Search

Liquid metal flows exposed to intense magnetic fields play a fundamental role in the development of blankets for nuclear fusion reactors, where they serve to produce the plasma-fuel component tritium and transport the generated heat. When the liquid metal circulates in the blanket, it interacts with the plasma-confining magnetic field leading to the induction of electric currents. Electromagnetic forces and thermal buoyancy affect significantly velocity and pressure distributions. The resulting magneto-convective flow has peculiar features that have to be taken into account in order to evaluate heat transfer properties in the liquid metal. In the breeding zone of the water-cooled lead lithium blanket concept, the volumetric heat released in the liquid metal is removed by water-cooled circular pipes that represent obstacles for the liquid metal flow. In order to improve the understanding of the underlying physical phenomena and obtain a database for code validation, model experiments have been performed to investigate magneto-convective flow and heat transfer at two differentially heated parallel horizontal tubes immersed in a box filled with liquid metal. Among other properties, electric potential has been recorded on the surface of the test-section and compared with 3D numerical simulations. Computational results provide the basis for interpretation of the measured data, since they allow differentiating between flow-induced electric potential and thermoelectric effects. • Magneto-convection in a liquid metal is investigated numerically and experimentally in a model geometry for WCLL blankets. • Measurements of electric potential in MHD convective flows is significantly affected by thermoelectric effects, since measured data is given by voltage between electrodes at different temperatures. • Flow-induced electric potential and thermoelectric contribution have to be separated to use the recorded wall potential to reconstruct velocity features. • A physics-based correct interpretation of electric potential measurements is achieved via numerical simulations. • A good agreement between experiments and simulations is found. [ABSTRACT FROM AUTHOR]

The present paper gives an overview of liquid metal research activities performed in recent years at the Karlsruhe Institute of Technology (KIT). The work is motivated by applications in liquid metal blankets for a DEMO fusion reactor where lead lithium (PbLi), which serves as a neutron multiplier and tritium breeder, interacts with the plasma-confining magnetic field as it flows in the blanket covering the inner walls of the reactor. Liquid metal magnetohydrodynamic (MHD) research at KIT supports blanket design activities through theoretical and experimental investigations. Predictive computational tools are developed and validated by empirical data obtained for fundamental problems, such as flows in non-uniform magnetic fields or magneto-convective heat transfer from submerged obstacles. In addition, technological developments like pressure drop reduction by insulating flow channel inserts are pursued both theoretically and experimentally. Two complementary experimental facilities (MEKKA and MaPLE) provide a unique and versatile platform for MHD investigations at fusion relevant parameters. Using NaK as a model fluid in MEKKA allows experiments to be conducted at high Hartmann numbers in large complex geometries, such as scaled blanket mock-ups of ITER test blanket modules. For magneto-convection and heat transfer studies, MaPLE is well suited since it enables experiments with the prototypical fluid PbLi in test sections inclined at various orientations with respect to gravity. Some recent results have been selected to illustrate the broad spectrum of MHD activities at KIT. [ABSTRACT FROM AUTHOR]

The flow of the electrically conducting lead lithium, foreseen in liquid metal blankets as breeder and coolant, under the influence of the plasma-confining magnetic field induces electric currents that create strong electromagnetic Lorentz forces and high magnetohydrodynamic pressure drop. Electrically insulating flow channel inserts (FCI) may be applied for electrically decoupling the liquid metal flow from the well-conducting channel walls. This reduces currents and associated Lorentz forces that are responsible for the major contribution to pressure drop in the blankets. For fabrication reasons, it is unavoidable that gaps between inserts are present. Gaps will interrupt the electrical insulation, thus providing unwanted local short circuit for electric currents. For experimental investigations of 3D effects at junctions of FCIs, a test section has been manufactured and experiments have been performed in the MEKKA facility at the Karlsruhe Institute of Technology (KIT). The present experimental study shows the benefits of FCIs for pressure drop reduction in fully developed flows and quantifies the deterioration of pressure drop reduction by the presence of uninsulated gaps between FCIs. [ABSTRACT FROM AUTHOR]

• Numerical results of three dimensional MHD investigations in a DCLL blanket. • MHD flow behavior becomes inertialess. • Bends and U-turn do not create large additional pressure drops. • Pressure drop in blanket ducts resembles fully developed laminar duct flow. In the dual coolant lead lithium blanket (DCLL) the liquid metal PbLi is used as coolant to remove the volumetrically generated heat. This requires higher flow velocities than in separately cooled blanket concepts, where the liquid metal serves exclusively as breeder, neutron multiplier and shield against high neutron radiation. The relatively fast movement of the electrically conducting fluid in the strong magnetic field confining the fusion plasma induces electric currents, which are responsible for strong Lorentz forces, high magnetohydrodynamic (MHD) pressure drop, and substantial modifications of velocity profiles compared with hydrodynamic flows. In order to support the design activities of a DCLL blanket, 3D numerical MHD simulations for one column of such a blanket module have been performed. The simulations take into account, that currents may close along the inner conducting sheet of sandwich-type flow channel inserts that are foreseen for pressure drop reduction. The applied numerical code has been carefully validated. It applies for magnetic fields of arbitrary orientation and is capable of simulating the liquid metal MHD flow in the blanket in the parameter range relevant for fusion applications. [ABSTRACT FROM AUTHOR]

Abstract Thermohydraulic measurements in liquid metal test blanket modules for ITER require reliable sensors for detection of temperature and magnetohydrodynamic properties of the flow such as the distribution of electric potential. The latter one can be directly compared with numerical simulations or interpreted in terms of local liquid metal flow rates. Combined temperature-potential probes may serve for both purposes for sensing temperature and potential values, while remaining relatively small in size. By comparison with results obtained using other sensing techniques or theoretical results, the present paper confirms that such measurements in magnetohydrodynamic liquid metal flows may yield accurate data for potential. [ABSTRACT FROM AUTHOR]

In the frame of the EUROfusion breeding blanket research activities, the water cooled lead lithium (WCLL) blanket is considered as reference liquid metal blanket design to be tested in ITER and to be used in a DEMO reactor. For the study of pressure and flow distribution in a WCLL blanket module, magnetohydrodynamic (MHD) phenomena, which result from the motion of the electrically conducting breeder in the intense plasma-confining magnetic field, have to be taken into account. They are due to the induction of electric currents that generate strong electromagnetic forces, which modify the velocity distribution in the blanket compared to hydrodynamic conditions and increase pressure losses. In the WCLL test blanket module (TBM) for ITER a basic geometry, consisting of a rectangular box, representing the breeding zone, is repeated along the poloidal direction. The manifold that distributes and collects the liquid metal features two long poloidal ducts, electrically connected across a common wall. The final goal of the present work is to simulate liquid metal MHD flows in strong magnetic fields in a complete column of a WCLL TBM including poloidal manifolds and 8 breeder units. With this aim, the complexity of the model geometry has been progressively increased and various topologies of computational grids have been considered. First simulations have been performed in a single breeder zone connected with a portion of manifolds in order to test the suitability of the generated grid and the performance of the numerical code for this type of problem. In a second step, the MHD flow at moderate magnetic fields in an entire TBM mock-up has been simulated to get pressure distribution along the manifolds and flow partitioning among breeder units. • Efficient mesh workflow for simulations of MHD flows in complex geometries. • Numerical schemes with non-orthogonality corrections have been employed. • Robust gradient discretization scheme for electric potential gradient. • First simulations of MHD flows in an entire experimental mock-up of a WCLL TBM. [ABSTRACT FROM AUTHOR]

In the helium cooled lead lithium (HCLL) blanket concept the eutectic alloy PbLi is used as breeder material and the heat is removed by helium. The liquid metal flows in rectangular boxes, called breeder units (BUs), which are arranged in columns to form a blanket module. Experiments have been performed to study MHD flows in a HCLL blanket mock-up that consists of 8 BUs fed and drained by poloidal manifolds. Electromagnetic flow coupling of adjacent fluid domains, due to an exchange of electric currents, plays an essential role in determining the velocity distribution in the breeding zone. During experiments electric potential on the walls perpendicular to the magnetic field has been recorded. In general, the electric potential serves as approximate streamfunction of the flow if walls are thin and induced currents are small. However, if the walls of the mock-up are thick, as in the performed experiments, induced currents should be taken into account. In order to correctly interpret surface electric potential measurements in terms of velocity distribution inside the test-section, a comparison with numerical simulations is required. A procedure has been developed to derive global solutions for magnetohydrodynamic (MHD) flows and to predict the velocity distribution in the test-section starting from experimental potential measurements and by comparing them with numerical results. [ABSTRACT FROM AUTHOR]

In liquid metal blanket concepts, such as the dual coolant lead lithium (DCLL) blanket, in which the PbLi serves to cool the breeding zone, the liquid metal has to flow with sufficiently large velocity to guarantee a suitable removal of the volumetric heat generated in the fluid. Large velocities lead to a high pressure drop caused by magnetohydrodynamic (MHD) interactions of the liquid metal with the plasma confining magnetic field. In order to reduce MHD pressure losses in DCLL blankets, it is foreseen to decouple electrically the PbLi from the conducting wall of the ducts. This can be achieved by inserting insulating layers inside pipes and channels. In this paper we investigate MHD flows in well-conducting pipes with flow channel inserts (FCI). The influence on the pressure drop of the finite electric conductance of the layer and of the presence of an interruption in the insulation is analyzed. Numerical results show that under realistic operating conditions, i.e. for sufficiently small electric conductance of the insert and large magnetic fields, the ensuing additional MHD pressure losses are still acceptable. However, the discontinuity of the insulation causes significant modifications of the velocity distribution in the pipe, which have to be taken into account, since they can affect heat and mass transfer phenomena. [ABSTRACT FROM AUTHOR]

Liquid metal magnetohydrodynamic (MHD) flows in a scaled mock-up of a Helium cooled lead lithium blanket have been investigated experimentally in the MEKKA laboratory at the Karlsruhe Institute of Technology. The test section consists of 8 breeder units (BUs) connected two by two at the first wall (FW) through a series of small openings. The large boxes are fed and drained by poloidal manifolds formed by two parallel channels separated by a step-shaped wall, which is electrically conducting. The complexity of the geometry and the strong 3D MHD effects that characterize the flow in the mock-up are at the present time still very challenging for numerical simulations. Therefore, experiments are required to obtain an overview of pressure and velocity distributions in such geometry and to derive scaling laws to extrapolate the results to realistic ITER operating conditions. Pressure differences between several points on the test section have been measured. The influence of magnetic field strength and flow rate on pressure losses in the mock-up has been studied. The largest pressure drops occur along the manifolds. Another significant part of the entire pressure drop comes from flow near the FW, where the liquid metal passes through small gaps into the adjacent BU. This pressure drop between external pairs of BUs is at least 4–5 times larger than that in the internal ones. This suggests an inhomogeneous partitioning of the liquid metal flow among the BUs, which is related to the geometry of the manifold. [ABSTRACT FROM AUTHOR]

In the water cooled lead lithium (WCLL) blanket liquid lead lithium PbLi is used as breeder, neutron multiplier, and as heat transfer medium. The released heat is removed by water that flows at a pressure of 155 bar through cooling pipes immersed in the liquid-metal pool. In order to withstand disruption-induced forces and water pressure in case of accidental conditions, the breeder zone is stiffened by internal plates that form rectangular ducts in which the liquid metal is confined. Numerical simulations have been performed to predict liquid metal flows in model geometries for WCLL blankets under the influence of external magnetic fields. In the analysis the flow is driven by an applied pressure gradient or by buoyancy due to the presence of volumetric heat sources in the fluid and heat removal at the water-cooled pipes. For strong magnetic fields the flow exhibits several quasi-2D cores with jumps in core velocity across thin internal layers tangent to the cooling pipes. [ABSTRACT FROM AUTHOR]

In 2008–2009 experiments have been performed to investigate liquid metal magnetohydrodynamic (MHD) flows in a scaled mock-up of a helium cooled lead lithium (HCLL) blanket. In order to improve the mechanical stiffness of HCLL blanket modules the design of the stiffening plate between two hydraulically connected breeder units (BUs) has been later modified. In the former design the liquid metal passed from one BU into the adjacent one by flowing through a narrow gap that extended along the entire width of the BU. In the most recent design this opening has been replaced by a series of smaller gaps. Therefore the velocity increases locally owing to the reduced cross-section along the flow path and the liquid metal contracts and expands along magnetic field lines to enter the neighboring BU. These flow conditions are known to create additional MHD pressure losses. In order to estimate the influence of design changes on MHD flow and pressure drop the previous test section has been adapted to the new design. Experimental results show that the modifications of the design at the first wall lead to an increase of pressure drop near the first wall by a factor 3–3.5. As a consequence the total pressure drop becomes larger. The 3D MHD phenomena that occur at the first wall seem to be mainly related to inertia effects that are confined in layers parallel to the magnetic field. [ABSTRACT FROM AUTHOR]

The current reference concept for a European liquid metal test blanket module (TBM) for ITER is a water-cooled lead lithium (WCLL) blanket. The liquid alloy lead-lithium (PbLi) is used as breeder, neutron multiplier, and heat carrier. Pressurized water cools the blanket structure and the breeding zone by a large number of cooling pipes immersed in the liquid metal. The spatial distribution of these pipes interfering the breeding zone and manifolds is a critical issue. Efficient heat removal from the liquid metal has to be ensured to avoid local hotspots in the fluid or in blanket walls. The highest liquid metal velocities occur in the feeding and draining manifolds, which represent quite complex geometrical structures. It is expected that the magnetohydrodynamic (MHD) flow of PbLi in the manifolds generates the major fraction of pressure drop. Moreover, the design of the manifolds has a decisive influence on the flow partitioning in the blanket, and their contribution to the overall performance of the module is significant. In order to investigate the MHD impact on the liquid metal flow, a scaled mockup of the WCLL TBM has been designed with focus on detailed measurements of flow quantities in the manifolds. The geometry of the mockup has been derived from the ITER TBM design. It takes into account internal solid obstacles such as cooling pipes, which block parts of the fluid volume and force the flow to complex 3D meandering paths. The present paper gives an overview of the WCLL mockup design, fabrication and measurement system. This information could be quite useful for engineers who are planning future experiments to see how some technical problems have been solved here. The mockup has been integrated in the liquid metal loop of the MEKKA facility of the Karlsruhe Institute of Technology for MHD experiments. Results will improve the current understanding of MHD flows in WCLL manifolds and help to improve and optimize the design of WCLL blankets. [ABSTRACT FROM AUTHOR]

An efficient hybrid model has been developed for prediction of magnetohydrodynamic pressure drop in electrically couple manifolds of liquid metal blankets and flow partitioning in breeder units. The tool combines global mass conservation and pressure drop correlations with detailed 3D simulations. When applied to a TBM-like geometry, the model reveals strong electromagnetic coupling with increased flow in BUs is near both ends of the module, while weak coupling with almost no flow is found in BUs in the middle of the module. The method is very efficient and it applies for a large number of BUs as foreseen in a DEMO design. It will be applied in future to determine the optimum position of the baffle plates in manifolds that guarantees uniform flow partitioning in all breeder units. [ABSTRACT FROM AUTHOR]

In the framework of the development of liquid metal blankets to be tested in ITER and for applications in a DEMOnstration reactor, magnetohydrodynamic (MHD) interactions between the electrically conducting lead lithium alloy and intense magnetic fields have been investigated numerically. In all proposed liquid metal blanket designs the presence of leakage currents that flow from one fluid domain into the adjacent one crossing electrically conducting walls has to be taken into account since it influences significantly the flow partitioning in parallel ducts. The way in which electrically coupled flows affect each other depends on orientation of common walls with respect to the magnetic field, on flow direction in adjacent channels and on the path of leakage currents. The present paper aims at providing an overview of main phenomena that occur in multiple channels where flows are coupled by an exchange of electric currents. The understanding of those basic flows is required for the study of complex 3D coupling conditions. [ABSTRACT FROM AUTHOR]

Theoretical studies are performed to predict magnetohydrodynamic flow pattern and pressure drop in typical geometric components of the liquid metal manifold for a HCLL blanket module. The interaction of the flow with the strong magnetic field that confines the fusion plasma is analyzed by asymptotic techniques and by numerical simulations. The flows in feeding and draining manifolds are electrically coupled to each other via leakage currents which may cross the common wall that hydraulically separates both channels. This results in complex flow paths in the manifold and in additional pressure drop. [ABSTRACT FROM AUTHOR]

Since many years, liquid metal flows for applications in fusion blankets have been investigated worldwide. A review is given about modeling requirements and existing experimental facilities for investigations of liquid metal related issues in blankets with the focus on magnetohydrodynamics (MHD). Most of the performed theoretical and experimental works were dedicated to fundamental aspects of MHD flows under very strong magnetic fields as they may occur in generic elements of fusion blankets like pipes, ducts, bends, expansions and contractions. Those experiments are required to progressively validate numerical tools with the purpose of obtaining codes capable to predict MHD flows at fusion relevant parameters in complex blanket geometries, taking into account electrical and thermal coupling between fluid and structural materials. Scaled mock-up experiments support the theoretical activities and help deriving engineering correlations for cases which cannot be calculated with required accuracy up to now. [ABSTRACT FROM AUTHOR]

Heat and mass transport phenomena in liquid metal blankets are strictly related to velocity distributions near the walls. It is known that the thin boundary layers, which form along walls parallel to an imposed magnetic field, can lose their stability due to the presence of high velocity jets with inflection points. This can lead to the formation of complex vortical structures whose size and features depend on mean flow rate and magnetic field strength. With the aim of investigating experimentally flow instabilities in liquid metal MHD duct flows, a loop has been manufactured and fully instrumented, in which GaInSn is used as model fluid. The circuit is inserted in a large dipole magnet that is strong enough to reach flow parameters close to those in ITER applications. Electric potential distributions are detected on the surface of the test section and in the liquid metal by means of a movable 4-pole potential probe. Detection of time-dependent signals allows identifying the critical Reynolds number for the onset of instabilities. Time-averaged measured data gives further information about the mean velocity distribution in the duct. Those results can be directly compared with numerical simulations, which support the physical interpretation of the observed MHD phenomena. [ABSTRACT FROM AUTHOR]

In liquid metal blankets the interaction of the moving breeder with the intense magnetic field that confines the fusion plasma results in significant modifications of the velocity distribution and increased pressure drop compared to hydrodynamic flows. Those changes are due to the occurrence of electromagnetic forces that slow down the core flow and which are balanced by large driving pressure heads. The resulting magnetohydrodynamic (MHD) pressure losses are proportional to the electric current density induced in the fluid and they can be reduced by electrically decoupling the wall from the liquid metal. For applications to dual coolant blankets it is foreseen to loosely insert electrically insulating liners into the ducts. In long channels the insulation could consist of a number of shorter inserts, which implies a possible local interruption of the insulation. Three dimensional numerical simulations have been performed to investigate MHD flows in electrically well-conducting channels with internal discontinuous insulating inserts. The local jump in the electric conductivity of the duct wall results in induced 3D electric currents and related electromagnetic forces yielding additional pressure losses and increased velocity in boundary layers parallel to the magnetic field. [ABSTRACT FROM AUTHOR]

An interesting option for plasma-facing components such as divertors seems to be the use of liquid–metal saturated capillary porous structures, on which a stable liquid surface could be obtained by capillary forces. The aim of the present study is providing a first description of liquid metal flows in a capillary porous system (CPS) when an external uniform magnetic field is imposed, in order to judge the feasibility, from the magnetohydrodynamic (MHD) point of view, of the CPS technology for steady-state divertor applications. Numerical results show that, independently of the orientation of the magnetic field, the permeability strongly reduces in porous MHD flows compared to those in the corresponding hydrodynamic Darcy regime. Main parameters that influence MHD flows in CPS are identified. [ABSTRACT FROM AUTHOR]

Highlights: [•] Liquid metal flows in two coupled breeder units of a HCLL TBM have been investigated by asymptotic analysis and numerical simulations. [•] Pressure drop at the first wall gap may reach a similar order of magnitude as the one along the radial extension of the module. [•] Pressure drop strongly depends on the radial distribution of the magnetic field and its value at the first wall. [•] Electric potentials along the side walls, i.e. along the stiffening plates, depend also significantly on the variation of B. [•] The distribution of flow rates remains practically unchanged in comparison with the case of a uniform magnetic field. [Copyright &y& Elsevier]

Abstract: A liquid metal blanket concept using a PbLi alloy as breeder material and helium as coolant is one of the proposed designs that will be tested in the experimental fusion reactor ITER. In order to finalize the design of the helium cooled lead lithium (HCLL) blanket, studies are still required to fully understand the behavior of the electrically conducting breeder under the influence of the intense magnetic field that confines the fusion plasma. Liquid metal HCLL blanket flows are expected to be mainly driven by buoyancy forces caused by non-isothermal operating conditions due to neutron volumetric heating, since only a weak forced flow is required for tritium extraction. The present numerical study aims at clarifying the influence of electromagnetic and thermal coupling of neighboring fluid domains on magneto-convective flows in geometries relevant for the HCLL blanket concept. Velocity and temperature distributions are discussed for various volumetric heat sources. [Copyright &y& Elsevier]

Abstract: The helium cooled lead lithium (HCLL) blanket concept is one of the designs that will be tested and validated in the experimental reactor ITER. The liquid metal, which is slowly circulated in the blanket, interacts with the magnetic field that confines the fusion plasma leading to modifications of velocity and pressure distributions compared to the ones in hydrodynamic flows. Experiments have been performed to investigate magnetohydrodynamic flows in a scaled mock-up of a HCLL blanket. One of the aims is identifying the geometric elements that give rise to the main contributions to the total pressure head in the module. Parametric studies have been carried out to assess the influence of flow rate and magnetic field intensity on pressure drops. [Copyright &y& Elsevier]

Abstract: European test blanket modules (TBM) for both European blanket concepts, namely the Helium Cooled Lithium Lead (HCLL) and the Helium Cooled Pebble Bed (HCPB), and their auxiliary systems have been designed in the last years. This paper presents the experimental facilities available in Europe for the qualification tests of European TBMs. In particular, devices designed to analyze specific technological issues, for example corrosion under strong magnetic fields or tritium management and extraction, as well as large scale facility to perform integral tests are described. A preliminary overview on the qualification strategy is presented and discussed. [ABSTRACT FROM AUTHOR]

Abstract: Magnetohydrodynamic experiments in a scaled mock-up of a helium-cooled lead lithium test blanket module designed for being tested in ITER have been performed in the MEKKA liquid metal laboratory at the Forschungszentrum Karlsruhe. The mock-up consists of four breeder units connected by feeding and draining manifolds. Each breeder unit contains five internal walls simulating the helium-cooled plates which remove in the original design the heat from the breeding zone. Due to the electric conductance of all walls the whole structure represents a complex coupled system, since currents induced by the motion of the fluid under the influence of the strong plasma-confining magnetic field can cross the walls and couple electrically neighboring fluid domains. The present work is focussed on the experimental study of flow partitioning in subchannels of breeder units, depending on flow rate and strength of the applied magnetic field. For strong magnetic fields one can observe that the flow distribution is similar as assumed in the design. In accordance with theoretical predictions the outer subchannels of breeder units show higher velocities than those in central ducts. This is a consequence of the established strong coupling of flow domains inside breeder units and by the fact of a weak coupling of neighboring breeder units. With increasing flow rate or decreasing magnetic field uneven flow distributions and even flow reversal are observed. Such unfavorable flow regimes, however, occur only at parameters that are out of scope of normal ITER operation. [Copyright &y& Elsevier]

Abstract: One of the blanket concepts proposed to be tested in ITER as part of the test blanket module program of the European Union is the helium cooled lead lithium blanket design. In this configuration the so called breeder units are arranged in an array, separated by a stiffening grid, to form blanket modules. The deposited thermal energy is removed by helium flowing at high pressure and speed in channels integrated both in the walls and in cooling plates that subdivide the breeder units into flat ducts where the lead lithium circulates under the influence of the strong plasma confining magnetic field. This gives rise to magnetohydrodynamic (MHD) phenomena whose effects on flow distribution have to be investigated to evaluate the performance of the proposed design. The established MHD flow is affected by the presence of helium channels in cooling and stiffening plates that results in non-homogeneous wall conductance. In support to the conceptual study of a liquid metal blanket, numerical investigations of fully developed MHD flows in a central cross-section of breeder units have been performed, taking into account both the presence of helium channels in the walls and the multi-channel effects caused by the exchange of currents through walls separating different fluid domains. [Copyright &y& Elsevier]

Abstract: In the helium-cooled lead lithium (HCLL) blanket, which has been chosen as a reference concept for a liquid-metal breeding blanket to be tested in ITER, the heat is removed by helium cooled plates aligned with the strong toroidal magnetic field that confines the fusion plasma. The liquid breeder lead lithium circulates through gaps of rectangular cross-section between the cooling plates to transport the generated tritium towards external extraction facilities. Under the action of the strong magnetic field, liquid metal flows in conducting rectangular ducts exhibit jet-like velocity profiles in the thin boundary layers near the side walls, which are parallel to the magnetic field like the cooling plates in HCLL blankets. The velocity in these side layers may exceed several times the mean velocity in the duct and it is known that these layers become unstable for sufficiently high Reynolds numbers. The present paper summarizes experimental results for such unstable time-dependent flows in strong magnetic fields, which have been obtained in the MEKKA liquid metal laboratory of the Forschungszentrum Karlsruhe. In particular, spatial and temporal scales of perturbation patterns are identified. The results suggest that the flow between cooling plates in a HCLL blanket is laminar and stable. The observed time-dependent flow behavior appears at larger velocities so that the present results are more relevant for applications in dual coolant concepts where high-velocity jets have been predicted along side walls. [Copyright &y& Elsevier]

• Features of 2D and 3D MHD flows typical in liquid metal blankets are reviewed together with available correlations for MHD pressure losses. • Design MHD guidelines that support a preliminary estimate of MHD effects in a blanket concept, in terms of pressure drop and flow distribution, are discussed. • Considerations on MHD issues for blanket applications focus on present status of the analysis of major problems and on future research needs. • Description of experimental methodology to be followed in order to produce experimental data for code validation purposes. The design of breeding blankets represents the major challenge for fusion reactor engineering because of performance requirements and severe operating conditions in terms of heat load and neutron flux. Liquid metal alloys such as lead-lithium, due to their lithium content, can be used to breed tritium, one of the plasma fuel components, and owing to their high thermal conductivity, they may serve as coolants. On the other hand, there are technical issues related to the fact that the liquid metals are electrically conducting and interact with the plasma-confining magnetic field. Induced electric currents and generated electromagnetic forces affect velocity and pressure distribution in the blankets. Magnetohydrodynamic (MHD) flows for fusion applications have been often investigated in simplified geometries, such as pipes, ducts, bends, with focus on their fundamental features. These analyses are essential, since results remain valid as background for the development of blanket designs, even when a concept is dismissed. However, the conceptual study of fusion blankets requires to take into account the global multiple effects, that arise when the full system is considered. Progress made in fusion-related MHD research results from combined numerical and experimental activities. In this paper we review and summarize features of 2D and 3D MHD flows that are typical in liquid metal blankets, together with available correlations for MHD pressure losses. This knowledge can provide simple design MHD guidelines that support a preliminary estimate of MHD effects in a blanket concept, in terms of pressure drop and flow distribution. [ABSTRACT FROM AUTHOR]

Abstract: For experimental determination of velocities in magnetohydrodynamic (MHD) liquid metal flows, probes that measure electric potential gradient are often used. They are known as Liquid-metal Electromagnetic Velocity Instruments (LEVI) or as Conduction Anemometers (CA) and they have been used in the past preferentially for investigating more or less fully developed flows in poorly conducting ducts and flows with smooth variations along the channel axis. For such applications, where electric current density is negligible, the probe was assumed to give reliable results since the potential gradient signals can be directly interpreted as a velocity measure. If the flow varies along its path on very short length scales, like in ducts with abrupt change of cross-section or in manifolds, 3D electric currents may occur that are not negligible any more so that the LEVI readings may become inaccurate. Moreover the presence of the probe itself may introduce already strong perturbations to the flow field since internal layers, which touch the shaft and body of the probe at both sides develop along magnetic field lines. This effect perturbs the flow not only locally in the vicinity of the probe tips but also along an entire region located between the internal layers. The present paper aims at quantifying these effects in order to obtain reliable velocity data. [Copyright &y& Elsevier]

Abstract: A modular helium cooled lead lithium (HCLL) blanket concept, in which helium is used to cool the breeder zone through cooling plates immersed in the liquid metal as well as the whole structure (first wall and stiffening plates), has been selected as part of the test blanket module (TBM) program of the European Union. In this blanket concept there are issues related to the magnetohydrodynamic (MHD) interaction of the moving liquid metal with the magnetic field that have to be addressed in order to assess the feasibility of this blanket design. Since the walls and the cooling plates are electrically conducting, an exchange of electric currents may occur that leads to an electric coupling of the flow in adjacent fluid domains. Fully developed MHD flows have been investigated numerically in four breeder units containing cooling plates. Results obtained for a pure toroidal magnetic field and for fields having a certain inclination with respect to the toroidal direction are summarized in this paper focusing on current and velocity distribution. [Copyright &y& Elsevier]

Abstract: The available experimental database for three-dimensional magnetohydrodynamic flows in strong magnetic fields, as required for applications to nuclear fusion blankets, is very limited. In order to broaden the knowledge at relevant parameters, experiments have been performed to investigate such flows through a sudden expansion of rectangular ducts at high Hartmann numbers. In accordance with the design for helium cooled lithium lead blankets the duct walls are electrically conducting. Experimental results for pressure distribution along the duct walls have been recorded and compared with results from asymptotic theory. More than 300 electric potential probes have been used to measure the potential distribution on the surface of the ducts. At the expansion the additional pressure drop due to 3D effects caused by the flow redistribution and by the occurrence of recirculating additional electric currents has been identified. The measurements of surface potential confirm the theoretical predictions that at the expansion a large fraction of the flow bypasses the core by flowing in thin layers along the side walls. [Copyright &y& Elsevier]

Abstract: Three-dimensional magnetohydrodynamic flows in sudden expansions of rectangular channels with walls of finite electric conductivity have been investigated. The used numerical tool, based on an extended version of the commercial code CFX, gives accurate results up to Hartmann numbers . For stronger magnetic fields an asymptotic-numeric approach has been used. The results of the study show that for moderate magnetic fields vortical recirculations form behind the expansion. They are, however, suppressed for sufficiently high Hartmann numbers. Moreover, regions with very small velocities may appear. The analysis highlights also that inertial effects become smaller with increasing the interaction parameter N, but even for large N an inertial influence is still present. The additional pressure drop caused by the three-dimensional flow at the expansion and the flow partitioning between cores, internal and side layers have been analyzed in detail. Some of the main outcomes are summarized in this paper. [Copyright &y& Elsevier]

Abstract: In the helium-cooled lithium lead (HCLL) blanket, liquid metal circulation is primarily required for tritium removal. Larger flow rates are desirable in order to keep the tritium permeation losses low. Because of electrically conducting structures, pressure drop becomes already large at small flow rates. Moreover, tritium permeation depends sensitively on velocity distributions in the ducts; MHD issues are, therefore, also relevant for the HCLL. In the paper, first the MHD specific flow features of the HCLL blanket are outlined and the status of pressure drop assessments is reported. Then, results of analyses are presented for velocity distributions influenced by flow coupling effect and thermal convection flows. [Copyright &y& Elsevier]

Abstract: Magnetohydrodynamic flows in a helium cooled lead lithium blanket have been analyzed by combined asymptotic-numeric computations. The flow has been considered in real 3D geometries. Of special concern are the connections between the circular access tubes with the breeder boxes through poloidal manifolds and distributing gaps, and the region near the first wall, where the fluid has to pass another narrow gap. For application to a DEMO-type reactor pressure drops of the order of 0.28 MPa have been estimated for an applied magnetic field of 10 T. [Copyright &y& Elsevier]

This paper describes an advanced fusion power reactor based on advanced plasma physics assumptions and large technological extrapolation compared with present-day knowledge. In-vessel components are based on the use of SiCf/SiC composite''s structure and the use of high-temperature Pb–17Li both as coolant and breeder for the blanket. The use of high-temperature super-conducting coils is also expected. A net electrical output of 1500 MWe is assumed. Tritium breeding self-sufficiency and acceptable operating parameters have been obtained. The resulting conceptual power plant has the potential for a thermal efficiency as high as 61% and for reaching good safety standard. The technological extrapolations assumed in this study provide an indication of the necessary R&D for addressing the most critical issues. [Copyright &y& Elsevier]

The major part of the magnetohydrodynamic pressure drop in the European water cooled concept for a fusion blanket arises in circular pipes which distribute the liquid metal breeder among the poloidal containers. These channels are surrounded by a massive structure of electrically conducting ferromagnetic material. This structure acts like a magnetic shielding for moderate external magnetic fields and reduces pressure drop. For fusion relevant conditions the magnetic shielding is not perfect since the wall material reaches magnetic saturation. The most interesting regime exists near magnetic saturation where curvilinear field lines inside the pipe are observed. They result in velocity profiles, which differ from the well known classical solutions. [Copyright &y& Elsevier]

The EVOLVE (EVaporation Of Lithium and Vapor Extraction) concept is a recently proposed liquid metal blanket. In EVOLVE a significant fraction of the fusion power is released in the liquid by nuclear volumetric heating and extracted from the fluid though a number of thin vapor channels by evaporation of lithium. The flow in the liquid phase under the influence of a strong, uniform, horizontal magnetic field is analyzed and the shape of vapor channel cross sections is determined. Vapor cross sections remain circular for strong surface tension but they may become highly elongated along magnetic field lines for strong magnetic fields. [Copyright &y& Elsevier]

Liquid metal magnetohydrodynamic pipe flow experiments have been performed in a non-uniform strong magnetic field in order to get further insight into the physics of this type of three-dimensional MHD flow and to enlarge the database for validation of numerical tools. In these experiments, in addition to measurements of the axial pressure distribution, the electric potential has been recorded at a large number of axial and circumferential positions by probes distributed around the periphery of the pipe. Since the wall of the pipe is relatively thick, the electric potential varies across the wall thickness, and surface potentials differ from those at the fluid-wall interface. However, the dense population of potential sensors on the pipe surface allows for a reconstruction of data on the fluid-wall interface, from which we can determine components of potential gradient in the fluid in a plane perpendicular to the magnetic field. The latter ones are closely related to the velocity field, because the electric potential may serve as approximate streamfunction of the flow. [ABSTRACT FROM AUTHOR]

Abstract: Magnetohydrodynamic experiments in a scaled mock-up of a helium-cooled lead lithium blanket module have been performed with the aim of determining the liquid–metal flow distribution inside breeder units and between sub-channels formed by the cooling plates. It has been observed that, even for strong magnetic fields, the distribution of the flow among these sub-channels is not completely uniform, i.e. inertial entrance and exit conditions still affect the flow partitioning in the breeder units. Moreover, depending on the balance between inertia, viscous and electromagnetic forces namely on Reynolds and Hartmann numbers, regions with almost stagnant or reversed flow can be observed. Since, in closed recirculation loops the accumulated tritium could represent a safety issue, their occurrence is investigated by a detailed parametric study. [Copyright &y& Elsevier]

• 3D mixed MHD convective flow has been simulated in poloidal ducts of DCLL blankets. • In rear duct colder fluid moves downwards as a fast stream along the internal wall. • In the jet the fluid removes a significant amount of heat in a half of the duct. • In the other half of the rear channel a closed recirculation is present. • For moderate B the recirculation extends along the full poloidal length of the duct. The dual coolant lead lithium (DCLL) blanket is one of the liquid metal blanket concepts investigated within the framework of the EUROfusion Consortium. In the proposed design, the liquid metal serves as tritium breeder, neutron multiplier, and as coolant for the breeding zone. Magnetohydrodynamic (MHD) effects resulting from the interaction of the moving liquid metal with the magnetic field that confines the fusion plasma are coupled with thermal phenomena. Induced electromagnetic forces lead to significant modifications of the hydrodynamic flow and affect in turn heat and mass transfer. The presence of a high-intensity neutron flux in the liquid metal yields volumetric heating and associated buoyant forces that drive complex flow patterns in the blanket. The resulting convective motions cause strong thermal mixing, and an increment in the effective heat transfer coefficient that may increase heat losses from the liquid metal into the helium coolant. This could deteriorate the thermal efficiency of the blanket. Buoyant MHD phenomena may also result in reversed flow and closed recirculations where tritium accumulates and temperature locally increases. The aim of the present investigation is to address by means of 3D numerical simulations the main features of magneto-convective flows in a geometry typical for DCLL blankets, in which PbLi moves upwards in a poloidal duct that faces the first wall and, after making a U-turn, flows downwards at the back of the blanket module. The liquid metal enters and leaves the model geometry by flowing radially in rectangular channels. It is found that in the front duct the velocity is characterized by jets in the boundary layers along walls parallel to the magnetic field, as expected in case of electrically conducting ducts. In the rear channel, along the wall that separates the two poloidal ducts, a downward flow occurs with much higher velocities than in the rest of the channel. The influence of the magnetic field on these flow features is investigated. [ABSTRACT FROM AUTHOR]

• Liquid metal flow in non-uniform magnetic field shows 3D magnetohydrodynamic effects. • Consistent and conservative formulation for spatially varying magnetic field. • Incoming turbulent flow transforms into magnetohydrodynamic laminar steady-state. • Thick pipe walls are numerically resolved for computation of electric potential. Motivated by experiments currently under investigation in the MEKKA laboratory at KIT, numerical simulations have been performed for a liquid metal pipe flow entering a strong magnetic field. Magnetohydrodynamic (MHD) flows in non-uniform magnetic fields are relevant in blanket engineering, in regions where the liquid breeder PbLi enters or exits the magnetic field confining the fusion plasma. Simulations provide insight into 3D phenomena near the non-uniform region of the magnetic field in particular velocity distribution and electric current pattern that cannot be directly observed in the experiments. Moreover, simulations describe the transition from a highly turbulent hydrodynamic entrance flow to a fully established laminar MHD flow. The present work addresses in particular a conservative formulation of the magnetic field, generation and decay of inlet turbulence by means of direct numerical simulation, as well as fluid–wall interaction by electric currents that are exchanged between the liquid metal and the thick electrically conducting pipe wall. [ABSTRACT FROM AUTHOR]

Insulating flow channel inserts (FCI) in ducts of liquid metal blankets decouple electrically the fluid domain from the conducting thick-walled structure. They interrupt current paths across well-conducting walls and thereby reduce significantly magnetohydrodynamic (MHD) pressure drop. Pipe flow of an electrically conducting liquid metal entering a region where walls are covered by a FCI is investigated experimentally under the influence of a strong uniform magnetic field. The abrupt change in wall conductivity leads to 3D MHD effects that are studied by measuring pressure and electric potential distribution along the duct wall. It is shown that FCIs may reduce pressure drop by at least one order of magnitude but 3D effects at the entrance of FCIs may reduce the efficiency of the insulating inserts by some degree. [ABSTRACT FROM AUTHOR]

Experiments have been performed in the MEKKA laboratory of the Karlsruhe Institute of Technology to characterize the influence of a magnetic field on liquid metal flows in a scaled mock-up of the water-cooled lead-lithium test blanket module of ITER. The test section consists of eight breeder units that are fed and drained by long electrically coupled manifolds oriented along the poloidal direction. Pressure differences between several points of the mock-up have been recorded for various liquid metal flow rates and strengths of the imposed magnetic field. The main contributions to the total pressure drop in the test section have been identified as a function of characteristic flow parameters. For sufficiently strong magnetic fields, the experimental data shows that the non-dimensional pressure loss is practically independent of the flow rate, i.e. inertia forces become negligible. The experiments also confirm previous conclusions of magnetohydrodynamic experiments performed in a scaled-down mock-up of a helium-cooled lead-lithium blanket, namely that the largest pressure drop in the blanket module originates from the flow in the distributing and collecting manifolds. Moreover, the experimental study demonstrates that the current manifold design does not allow the flow to be uniformly distributed among all the breeder units. [ABSTRACT FROM AUTHOR]

New manufacturing methods for the production of key components for nuclear fusion reactors by selective laser melting (SLM) are currently under investigation at Karlsruhe Institute of Technology. SLM offers great potential compared to conventional manufacturing methods. Within the framework of feasibility studies, complex 3D structures, such as a thin- and double-walled flow channel inserts (FCIs) for dual-coolant lead lithium (DCLL) blankets, were successfully manufactured and subjected to preliminary tests. The paper shows that fabrication of thin-walled FCIs by the SLM technique is feasible in principle. The complexity of the fabrication process is outlined and application of the SLM technique for production of other fusion blanket sub-components, such as thin-walled cooling plates with internal cooling channels, is addressed. [ABSTRACT FROM AUTHOR]

• The preliminary Pb15.7Li Loop layout was shown in the paper. • The Magneto Hydro Dynamic effect induced in the flowing Pb-16Li inside BB and the pressure drops generated were analysed. • The developed anti-permeation and corrosion barriers inside the BB where shown. • The code developed to analyse the In-box Loca inside WCLL BB was introduced. • Preliminary integration of Pb-16Li loops inside the tokamak building was shown. Three of the four breeder blanket concepts currently under investigation for the European DEMO Reactor use the eutectic Pb-16Li as breeder material. Those are the Helium Cooled Lithium Lead (HCLL), Water Cooled Lithium Lead (WCLL) and Dual Coolant Lithium Lead (DCLL) blankets. Moreover, the WCLL is one of the blanket concepts that will be qualified in the ITER reactor, therefore the development and design of lead lithium loops and auxiliary systems is essential. The main functional requirements that Pb-16Li systems have to fulfill are: • to circulate the liquid Pb-16Li through the blanket and ancillaries; • to extract the tritium produced inside the breeder modules from Pb-16Li; • to control Pb-16Li chemistry and to remove accumulated impurities; The present work aims to describe the activities performed in order to achieve the following objectives: i) design and integration of the Pb-16Li loops inside the tokamak building, ii) development and characterization of antipermeation and anticorrosion coatings on structures in contact with Pb-16Li, iii) development and design of an activation products removal system, iv) design of a chemistry control system for Pb-16Li loops, v) performing magnetohydrodynamic analyses taking into account the impact on heat transfer and tritium transport in breeding blankets and performing safety analyses of water/Pb-16Li interaction due to LOCA inside the WCLL blanket. [ABSTRACT FROM AUTHOR]

We propose a new activity on verification and validation (V&V) of MHD codes presently employed by the fusion community as a predictive capability tool for liquid metal cooling applications, such as liquid metal blankets. The important steps in the development of MHD codes starting from the 1970s are outlined first and then basic MHD codes, which are currently in use by designers of liquid breeder blankets, are reviewed. A benchmark database of five problems has been proposed to cover a wide range of MHD flows from laminar fully developed to turbulent flows, which are of interest for fusion applications: (A) 2D fully developed laminar steady MHD flow, (B) 3D laminar, steady developing MHD flow in a non-uniform magnetic field, (C) quasi-two-dimensional MHD turbulent flow, (D) 3D turbulent MHD flow, and (E) MHD flow with heat transfer (buoyant convection). Finally, we introduce important details of the proposed activities, such as basic V&V rules and schedule. The main goal of the present paper is to help in establishing an efficient V&V framework and to initiate benchmarking among interested parties. The comparison results computed by the codes against analytical solutions and trusted experimental and numerical data as well as code-to-code comparisons will be presented and analyzed in companion paper/papers. [ABSTRACT FROM AUTHOR]

The scope of the EUFOfusion Work Package Breeding Blanket is to develop a blanket concept for the EU DEMO reactor; this includes the blanket segments inside the Vacuum Vessel and the related Tritium Extraction/Removal Systems. In the Pre-Concept Design (PCD) Phase, two concepts have been selected as candidates; a solid and a liquid breeder blanket cooled with helium and water, respectively. The design of these two blanket systems has been adapted to the DEMO plant design developed in the PCD Phase and performances assessed. A large R&D programme has been implemented with the scope to evaluate different technologies for these blankets; including the development of breeders, tritium extraction and cooling technologies, and the manufacturing of the blanket system. A major milestone in the subsequent Concept Design Phase is the final selection of the blanket concept for DEMO. [ABSTRACT FROM AUTHOR]

Abstract: This paper describes the work performed for the power plant conceptual studies (PPCS) reactor model AB, a ‘near-term’ fusion power reactor based on limited extrapolations from today knowledge on technology and plasma physic assumption, featuring a helium-cooled lithium lead (HCLL) blanket concept and a He cooled divertor. The main design rational and characteristics of the PPCS model AB as well as the main assumption and achieved results are reported. The reactor is able to supply to the grid a net electrical power of 1.5GWe for a fusion power of 4.2GWth and features a major radius of 9.56m. A tritium breeding ratio (TBR) of 1.13 will assure the tritium self sufficiency. [Copyright &y& Elsevier]