coated versus non-coated Au/Al bolometer sensors. Despite low electron temperatures of the plasma (1-5 eV), the observed power densities were found to be 10-37 times the sensor noise floor of ∼0.1 W m-2. A synthetic diagnostic based on collisional radiative model calculations from ADAS could well match observed values from H and Ne plasmas while the measured values for Ar and He were more difficult to reproduce. The obtained findings allow for approximate power balance calculations in Magnum-PSI indicating that maximally ∼47% and ∼14% of the total power is lost by radiation in the cases of Ar and Ne/He respectively. The results demonstrate the feasibility of resistive bolometry in low temperature high density plasma regions and on long timescales (>450 s) which is of relevance to ITER. Due to long-term temperature drifts which were observed, a recent upgrade involved the installation of a shutter and FPGA-based electronics for increased accuracy.

The interaction of intense plasma impacting on the wall of a fusion reactor is an area of high and increasing importance in the development of electricity production from nuclear fusion. In the Magnum-PSI linear device, an axial magnetic field confines a high density, low temperature plasma produced by a wall stabilized DC cascaded arc into an intense magnetized plasma beam directed onto a target. The experiment has shown its capability to reach conditions that enable fundamental studies of plasma-surface interactions in the regime relevant for fusion reactors such as ITER: 1023–1025 m−2s−1 hydrogen plasma flux densities at 1–5 eV for tens of seconds by using conventional electromagnets. Recently the machine was upgraded with a superconducting magnet, enabling steady-state magnetic fields up to 2.5 T, expanding the operational space to high fluence capabilities for the first time. Also the diagnostic suite has been expanded by a new 4-channel resistive bolometer array and ion beam analysis techniques for surface analysis after plasma exposure of the target. A novel collective Thomson scattering system has been developed and will be implemented on Magnum-PSI. In this contribution, the current status, capabilities and performance of Magnum-PSI are presented.

In Magnum-PSI (MAgnetized plasma Generator and NUMerical modeling for Plasma Surface Interactions), the high density, low temperature plasma of a wall stabilized dc cascaded arc is confined to a magnetized plasma beam by a quasi-steady state axial magnetic field up to 1.3 T. It aims at conditions that enable fundamental studies of plasma-surface interactions in the regime relevant for fusion reactors such as ITER: 1023-1025 m -2 s-1 hydrogen plasma flux densities at 1-5 eV. To study the effects of transient heat loads on a plasma-facing surface, a high power pulsed magnetized arc discharge has been developed. Additionally, the target surface can be transiently heated with a pulsed laser system during plasma exposure. In this contribution, the current status, capabilities and performance of Magnum-PSI are presented.

The thermal shock behaviour of tungsten exposed to high-flux plasma is studied using a high-power laser. The cases of laser-only, sequential laser and hydrogen (H) plasma and simultaneous laser plus H plasma exposure are studied. H plasma exposure leads to an embrittlement of the material and the appearance of a crack network originating from the centre of the laser spot. Under simultaneous loading, significant surface melting is observed. In general, H plasma exposure lowers the heat flux parameter (FHF) for the onset of surface melting by ∼25%. In the case of He-modified (fuzzy) surfaces, strong surface deformations are observed already after 1000 laser pulses at moderate FHF = 19 MJ m-2 s-1/2, and a dense network of fine cracks is observed. These results indicate that high-fluence ITER-like plasma exposure influences the thermal shock properties of tungsten, lowering the permissible transient energy density beyond which macroscopic surface modifications begin to occur.

The application of liquid metal, especially liquid lithium, as a plasma facing component (PFC) has the capacity to offer a strong alternative to solid PFCs by reducing damage concerns and enhancing plasma performance. The liquid-metal infused trenches (LiMIT) concept is a liquid metal divertor alternative which employs thermoelectric current from either plasma or external heating in tandem with the toroidal field to self-propel liquid lithium through a series of trenches. LiMIT was tested in the linear plasma simulator, Magnum PSI, at heat fluxes of up to 3 MW m-2. Results of these experiments, including velocity and temperature measurements, as well as power handling considerations are discussed, focusing on the 80 shots performed at Magnum scanning magnetic fields and heat fluxes up to ∼0.3 T and 3 MW m-2. Comparisons to predictions, both analytical and modelled, are made and show good agreement. Concerns over MHD droplet ejection are additionally addressed.

Tungsten samples were exposed to combined steady state and edge localised mode transient replication experiments in a linear plasma device; either in combined hydrogen plasma and high powered laser exposures at Magnum-PSI or steady state hydrogen plasma and superimposed plasma pulses created using a capacitor bank in Pilot-PSI. With each transient the surface temperature of the sample was elevated above the melting point creating a shallow molten layer. An apparent heat transport reduction in the near surface layer increases as a function of pulse number. Scanning electron microscopy analysis of the samples shows large scale grain enlargement down to several hundred to thousand micrometers below the surface as well as melt layer motion due to evaporative recoil from the molten surface. In the ITER divertor this would lead to severe embrittlement and enhanced erosion of the tungsten surface leading to reduced lifetimes for the plasma facing material.

The performance of the full-W ITER divertor may be significantly affected by the interplay between steady-state plasma exposure and transient events. To address this issue, the effect of a high-flux H plasma on the thermal shock response of W to ELM-like transients has been investigated. Transient heating of W targets is performed by means of a high-power Nd:YAG laser with simultaneous exposure to H plasma in the linear device Magnum-PSI. The effects of simultaneous exposure to laser and plasma have been compared to those sequentially and to laser only. Transient melting is found to be aggravated during plasma exposure and to occur at lower heat flux parameters. Roughness and grain growth are observed to be driven by peak temperature, rather than by the loading conditions. The temperature evolution of the W surface under a series of transients is recorded by fast infrared thermography. By accounting for changes in the reflectivity at the damaged surface as measured by ellipsometry, a reduction in power handling capabilities of the laser/plasma affected W is concluded. The evidence of reduced power handling of the W surface under conditions as described here is of great concern with respect to the durability of W PFCs for application in fusion devices.

The infrared imaging video bolometer (IRVB) measures plasma radiated power images using a thin metal foil. Two different designs with a tangential view of NSTX-U are made assuming a 640 × 480 (1280 × 1024) pixel, 30 (105) fps, 50 (20) mK, IR camera imaging the 9 cm × 9 cm × 2 μm Pt foil. The foil is divided into 40 × 40 (64 × 64) IRVB channels. This gives a spatial resolution of 3.4 (2.2) cm on the machine mid-plane. The noise equivalent power density of the IRVB is given as 113 (46) μW/cm2 for a time resolution of 33 (20) ms. Synthetic images derived from Scrape Off Layer Plasma Simulation data using the IRVB geometry show peak signal levels ranging from ∼0.8 to ∼80 (∼0.36 to ∼26) mW/cm2.

For DEMO and beyond, liquid metal plasma-facing components are considered due to their resilience to erosion through flowed replacement, potential for cooling beyond conduction and inherent immunity to many of the issues of neutron loading compared to solid materials. The development curve of liquid metals is behind that of e.g. tungsten however, and tokamak-based research is currently somewhat limited in scope. Therefore, investigation into linear plasma devices can provide faster progress under controlled and well-diagnosed conditions in assessing many of the issues surrounding the use of liquid metals. The linear plasma devices Magnum-PSI and Pilot-PSI are capable of producing DEMO-relevant plasma fluxes, which well replicate expected divertor conditions, and the exploration of physics issues for tin (Sn) and lithium (Li) such as vapour shielding, erosion under high particle flux loading and overall power handling are reviewed here. A deeper understanding of erosion and deposition through this work indicates that stannane formation may play an important role in enhancing Sn erosion, while on the other hand the strong hydrogen isotope affinity reduces the evaporation rate and sputtering yields for Li. In combination with the strong redeposition rates, which have been observed under this type of high-density plasma, this implies that an increase in the operational temperature range, implying a power handling range of 20-25 MW m-2 for Sn and up to 12.5 MW m-2 for Li could be achieved. Vapour shielding may be expected to act as a self-protection mechanism in reducing the heat load to the substrate for off-normal events in the case of Sn, but may potentially be a continual mode of operation for Li.

Measurements of radiated power in magnetically confined plasmas are important for exhaust studies in present experiments and expected to be a critical diagnostic for future fusion reactors. Resistive bolometer sensors have long been utilized in tokamaks and helical devices but suffer from electromagnetic interference (EMI). Results are shown from initial testing of a new bolometer concept based on fiber-optic temperature sensor technology. A small, 80 μm diameter, 200 μm long silicon pillar attached to the end of a single mode fiber-optic cable acts as a Fabry–Pérot cavity when broadband light, λo ∼ 1550 nm, is transmitted along the fiber. Changes in temperature alter the optical path length of the cavity primarily through the thermo-optic effect, resulting in a shift of fringes reflected from the pillar detected using an I-MON 512 OEM spectrometer. While initially designed for use in liquids, this sensor has ideal properties for use as a plasma bolometer: a time constant, in air, of ∼150 ms, strong absorption in the spectral range of plasma emission, immunity to local EMI, and the ability to measure changes in temperature remotely. Its compact design offers unique opportunities for integration into the vacuum environment in places unsuitable for a resistive bolometer. Using a variable focus 5 mW, 405 nm, modulating laser, the signal to noise ratio versus power density of various bolometer technologies are directly compared, estimating the noise equivalent power density (NEPD). Present tests show the fiber-optic bolometer to have NEPD of 5-10 W/m2 when compared to those of the resistive bolometer which can achieve