Tungsten Armor Research Articles

Simulation of residual thermostress in tungsten after repetitive ELM-like heat loads

Brittle destruction of tungsten armour under action of edge localised modes of plasma instabilities (ELMs) in ITER is an important issue determining the lifetime of the divertor. Besides, cracking of the armour produces tungsten dust with characteristic size of 1–10 μm flying from the armour surface with velocities up to 10 m/s. Influx of the tungsten dust into the ITER confinement decreases the temperature of the plasma, reduces the thermonuclear gain and even may run the confinement into disruption. This paper describes experiments in QSPA-Kh50 plasma gun and modeling, which has been performed for providing more insight into the physics of tungsten cracking under action of ELMs and for confirmation of the important result on stabilization of the crack development at the tungsten armour surface, predicted in our previous paper – the same authors, 2010. The threshold value of energy density deposition for start of tungsten cracking has been measured as 0.3 MJ/m 2 after 5–10 shots. From analytical considerations three times smaller threshold value has been predicted with increasing number of shots.

Development of the armoring technique for ITER Divertor Dome

This paper describes the current status of the technique for armoring of Plasma Facing Units (PFUs) of the ITER Divertor Dome with flat tungsten tiles planned for application at the procurement stage. Application of high-temperature vacuum brazing for armoring of High Heat Flux (HHF) plasma facing components was traditionally developed at the Efremov Institute and successfully tried out at the ITER R&D stage by manufacturing and HHF testing of a number of W- and Be-armored mock-ups [1,2]. Nevertheless, the so-called “fast brazing” technique successfully applied in the past was abandoned at the stage of manufacturing of the Dome Qualification Prototypes (Dome QPs), as it failed to retain the mechanical properties of CuCrZr heat sink of the substrate. Another problem was a substantially increased number of armoring tiles brazed onto one substrate. Severe ITER requirements for the joints quality have forced us to refuse from production of W/Cu joints by brazing in favor of casting. These modifications have allowed us to produce ITER Divertor Dome QPs with high-quality tungsten armor, which then passed successfully the HHF testing. Further preparation to the procurement stage is in progress.

Simulation of tungsten armour cracking due to small ELMs in ITER

Simulations of tungsten armour cracking under small ELM-like plasma heat load, which does not cause surface melting, have been performed using the PEGASUS-3D code. A dedicated series of experiments have been performed in the QSPA-Kh50 facility for measurements of the unknown tungsten thermophysical properties and for verification of the PEGASUS-3D simulation results. The simulations revealed that a cellular crack network with average mesh size Λ ∼ 0.5 mm formed after first ELMs and the pattern does not change further. With increasing number of repetitive ELMs loads, the average crack width Δ( n) has a maximum value Δ m . The ratio of Δ m / Λ is equal to the tungsten thermal expansion at the maximum surface temperature. Δ( n) tends to this value exponentially. The number of ELMs n m needed for Δ stabilization depends on the ELMs energy density and time duration, n m ∼ 300 for the simulated ELMs of 0.45 MJ/m 2 and 0.25 ms duration. The PEGASUS-3D code is prepared for simulations of tungsten armour damage under action of ELMs of various energy deposition and time duration. These parameters of ELMs depend on ITER regimes of operation and on how successful will be the efforts on ELMs mitigation.

Control of the Composition Profile of Tungsten-Copper Functionally Graded Materials for Fusion Technology Application

This work aims to process W-Cu FGM for fusion material applications, in particular plasma facing components (PFC). Currently, PFCs are made by multi-material assembly and rely on tungsten armour tile (high heat flux side) which needs to be cooled by the attachment to a heat sink (Cu-alloys). The interfaces in multi-material assembly are unfavourable for the application (high working temperature, thermal fatigue). The use of FGM can be a new promising way. The aim of this study is to analyze the effect of a composition/grain size variation on the liquid phase migration during liquid phase sintering of FGM, in order to control the final composition profile. Bi-layer materials are processed by powder metallurgy route. W-Cu compositions with 10 and 20wt%Cu and with different W-particle sizes are processed starting with attritor-milled W-CuO powder mixtures which are reduced at 350°C. Analysis of copper liquid phase migration for different composition/grain size associations indicates that the phenomenon is driven by differential sinterability in the gradient. The copper liquid migration depends on the differences in sinterability, on the available liquid and open porosity in the structure beyond the melting point of copper. From these analyses, a way to control the gradient profile of W-Cu structure can be proposed.

Self-consistent analysis of the effect of runaway electrons on plasma facing components in ITER

Physical and computational models are developed, used and benchmarked for studying the response of ITER tokamak plasma facing components to runaway electron impact following a plasma disruption. The energy deposition, temperature evolution and material melting thickness are calculated for a wide range of runaway electron parameters, namely, electron kinetic energy, magnetic field, energy partition ratio (along and across magnetic field direction) impact duration, and wall material composition. It is shown that the electron energy partition ratio will have a significant effect on the wall heat load with melting of the first wall with beryllium armor possible. If tungsten armor is used instead, the surface of the mockup is overheated and melted for all ranges of studied parameters of the runaway electrons. Using an insert of a thin layer of a high-Z material inside the beryllium armor can mitigate the heat load in the armor and heat sink structure.

Conceptual design of a fast-ignition laser fusion reactor based on a dry wall chamber

The fast ignition is quite attractive for a compact laser fusion reactor, because a sufficiently high pellet gain is available with a small input energy. We designed an inertial fusion reactor based on Fast-ignition Advanced Laser fusion reactor CONcept, called FALCON-D, where a dry wall is employed for a chamber wall. A simple point model shows that the pellet gain G∼100 is available with laser energies of 350kJ for implosion, 50kJ for heating. This results in the fusion yield of 40 MJ in one shot. By increasing the repetition rate up to 30 Hz, the fusion power of 1.2 GWth becomes available. Plant system analysis shows the net electric power to be about 0.4 GWe In the fast ignition it is available to employ a low aspect ratio pellet, which is favorable for the stability during the implosion phase. Here the pellet aspect ratio is reduced to be 2 ∼ 4, and the optimization of the pulse shape for the implosion laser are carried out by using the 1-D hydrodynamic simulation code ILESTA-1D. A ferritic steel with a tungsten armour is employed for the chamber wall. The feasibility of this dry wall concept is studied from various engineering aspects such as surface melting, physical and chemical sputtering, blistering and exfoliation by helium retention, and thermo-mechanical fatigue, and it is found that blistering and exfoliation due to the helium retention and fatigue failure due to cyclic thermal load are major concerns. The cost analysis shows that the construction cost is moderate but the cost of electricity is slightly expensive.

Failure Strength Measurements of VPS Tungsten Coatings for HAPL First Wall Armor

The High Average Power Laser (HAPL) project is pursuing development of an IFE power reactor using a solid first wall chamber. Tungsten has been chosen as the primary candidate armor material protecting the low activation ferritic steel chamber wall structure. The tungsten armor is less than 1-mm thick and is applied by vacuum plasma spraying (VPS). The failure strength of the tungsten-armor is critical, which is measured using a state-of-the-art spallation technology developed at UCLA. A nano-second laser is used to propagate a compression/tension stress wave through the composite layered structure. The tensile strength in the coating is then related to the displacement velocity of the free surface of the tungsten coating. VPS tungsten coated steel samples were tested using the laser spallation technique and coating strengths were evaluated and are reported.

High heat load properties of actively cooled tungsten/copper mock-ups by explosive joining

Two actively cooled mock-ups with 2mm thick tungsten armor joined to chromium copper alloy by explosive welding were developed, tested and analyzed in ASIPP of China. The heat load limit of the mock-up is 7MW/m2, and delamination was observed along the W/Cu interface. The mock-ups can withstand the maximum heat flux of 6MW/m2. SEM investigations showed that some cracks and failures appeared in filler materials and at copper/filler interface, but intactness was found at tungsten/filler interface. Numerical simulations indicated that if the distance decreases between the vertex of channel and the interface, the mock-up had the better structure reliability due to the reduction of temperature and stress. Under 4MW/m2 heat flux, the mock-up showed good heat transfer capabilities and high joint reliabilities. All these results demonstrate the explosive joining is also an alternative way to realize the W/Cu joint for the plasma facing components.

Recent developments toward the use of tungsten as armour material in plasma facing components

Future fusion experiments will rely on tungsten armour tile for their plasma facing components. In order to sustain steady state operation, the components need to be cooled through an attachment to a heat sink. All current reference concepts rely on contact bonds, unfavourable for long-term application (high temperature service, cycle fatigue, thermal shocks). Three routes toward the development of thick tungsten bonds are presented here, namely functionally graded tungsten copper assembly, electron beam welding of tungsten, and coating processes. All present favourable prospects, and tend to indicate that a thick bond is possible with tungsten. Dedicated programs as well as industrial implication are however required if such concepts are to be used actually for the fabrication of large components series.

Material Integration Design for the Divertor Target Plate of EAST Device

In the next generation of fusion device in China, e.g., the Experimental Advanced Superconducting Tokamak (EAST), the divertor target will be exposed to high heat loads up to 5 MW/m2 for about 1000 s. An actively water-cooled target plate element with flat tungsten tile armored on CuCrZr heat sink was designed for EAST. A two-dimensional finite element method (FEM) code was used to analyze its thermal and mechanical properties under high heat flux of 10 MW/m2 for the selection of an appropriate cross section. To meet the integrated requirements of temperature and stress in the target element, twisted tapes have to be inserted into the cooling channels to strengthen the heat transfer efficiency, and a tungsten armor thickness of 4 mm and a distance of 2 mm from the interface to the vertex of the cooling channel were ultimately selected. The thermal and mechanical properties of two kinds of tungsten armor (sintered and plasma sprayed) were also analyzed and discussed in the FEM calculations. The designed structure can be used under the 5 MW/m2 heat load expected for normal operation of EAST device, but it would suffer from cracks/failure danger under higher heat load, up to 10 MW/m2.

Simulation of thermal response and ion deposition in the HAPL target chamber 1 mm tungsten armor layer using the improved BUCKY code

Improved radiation hydrodynamics 1-D simulations were performed with the BUCKY code using new models for the plasma–wall interface and for ion source term transport. The new models were used to investigate the thermal response and ion implantation in a tungsten armor shell at a radius of 10.5 m with a thickness of 1 mm, which correspond to the nominal parameters for the high average power laser (HAPL) reactor target chamber. The simulations were performed using a chamber buffer gas of xenon at pressures of 8 μTorr and 8 mTorr. Two direct-drive target explosions were simulated: a 343 MJ empty foam target and a 365 MJ Pd–Au coated target. The simulations were run to 12.5 μs to ensure that all of the simulated ions were stopped in the tungsten armor. Surface temperature peaks due to X-rays, debris ions and kinetic ions were found for both the empty foam and the Pd–Au coated targets. Ion implantation results from the BUCKY simulations were compared to SRIM simulations of the same ion spectra to assess the validity of using BUCKY to estimate ion stopping in the tungsten armor. Parametric calculations were performed to ascertain the effect of buffer gas pressure on surface temperature rise in the tungsten armor.

Transmutation and Phase Stability of Tungsten Armor in Fusion Power Plants

We present calculations of the transmutation of initially pure tungsten first-wall and divertor plasma-facing armor into W-Re-Os alloys in the European Union Power Plant Conceptual Study (PPCS) fusion plant models A, B, and AB. The fusion neutron spectrum was modeled using the MCNP Monte Carlo code including resonance self-shielding effects, and we have calculated the evolution of the W-Re-Os alloy compositions. Trajectories of the alloys in the thermodynamic phase diagram show that the alloys remain in the single body-centered-cubic phase for their service lifetimes. Results for PPCS models A and B with soft neutron spectra show that the first-wall armor transmutes to an end-of-service alloy composition of approximately 91 at.% tungsten, 6 at.% rhenium, and 3 at.% osmium at its rear face. On the plasma-facing side of the tungsten, the effect of neutron shielding is larger. For PPCS model AB, the neutron spectrum is energetically harder, resulting in significantly lower tungsten transmutation rates.

Systems modeling for laser IFE

A systems model of a laser-driven IFE power plant is being developed to assist in design trade-offs and optimization. The focus to date has been on modeling the fusion chamber, blanket and power conversion system. A self-consistent model has been developed to determine key chamber and thermal cycle parameters (e.g., chamber radius, structure and coolant temperatures, cycle efficiency, etc.) as a function of the target yield and pulse repetition rate. Temperature constraints on the tungsten armor, ferritic steel wall, and structure/coolant interface are included in evaluating the potential design space. Results are presented for a lithium cooled first wall coupled with a Brayton power cycle.

Key achievements in elementary R&D on water-cooled solid breeder blanket for ITER test blanket module in JAERI

This paper presents the significant progress made in the research and development (R&D) of key technologies on the water-cooled solid breeder blanket for the ITER test blanket modules in JAERI. Development of module fabrication technology, bonding technology of armours, measurement of thermo-mechanical properties of pebble beds, neutronics studies on a blanket module mockup and tritium release behaviour from a Li2TiO3 pebble bed under neutron-pulsed operation conditions are summarized.With the improvement of the heat treatment process for blanket module fabrication, a fine-grained microstructure of F82H can be obtained by homogenizing it at 1150 °C followed by normalizing it at 930 °C after the hot isostatic pressing process. Moreover, a promising bonding process for a tungsten armour and an F82H structural material was developed using a solid-state bonding method based on uniaxial hot compression without any artificial compliant layer. As a result of high heat flux tests of F82H first wall mockups, it has been confirmed that a fatigue lifetime correlation, which was developed for the ITER divertor, can be made applicable for the F82H first wall mockup. As for R&D on the breeder material, Li2TiO3, the effect of compression loads on effective thermal conductivity of pebble beds has been clarified for the Li2TiO3 pebble bed. The tritium breeding ratio of a simulated multi-layer blanket structure has successfully been measured using 14 MeV neutrons with an accuracy of 10%. The tritium release rate from the Li2TiO3 pebble has also been successfully measured with pulsed neutron irradiation, which simulates ITER operation.

Testing IFE materials on Z

On a single-pulse basis, the tungsten armor for the chamber walls in a laser inertial fusion energy power plant must withstand X-ray fluences of 0.4–1.2 J/cm 2 with almost no mass loss, and preferably no surface changes. We have exposed preheated tungsten samples to 0.27 and 0.9 J/cm 2 X-ray fluence from the Z accelerator at Sandia National Laboratories to determine the single-shot X-ray damage threshold. Earlier focused ion beam analysis has shown that rolled powdered metal formed tungsten and tungsten alloys, will melt when exposed to 2.3 J/cm 2 on Z, but not at 1.3 J/cm 2. Three forms of tungsten – single-crystal (SING), chemical-vapor-deposited (CVD), and rolled powdered metal (PWM) – were exposed to fluence levels of 0.9 J/cm 2 without any apparent melting. However, the CVD and PWM sample surfaces were rougher after exposure than the SING sample, which was not roughened. BUCKY (1D) calculations show a threshold of 0.5 J/cm 2 for melting on Z. The present experiments indicate no melting but limited surface changes occur with polycrystalline samples (PWM and CVD) at 0.9 J/cm 2 and no surface changes other than debris for samples at 0.27 J/cm 2.

An overview of the development of the first wall and other principal components of a laser fusion power plant

This paper introduces the JNM Special Issue on the development of a first wall for the reaction chamber in a laser fusion power plant. In this approach to fusion energy a spherical target is injected into a large chamber and heated to fusion burn by an array of lasers. The target emissions are absorbed by the wall and encapsulating blanket, and the resulting heat converted into electricity. The bulk of the energy deposited in the first wall is in the form of X-rays (1.0–100 keV) and ions (0.1–4 MeV). In order to have a practical power plant, the first wall must be resistant to these emissions and suffer virtually no erosion on each shot. A wall candidate based on tungsten armor bonded to a low activation ferritic steel substrate has been chosen as the initial system to be studied. The choice was based on the vast experience with these materials in a nuclear environment and the ability to address most of the key remaining issues with existing facilities. This overview paper is divided into three parts. The first part summarizes the current state of the development of laser fusion energy. The second part introduces the tungsten armored ferritic steel concept, the three critical development issues (thermo-mechanical fatigue, helium retention, and bonding) and the research to address them. Based on progress to date the latter two appear to be resolvable, but the former remains a challenge. Complete details are presented in the companion papers in this JNM Special Issue. The third part discusses other factors that must be considered in the design of the first wall, including compatibility with blanket concepts, radiological concerns, and structural considerations.

Threats, design limits and design windows for laser IFE dry wall chambers

The High Average Power Laser (HAPL) program is carrying out a coordinated effort to develop inertial fusion energy based on lasers, direct-drive targets and a dry wall chamber. The dry wall must accommodate the ion and photon threat spectra from the fusion micro-explosion over its required lifetime. This paper summarizes the current HAPL strategy on the armor/first wall configuration based on tungsten and ferritic steel as preferred armor and structural materials, respectively. The thermal performance of an example fully dense tungsten armor configuration on a ferritic steel first wall is described showing the basis for separating the high energy accommodation function of the armor from the structural function of the first wall. Example design operating windows for the armor, first wall and blanket are presented based on different requirements and constraints. The possibility of utilizing an engineered porous armor is discussed. Key chamber wall and armor issues are summarized.

Micro-engineered first wall tungsten armor for high average power laser fusion energy systems

The high average power laser program is developing an inertial fusion energy demonstration power reactor with a solid first wall chamber. The first wall (FW) will be subject to high energy density radiation and high doses of high energy helium implantation. Tungsten has been identified as the candidate material for a FW armor. The fundamental concern is long term thermo-mechanical survivability of the armor against the effects of high temperature pulsed operation and exfoliation due to the retention of implanted helium. Even if a solid tungsten armor coating would survive the high temperature cyclic operation with minimal failure, the high helium implantation and retention would result in unacceptable material loss rates. Micro-engineered materials, such as castellated structures, plasma sprayed nano-porous coatings and refractory foams are suggested as a first wall armor material to address these fundamental concerns. A micro-engineered FW armor would have to be designed with specific geometric features that tolerate high cyclic heating loads and recycle most of the implanted helium without any significant failure. Micro-engineered materials are briefly reviewed. In particular, plasma-sprayed nano-porous tungsten and tungsten foams are assessed for their potential to accommodate inertial fusion specific loads. Tests show that nano-porous plasma spray coatings can be manufactured with high permeability to helium gas, while retaining relatively high thermal conductivities. Tungsten foams where shown to be able to overcome thermo-mechanical loads by cell rotation and deformation. Helium implantation tests have shown, that pulsed implantation and heating releases significant levels of implanted helium. Helium implantation and release from tungsten was modeled using an expanded kinetic rate theory, to include the effects of pulsed implantations and thermal cycles. Although, significant challenges remain micro-engineered materials are shown to constitute potential candidate FW armor materials.

Flow Field and Thermal Analysis of the Divertor Target Plate for HL-2A Tokamak

In the initial phase of the physics experiment, the double-null divertor plates used consist of graphite armor tiles, Mo-alloy intermediate layers and Cu-alloy coolant tubes. In the later operating phase, tungsten will be used as armor tiles.A multi-physical field numerical analysis method is used in this paper. Its analysis model reflects more realistically the real divertor structure than other models. Two-dimensional (2D) and three-dimensional (3D) fluid flow field, temperature distribution and thermal stress analyses of the divertor plates are carried out by the ANSYS code. During the physics experimental phase with a heat flux of 1 MW/m2, a coolant velocity of 5.48 m/s, and a thermal stress of 750 kg/cm2, the graphite armor tiles successfully meet the requirements of temperature, thermal stress and sputtering erosion. The tungsten armor will be considered as a second candidate. The result of simulation can be used for upgrading the design parameters of the HL-2A poloidal divertor.

Simulation of multi-atomic interactions in H–O–W system with the MD code CADAC

For future tokamak reactors, chemical erosion of tungsten armour surfaces under impact of hot deuterium–tritium plasma that contains impurities, for instance oxygen, is an important issue. Oxygen can form volatile molecular complexes O x W y at the surface, and the retained H-atoms form the volatile complexes H x O y , which mitigates the erosion (H states for hydrogen isotopes). The plasma impact can substantially destroy the complexes. To describe this H–O–W system, the molecular dynamics (MD) code CADAC was earlier developed using only pair–atomic interactions. Now CADAC is extended for multi-body forces to simulate molecular organization of atoms near the tungsten surface. The approach uses the Abell's model of empirical bond-order potentials in addition combined, for the first time, with a valence concept. CADAC simulates chemical features using atomic valences and the Morse potentials. The new model is introduced and model parameters are estimated.