- Email: [email protected]
Overview of the last progresses for the European Test Blanket Modules projects
Fusion Engineering and Design 82 (2007) 2105–2112
Overview of the last progresses for the European Test Blanket Modules projects J.-F. Salavy a,∗ , L.V. Boccaccini b , R. L¨asser c , R. Meyder b , H. Neuberger b , Y. Poitevin c , G. Rampal a , E. Rigal d , M. Zmitko c , A. Aiello e a CEA Saclay, DEN/DM2S, F-91191 Gif-sur-Yvette, France Forschungszentrum Karlsruhe, P.O. Box 3640, D-76021 Karlsruhe, Germany c EFDA, Close Support Unit, D-85748 Garching, Germany CEA Grenoble, Direction de la Recherche Technologique, F-38054 Grenoble, France e ENEA Brasimone, P.O. Box 1, I-40032 Camugnano (Bo), Italy b
d
Received 23 August 2006; received in revised form 12 February 2007; accepted 12 February 2007 Available online 21 March 2007
Abstract This paper gives an overview of the last progresses in terms of system design, calculations, safety and R&D done these last 2 years for the reference Test Blanket Modules developed in Europe, namely the Helium Cooled Lithium-Lead (HCLL) breeder blanket and the Helium Cooled Pebble Bed (HCPB), in order to cope with the ambitious objective to introduce two EU TBM systems for day-1 of ITER operation. The engineering design of the two systems is mostly concluded and the priority is now on the development and qualification of the fabrication technologies. From calculations point of view, the last modelling efforts related to the thermal–hydraulic of the first wall, the tritium behaviour, and the box thermal and mechanical resistance in accidental conditions are presented. Last features of the TBM and cooling system designs and integration in ITER reactor are highlighted. In particular, this paper also describes the safety and licensing analyses performed for each concept to be able to include the TBM systems in the ITER preliminary safety report (RPrS). Finally, overview of recent R&D progresses in fabrication, tritium experiments and test facilities development is given. © 2007 Elsevier B.V. All rights reserved. Keywords: ITER; HCLL; HCPB; Test Blanket Modules
1. Introduction
∗ Corresponding author at: CEA/Saclay, DEN/DM2S/SERMA/ LCA, Bˆat 470, 91191 Gif sur Yvette, CEDEX, France. Tel.: +33 1 69087179; fax: +33 1 69089935. E-mail address: [email protected] (J.-F. Salavy).
0920-3796/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.fusengdes.2007.02.017
The long-term objective of the EU Breeding Blankets program is the development of DEMO breeding blankets, which shall assure tritium self-sufficiency, an economically attractive use of the heat produced inside the blankets for electricity generation and a
2106
J.-F. Salavy et al. / Fusion Engineering and Design 82 (2007) 2105–2112
sufficiently high shielding of the superconducting magnets for long time operation. In the short-term socalled DEMO relevant Test Blanket Modules (TBMs) of these breeder blanket concepts shall be designed, manufactured, tested, installed, commissioned and operated in ITER for first tests in a fusion environment. The main objective of the EU test strategy related to TBMs in ITER is to provide the necessary information for the design and fabrication of breeding blankets for a future DEMO reactor. The two reference breeding blankets developed by Europe [1] are the Helium-Cooled Lithium-Lead (HCLL) blanket which uses the eutectic Pb–17Li as both breeder and neutron multiplier, and the HeliumCooled Pebble-Bed (HCPB) blanket which features lithiated ceramic pebbles (Li4 SiO4 or Li2 TiO3 ) as breeder and beryllium pebbles as neutron multiplier. Both blankets are using the pressurized He technology for the power conversion cycle (8 MPa, inlet/outlet temperature 300 ◦ C/500 ◦ C) and the 9% CrWVTa Reduced Activation Ferritic Martensitic (RAFM) steel as structural material, the EUROFER. Main features of the TBMs and circuits design, of the basic parameters and of the test strategy in ITER has been presented at SOFT23 in Venice 2004 [2] and at the ISFNT-7 in Tokyo [3]. This paper gives an overview of the last progresses done these last 2 years in terms of system design, calculations, safety and R&D.
2. TBMs design and system updates For HCLL TBM, a first set of modifications have been realized [4]. They first concerned specific functional features in order to better comply with the testing objectives and to improve TBM reliability, such as: Pb–17Li flow path, TBM instrumentation, Pb–17Li draining, Pb–17Li manifold and He manifolds. They were also some modifications related to fabrication issues. Actually, after an industrial expertise on the In-TBM preliminary design to identify fabrication and mounting sequence issues, some following design guidelines have been adopted: avoid welding triple points on the junction first wall/stiffening grid, suppress sharp points on some welding trajectories, avoid possible interference between welding beams and welded parts (Back Plates/Side Walls), optimize thick plates welding. Finally, a preliminary version of the module
attachment system to the frame, assuming flexible fixation and gliding shear keys system, has also been realized. More recently, two main modifications have been further introduced in the HCLL TBM design. The first one concerns the addition of a by-pass tube to extract part of the He first wall (FW) flow-rate in order to be able to increase up to 500 ◦ C the cooling-plate (CP) outlet temperature. This feature is necessary because of the difference in terms of Neutron Wall Loading comparing DEMO operation (∼2.4 MW/m2 ) and ITER DT Phase (0.78 MW/m2 ). Moreover, the He cooling path has been reviewed following thermal–hydraulics studies performed for the HCLL DEMO design: the new reference cooling scheme is a full in-series cooling, FW first, then Stiffening Plates and covers, and CP at the end. For HCPB, the design presented in 2004 has been in these years continuously optimised and dedicated to the first, so-called EM-TBM, to be installed into ITER. The major change that has been done since 2004 is a modification of the breeding unit. The breeding canister has been redesigned and its orientation changed from horizontal to vertical. Instead of the double canister with straight channels, the canister is now made from two cooling plates connected by a wrapper. Some of the coolant channels in the plates run along the circumference, some are meandering in toroidal direction some in the radial one. Moreover, the question of instrumentation of the HCPB EM-TBM has also been addressed. There are sensors in the high pressure zone of the helium coolant, the low pressure zone in the purge gas system and in the vacuum vessel on the surface of the TBM. Electrical and pressure signals have to be transferred. No sensor line should be interrupted in the vacuum vessel, requiring a continuous line from the sensor through the vacuum vessel vacuum boundary to the plug where the signal is passed to the data acquisition system. Handling of lines of the required length must be proven in advance with dedicated tests. The sensor plugs are very space consuming; they will be mounted on the pipe integration cask in the port cell. If a sensor fails no repair is possible, so there is the conflict of request for a high redundancy and the limited space. To cope with this problem extensive out of ITER tests are needed in the TBM mock-ups planned to gather experience with the sensors to be used and the issue to be investigated.
J.-F. Salavy et al. / Fusion Engineering and Design 82 (2007) 2105–2112
The primary loop for the HCPB-TBM, mainly common with the HCLL one, has been investigated with regard to layout definition, selection and dimensioning of components including piping and mechanical integration of the circuit into the different sections of ITER. The accommodation of the main components of the Helium loop into the Torus Coolant Water System (TCWS) vault turned out as the most challenging point [5].
3. Calculations updates For each European TBM concept, significant improvements in thermal–hydraulics, mechanical and T permeation calculations methods have been realized. They can be summarized as follows. 3.1. HCLL calculations progresses ◦ Thermal–hydraulics analyses: new Finite Elements (FE) models of the blanket that allow to obtain the temperature distribution in the coolant on the basis of the detailed power density distribution in the surrounding structures have been developed. Improved models, heat transfer coefficient review, description of the calculations and main conclusions, showing in particular that cooling the FW, SPs and CPs in-series is a viable solution, are described in [6]. ◦ At the same time, the models allow to improve the description of the heat transfer between the solid walls and the coolant. A review of the most used relations available in literature to estimate the heat transfer coefficient has then been performed and the most appropriate has been selected. The new models have been used to carry out a thermal hydraulic analysis of the cooling system of the HCLL breeding blanket. Also, an estimation of the pressure drops in the different cooling elements has been made. Also, counter current flow in the FW can be adopted to reduce the overall thermal stresses in the structure. However, a careful re-engineering of the geometry of the He channels in the cooling plates is needed to avoid hot spots in the steel and achieve a more uniform He outlet temperature. ◦ Mechanical analyses [7]: in order to validate the last improvements made to the design, several FEM analyses have been performed with CAST3M code on
2107
relevant portions of the TBM geometry such as a horizontal TBM slice, top cap and stiffening grid. For each component, results show a correct and admissible behaviour of this part of the structure regarding to the loads and criteria (linear analyses according to SDC-IC criteria verifications assuming 500 ◦ C as structure temperature). The attachment and supporting system of the TBM implemented in the design needs to be able to withstand the disruption loads and TBM weight, while allowing thermal expansions of the box. In order to verify and validate the behaviour of this system, a local model of the flexible cartridge and a global model of the back plates and shear keys were established. Stressed computed in these various parts of the attachment system show significant margins with regards to acceptable limits. ◦ Tritium permeation: a local finite element modelling of the tritium permeation rate through the HCLL breeder unit cooling plates has been developed and is presented in [8]. ◦ Safety and licensing: preliminary thermal and mechanical analysed in accidental transients have been performed but need to be updated due to recent precisions in ITER safety strategy related to TBM operations. In parallel, in view of preparing the licensing of the HCLL TBM for ITER, a first assessment of the impact of the TBM on its environment from a safety viewpoint during the different phases of its lifespan (operation, maintenance, disposal, accident) has been performed [9]. Based on activation calculations and envisaged accident scenarios which need to be modified, the results of this first study aiming at evaluating the impact of HCLL TBMs upon several aspects: releases, dose rates and waste, have not identified killing issues. 3.2. HCPB calculations progresses ◦ Thermal–hydraulics analyses: several models have been developed in the last years for the thermohydraulics analysis of the TBM and TBM systems: RELAP 5 model for the whole helium systems, % TBM model in ANSYS for the study of the temperature distribution, and detailed models with STAR CD of the single cooling channels and manifolds. Several calculations have been performed with these models to optimise the design of the helium cooling system and to select the most suitable working point.
2108
J.-F. Salavy et al. / Fusion Engineering and Design 82 (2007) 2105–2112
A particular effort has been dedicated on the thermohydraulic analysis of the first wall. The cooling of this component is critical due to an extreme one sided heat flux. Unexpected high radial temperature differences were found and asked for better cooling conditions. More details of this assessment can be found in [10]. The flow distribution in the manifolds was investigated both for the individual cooling channels and for the manifolds. It turned out that the mass flow deviations, without any adjustment are within 10–15% of the design values. ◦ Mechanical analyses: in order to validate the last improvements made to the design, several FEM analyses have been performed with ANSYS program on relevant portions of the TBM geometry such a quarter of the TBM, the manifold region, the attachment system, and a complete BU units. The mesh has been generated directly from the CAD drawings. The assessment according to the SDC-IC criteria verifications shows a correct and admissible behaviour of the structure regarding to the loads in operation and fault conditions. Example of this calculation for the first wall can be founded in [10]. The model used for the attachment system covers also the manifolds to allow for formulation of reasonable boundary conditions for the analysis. Dimensioning of the montage gap between the shear keys of the attachment is a crucial issue, to get more insight in this question a detailed investigation is under way with regard to find a compromise in between minimization of the gap to reduce the dynamic loads during a plasma disruption and thermal stresses occurring in the attachment system during operation. So far it was found that the temperature distribution in the manifolds lead to a bowing of the TBM back plate having negative effects on the contact zone of the opposing stub keys. As a general result we found that the thermal stresses in the attachment are of the same order as those due to disruption loads. ◦ Safety and licensing: the Preliminary Safety Analysis Report for the HCPB TBM is under preparation to get the construction permit first and then the operation licence later. In general, the reduced inventory of activation products and tritium associated with the TBM system makes its impact almost negligible to the overall safety risk of ITER. Release of radioactive effluents and radiation exposure of work-
ers connected to the operation of this system are under assessment, but are considered almost uncritical. This consideration can be extended also to the safety assessment of the TBM system related to accidental conditions. Nevertheless, the possibility to jeopardise the ITER safety concept has been analysed in connection to the consequences of specific accident sequences, e.g. the pressurisation of the Vacuum Vessel (VV) due to the He coolant blow-down, the hydrogen production from some breeder/multiplier-steam reactions, the possible interconnection between Port Cell and Vacuum vessel causing air ingress, and the necessity to assure the heat removal in the short and long period. A summary of preliminary studies on accidental sequences can be found in [11].
4. R&D progresses This section only refers to some particular features of the very extensive European R&D program runs these recent years for the two breeding blanket concepts. It concerns progresses in fabrication, tritium control experiments and test facilities development. 4.1. Subcomponent fabrication Each of the main European TBM subcomponents possesses internal cooling channels with a rectangular cross section. This feature makes them difficult to manufacture, even using advanced technologies such like Hot Isostatic Pressing-Diffusion Welding (HIP-DW). Basically, HIP-DW consists in applying a high gas pressure at high temperature to a stack of parts in order to achieve bonding between the surfaces. Gas must be prevented from entering gaps between parts. This is achieved by welding the periphery of the joints or by using a canister. With this process, very good joint properties can be achieved, especially with Eurofer [12]. Complex shapes can be achieved as well, as explained hereafter. A component with round embedded channels can be manufactured by figuring the channels thanks to round tubes, providing the channel bending radii are not too small. In this case, the ends of the tubes are welded to the canister in such a way that the gas pressure enters the channels without entering the joints, which prevents the channels to collapse.
J.-F. Salavy et al. / Fusion Engineering and Design 82 (2007) 2105–2112
Processes based on HIP-DW have been recently reviewed in view of their application to HCLL and HCPB TBMs subcomponents. As a result, prime candidate processes have been defined for each kind of subcomponent, as described in the following: - The first walls have rectangular channels. One proposed manufacturing process is based on the use of round tubes inserted between grooved plates [12–14]. During HIP the tubes expand and fit in the grooves. It appeared that this process is not reliable and not versatile enough because tubes may fail during HIP. One proposed solution is to expand the tubes at room temperature using hydraulic forming, then to insert the tubes in the grooves and apply HIP-DW. The use of thick rectangular tubes inserted between thin plates, as described in [12], is no more considered. - The stiffening plates and cooling plates have small, rectangular channels that must follow sharp turns, which makes impossible to use tubes. Two solutions are envisaged for these components, namely the two steps HIP [12–14] or a new process which consists in closing the top of machined grooves by welding thin strips, then adding a plate by HIP-DW. With this technique, the welds are encapsulated within the structure: the plate acts as a barrier preventing leakage between the coolant side and the breeder side. A 400 mm long HCLL cooling-plate mockup has been manufactured according to this process (Fig. 1). It is currently under thermomechanical testing in CEA/Cadarache. It is worth noting that the
2109
two above processes are applicable to the caps. The welding + HIP process is also, in principle, applicable to the first wall, despite its curvature makes it more difficult. An alternative procedure for the manufacturing of cooling plates is also under investigation. The plates are produced by diffusion welding in a press device, starting from two grooved half plates. The process has been successfully tested in laboratory conditions; at the present the transfer to industrial scale is on-going. 4.2. TBM assembly by welding The assembly of the HCPB and HCLL TBMs imply the development of sophisticated welding technologies and tools. In particular the following issues are a challenge: • Welding of thin (6–8 mm) to thick (35–40 mm) plates. • Access of tools in the internal grid structure (200 mm × 200 mm cells). • Avoid hot cracking (Eurofer like other RAFM steels is sensitive to hot cracking under high energy welding processes). • Limitation of the global deformation. • Optimization of the Post Welding Heat Treatment (PWHT) to retrieve the microstructure and mechanical properties of the joint and Heat Affected Zone (HAZ).
Fig. 1. HCLL cooling-plate mock-up tested in the CEA DIADEMO facility.
2110
J.-F. Salavy et al. / Fusion Engineering and Design 82 (2007) 2105–2112
EU has investigated the capabilities of the TIG and laser welding processes for assembling the TBM structure. Even if the final selection of a reference process will be done later on, the welding capability and strategy for both processes has been established. The final selection will come from optimization of the assembly deformation during fabrication and optimization of cost for tool development. TIG butt welding experiments on Eurofer samples have shown that no cracks are observed for welding plates relevant to the stiffening grid configuration (8 mm thickness). An optimization of the groove shape, process and electrode diameter has allowed avoiding lack of penetration of the root pass and minimization of distortion. Welds up to 150 mm length have been produced. The formation of delta ferrite and possible induced deviation phenomena will be further assessed for ensuring a full control of the process. Standard available laser tools are difficult to use under the geometrical constraints imposed by the TBM grid configuration (in particular for T welding). The standard volume of these tools would lead to perform T-shape welds with a non-negligible angle with regard to the perpendicular direction of the plates. For that purpose, the development of reduced size (14 mm diameter) or even more advanced laser tools are ongoing. Preliminary investigation has shown that standard YAG laser welding parameters applied on 8 mm thick
Eurofer samples result in apparition of hot cracking in the fusion zone of few samples, in particular in the first 50 mm of the weld length. This phenomenon is well known [15] and EDS and SIMS analyses have confirmed that no chemical segregation occurred around the crack, indicating that standard YAG welding parameters induce excessive stress level in the solidification zone (at high temperature, strength of the austenite phase is not sufficient to accommodate stress concentration flow for the this solidification time stage). A development program has been carried-out to eliminate hot cracking trouble. This program was based on a systematic investigation of parameters influencing hot cracking (welding travel speed, beam energy, beam spot profile, pre/post heat treatment, design adjustment, etc.). It was demonstrated that hot cracking can be avoided on Eurofer 8 mm thick plates (Fig. 2). Few pores (<0.5 mm diameter) were observed but experiments have shown that they were never at the origin of crack initiation during destructive examination. For the welding of thick walls (e.g. top/bottom covers, back plates) several processes are under investigation, from industrial processes like the Narrow Gap TIG welding operated in nuclear field to more advanced processes like the Laser MIG Hybrid process developed by CEA for several years in the frame of ITER Vacuum Vessel sector joining development.
Fig. 2. Adapted laser weld shape of Eurofer sample (8 mm thickness). No occurrence of hot cracking, presence of few pores (<0.5 mm).
J.-F. Salavy et al. / Fusion Engineering and Design 82 (2007) 2105–2112
4.3. Tritium control related experiments and test facilities Measurements of the diffusivity and the Sievert’s constants of the hydrogen isotopes protium and deuterium in liquid Pb–15.7 at.% Li, by means of LEDI and SOLE apparatuses [16], have been recently completed (ENEA-Brasimone) using the absorption technique. The hydrogen diffusivity was found to be about two orders of magnitude higher than previously selected reference literature data (Reiter’s data [17] obtained by desorption technique). Also, the values of the Sievert’s constant (a measure for solubility) are about one order of magnitude higher than stated in the literature. These data are very important for the HCLL breeding blanket concept having implications with respect to tritium control, tritium permeation and tritium extraction. If they were true, the tritium permeation problem in case of the HCLL concepts would be less severe as the driving force for permeation is reduced due to the higher solubility in Pb–Li. This is certainly true as long as Pb–Li flow is turbulent and the whole hydrogen inventory is available for permeation. In a strong magnetic field (e.g. in TBMs) where we expect to have laminar flow along the cooling plates, the reduction in permeation could get reversed by the two orders of magnitude higher diffusion coefficient. A repetition of the measurements of hydrogen diffusivity and Sievert’s constant in Pb–Li is currently planned with emphasis on comparison of the data to be obtained by desorption and absorption techniques and using an equipment with walls of extremely low hydrogen solubility. The construction of the TRIEX loop for study of tritium recovery from Pb–Li, to be operated in Brasimone (ENEA), has been completed. Details on loop components, operating parameters and test campaign are given in [18]. This facility will use Hydrogen (tritium) partial pressure sensors for measurement in Pb–Li. Recent developments of such sensors, using Nb and Fe Armco permeation membranes, are being tested in the temperature range of 400–550 ◦ C (Politecnico di Torino, ENEA-Brasimone).
5. Conclusions In order to cope with the ambitious objective to introduce two European TBM systems for day-1 of
2111
ITER operation, EU maintains a large effort in terms of design activities, modelling improvement and R&D development for the HCLL and HCPB blanket modules. Up to now, no killing issue has been identified in these various activities. This effort will be pursued in the framework of a European project whose technical plan has been detailed over the next 10 years [19].
Acknowledgements This work, supported by the European Communities under the contract of Association between EURATOM and CEA, was carried-out within the framework of the European Fusion Development Agreement. The views and opinions expressed herein do not necessarily reflect those of the European Commission.
References [1] L.V. Boccaccini, L. Giancarli, G. Janeschitz, S. Hermsmeyer, Y. Poitevin, A. Cardella, E. Diegele, Materials and design of the European DEMO blankets, J. Nucl. Mater. 329–333 (2004) 148–155. [2] Y. Poitevin, L.V. Boccaccini, A. Cardella, L. Giancarli, R. Meyder, E. Diegele, R. Laesser, G. Benamati, The European Breeding Blankets development and the test strategy in ITER, Fusion Eng. Des. 75–79 (2005) 741–749. [3] L.V. Boccaccini, J.-F. Salavy, R. L¨asser, A. Li Puma, R. Meyder, H. Neuberger, Y. Poitevin, G. Rampal, The European Test Blanket Module systems: design and integration in ITER, invited Paper to the ISFNT-7, Fusion Eng. Des. 81 (2006) 407–414. [4] G. Rampal, D. Gatelet, L. Giancarli, G. Laffont, A. Li-Puma, J.F. Salavy, E. Rigal, Design approach for the main ITER Test Blanket Modules for the EU Helium Cooled Lithium-Lead blankets, Fusion Eng. Des. 81 (2006) 695–700. [5] H. Neuberger, X. Jin, L.V. Boccaccini, B-E. Ghidersa, R. Meyder, Helium loop for the HCPB Test Blanket Module, this conference. [6] G. Aiello, F. Gabriel, L. Giancarli, G. Rampal, J.-F. Salavy, Thermal–hydraulic analysis of the HCLL DEMO Blanket, this conference. [7] G. Rampal, Design and analyses of the HCLL TBM including design of supporting system and instrumentation integration, CEA Report SEMT/BCCR/RT/06-004/A, September 2006. [8] F. Gabriel et al., A 2D finite element modeling of tritium permeation for the HCLL DEMO blanket module, this conference. [9] C. Lacressoniere, O. Gastaldi, TW2-TTBC-005-D02: TBM system detailed safety and licensing, CEA Technical note DTN/STPA/LPC/2005/059, 2005. [10] M.Ili´c, B. Dolensky, L.V. Boccaccini, R. Meyder, C. Polixa, P. Schanz, Thermo-hydraulics and mechanical design of the
2112
[11]
[12]
[13]
[14]
J.-F. Salavy et al. / Fusion Engineering and Design 82 (2007) 2105–2112 First Wall of the European HCPB Test Blanket Module, this conference. L.V. Boccaccini, R. Meyder, X. Jin, Analysis of three LOCAs for the European HCPB Test Blanket System, this conference. A. Cardella, E. Rigal, L. Bedel, P. Bucci, J. Fiek, L. Forest, L.V. Boccaccini, E. Diegele, L. Giancarli, S. Hermsmeyer, G. Janeschitz, R. L¨asser, A. Li Puma, J.D. Lulewicz, A. M¨oslang, Y. Poitevin, E. Rabaglino, The manufacturing technologies of the European breeding blankets, J. Nucl. Mater. 329–333 (2004) 133–140. T. Heider, K. Schleisiek, G. Reimann, Ch Dellis, E. Rigal, Development of manufacturing processes for the ITER heliumcooled pebble bed blanket test module, Fusion Eng. Des. 39–40 (1998) 819–824. E. Rigal, G. Le Marois, T. Lechler, G. Reimann, K. Schleisiek, L. Sch¨afer, P. Weimar, Development of FM steels diffusion
[15]
[16]
[17] [18]
[19]
bonding technologies for blanket manufacturing applications, Fusion Eng. Des. 49–50 (2000) 651–656. G. de Dinechin, C. Chagnot, Ph. Aubert, CEA Weldability Developments on RAFM Steels, Report CEA/DRT/LITEN/ UTIAC-04-RT, May 2004. A. Aiello, A. Ciampichetti, Measurement of Hydrogen Transport Parameters in Lead Lithium Alloy Using LEDI and SOLE devices, ENEA Report, LB-A-R-023, March 2005. F. Reiter, Solubility and diffusivity of hydrogen isotopes in liquid Pb–17Li, Fusion Eng. Des. 14 (1991) 207–211. A. Aiello, A. Ciampichetti, M. Utili, G. Benamati, The TRIEX experimental loop for tritium extraction systems qualification, this conference. Y. Poitevin, L.V. Boccaccini, G. Dell’Orco, E. Diegele, R. L¨asser, J.-F. Salavy, J. Sundstr¨om, M. Zmitko, The Test Blanket Modules project in Europe: from the strategy to the technical plan over next ten years, this conference.
Overview of the last progresses for the European Test Blanket Modules projects J.-F. Salavy a,∗ , L.V. Boccaccini b , R. L¨asser c , R. Meyder b , H. Neuberger b , Y. Poitevin c , G. Rampal a , E. Rigal d , M. Zmitko c , A. Aiello e a CEA Saclay, DEN/DM2S, F-91191 Gif-sur-Yvette, France Forschungszentrum Karlsruhe, P.O. Box 3640, D-76021 Karlsruhe, Germany c EFDA, Close Support Unit, D-85748 Garching, Germany CEA Grenoble, Direction de la Recherche Technologique, F-38054 Grenoble, France e ENEA Brasimone, P.O. Box 1, I-40032 Camugnano (Bo), Italy b
d
Received 23 August 2006; received in revised form 12 February 2007; accepted 12 February 2007 Available online 21 March 2007
Abstract This paper gives an overview of the last progresses in terms of system design, calculations, safety and R&D done these last 2 years for the reference Test Blanket Modules developed in Europe, namely the Helium Cooled Lithium-Lead (HCLL) breeder blanket and the Helium Cooled Pebble Bed (HCPB), in order to cope with the ambitious objective to introduce two EU TBM systems for day-1 of ITER operation. The engineering design of the two systems is mostly concluded and the priority is now on the development and qualification of the fabrication technologies. From calculations point of view, the last modelling efforts related to the thermal–hydraulic of the first wall, the tritium behaviour, and the box thermal and mechanical resistance in accidental conditions are presented. Last features of the TBM and cooling system designs and integration in ITER reactor are highlighted. In particular, this paper also describes the safety and licensing analyses performed for each concept to be able to include the TBM systems in the ITER preliminary safety report (RPrS). Finally, overview of recent R&D progresses in fabrication, tritium experiments and test facilities development is given. © 2007 Elsevier B.V. All rights reserved. Keywords: ITER; HCLL; HCPB; Test Blanket Modules
1. Introduction
∗ Corresponding author at: CEA/Saclay, DEN/DM2S/SERMA/ LCA, Bˆat 470, 91191 Gif sur Yvette, CEDEX, France. Tel.: +33 1 69087179; fax: +33 1 69089935. E-mail address: [email protected] (J.-F. Salavy).
0920-3796/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.fusengdes.2007.02.017
The long-term objective of the EU Breeding Blankets program is the development of DEMO breeding blankets, which shall assure tritium self-sufficiency, an economically attractive use of the heat produced inside the blankets for electricity generation and a
2106
J.-F. Salavy et al. / Fusion Engineering and Design 82 (2007) 2105–2112
sufficiently high shielding of the superconducting magnets for long time operation. In the short-term socalled DEMO relevant Test Blanket Modules (TBMs) of these breeder blanket concepts shall be designed, manufactured, tested, installed, commissioned and operated in ITER for first tests in a fusion environment. The main objective of the EU test strategy related to TBMs in ITER is to provide the necessary information for the design and fabrication of breeding blankets for a future DEMO reactor. The two reference breeding blankets developed by Europe [1] are the Helium-Cooled Lithium-Lead (HCLL) blanket which uses the eutectic Pb–17Li as both breeder and neutron multiplier, and the HeliumCooled Pebble-Bed (HCPB) blanket which features lithiated ceramic pebbles (Li4 SiO4 or Li2 TiO3 ) as breeder and beryllium pebbles as neutron multiplier. Both blankets are using the pressurized He technology for the power conversion cycle (8 MPa, inlet/outlet temperature 300 ◦ C/500 ◦ C) and the 9% CrWVTa Reduced Activation Ferritic Martensitic (RAFM) steel as structural material, the EUROFER. Main features of the TBMs and circuits design, of the basic parameters and of the test strategy in ITER has been presented at SOFT23 in Venice 2004 [2] and at the ISFNT-7 in Tokyo [3]. This paper gives an overview of the last progresses done these last 2 years in terms of system design, calculations, safety and R&D.
2. TBMs design and system updates For HCLL TBM, a first set of modifications have been realized [4]. They first concerned specific functional features in order to better comply with the testing objectives and to improve TBM reliability, such as: Pb–17Li flow path, TBM instrumentation, Pb–17Li draining, Pb–17Li manifold and He manifolds. They were also some modifications related to fabrication issues. Actually, after an industrial expertise on the In-TBM preliminary design to identify fabrication and mounting sequence issues, some following design guidelines have been adopted: avoid welding triple points on the junction first wall/stiffening grid, suppress sharp points on some welding trajectories, avoid possible interference between welding beams and welded parts (Back Plates/Side Walls), optimize thick plates welding. Finally, a preliminary version of the module
attachment system to the frame, assuming flexible fixation and gliding shear keys system, has also been realized. More recently, two main modifications have been further introduced in the HCLL TBM design. The first one concerns the addition of a by-pass tube to extract part of the He first wall (FW) flow-rate in order to be able to increase up to 500 ◦ C the cooling-plate (CP) outlet temperature. This feature is necessary because of the difference in terms of Neutron Wall Loading comparing DEMO operation (∼2.4 MW/m2 ) and ITER DT Phase (0.78 MW/m2 ). Moreover, the He cooling path has been reviewed following thermal–hydraulics studies performed for the HCLL DEMO design: the new reference cooling scheme is a full in-series cooling, FW first, then Stiffening Plates and covers, and CP at the end. For HCPB, the design presented in 2004 has been in these years continuously optimised and dedicated to the first, so-called EM-TBM, to be installed into ITER. The major change that has been done since 2004 is a modification of the breeding unit. The breeding canister has been redesigned and its orientation changed from horizontal to vertical. Instead of the double canister with straight channels, the canister is now made from two cooling plates connected by a wrapper. Some of the coolant channels in the plates run along the circumference, some are meandering in toroidal direction some in the radial one. Moreover, the question of instrumentation of the HCPB EM-TBM has also been addressed. There are sensors in the high pressure zone of the helium coolant, the low pressure zone in the purge gas system and in the vacuum vessel on the surface of the TBM. Electrical and pressure signals have to be transferred. No sensor line should be interrupted in the vacuum vessel, requiring a continuous line from the sensor through the vacuum vessel vacuum boundary to the plug where the signal is passed to the data acquisition system. Handling of lines of the required length must be proven in advance with dedicated tests. The sensor plugs are very space consuming; they will be mounted on the pipe integration cask in the port cell. If a sensor fails no repair is possible, so there is the conflict of request for a high redundancy and the limited space. To cope with this problem extensive out of ITER tests are needed in the TBM mock-ups planned to gather experience with the sensors to be used and the issue to be investigated.
J.-F. Salavy et al. / Fusion Engineering and Design 82 (2007) 2105–2112
The primary loop for the HCPB-TBM, mainly common with the HCLL one, has been investigated with regard to layout definition, selection and dimensioning of components including piping and mechanical integration of the circuit into the different sections of ITER. The accommodation of the main components of the Helium loop into the Torus Coolant Water System (TCWS) vault turned out as the most challenging point [5].
3. Calculations updates For each European TBM concept, significant improvements in thermal–hydraulics, mechanical and T permeation calculations methods have been realized. They can be summarized as follows. 3.1. HCLL calculations progresses ◦ Thermal–hydraulics analyses: new Finite Elements (FE) models of the blanket that allow to obtain the temperature distribution in the coolant on the basis of the detailed power density distribution in the surrounding structures have been developed. Improved models, heat transfer coefficient review, description of the calculations and main conclusions, showing in particular that cooling the FW, SPs and CPs in-series is a viable solution, are described in [6]. ◦ At the same time, the models allow to improve the description of the heat transfer between the solid walls and the coolant. A review of the most used relations available in literature to estimate the heat transfer coefficient has then been performed and the most appropriate has been selected. The new models have been used to carry out a thermal hydraulic analysis of the cooling system of the HCLL breeding blanket. Also, an estimation of the pressure drops in the different cooling elements has been made. Also, counter current flow in the FW can be adopted to reduce the overall thermal stresses in the structure. However, a careful re-engineering of the geometry of the He channels in the cooling plates is needed to avoid hot spots in the steel and achieve a more uniform He outlet temperature. ◦ Mechanical analyses [7]: in order to validate the last improvements made to the design, several FEM analyses have been performed with CAST3M code on
2107
relevant portions of the TBM geometry such as a horizontal TBM slice, top cap and stiffening grid. For each component, results show a correct and admissible behaviour of this part of the structure regarding to the loads and criteria (linear analyses according to SDC-IC criteria verifications assuming 500 ◦ C as structure temperature). The attachment and supporting system of the TBM implemented in the design needs to be able to withstand the disruption loads and TBM weight, while allowing thermal expansions of the box. In order to verify and validate the behaviour of this system, a local model of the flexible cartridge and a global model of the back plates and shear keys were established. Stressed computed in these various parts of the attachment system show significant margins with regards to acceptable limits. ◦ Tritium permeation: a local finite element modelling of the tritium permeation rate through the HCLL breeder unit cooling plates has been developed and is presented in [8]. ◦ Safety and licensing: preliminary thermal and mechanical analysed in accidental transients have been performed but need to be updated due to recent precisions in ITER safety strategy related to TBM operations. In parallel, in view of preparing the licensing of the HCLL TBM for ITER, a first assessment of the impact of the TBM on its environment from a safety viewpoint during the different phases of its lifespan (operation, maintenance, disposal, accident) has been performed [9]. Based on activation calculations and envisaged accident scenarios which need to be modified, the results of this first study aiming at evaluating the impact of HCLL TBMs upon several aspects: releases, dose rates and waste, have not identified killing issues. 3.2. HCPB calculations progresses ◦ Thermal–hydraulics analyses: several models have been developed in the last years for the thermohydraulics analysis of the TBM and TBM systems: RELAP 5 model for the whole helium systems, % TBM model in ANSYS for the study of the temperature distribution, and detailed models with STAR CD of the single cooling channels and manifolds. Several calculations have been performed with these models to optimise the design of the helium cooling system and to select the most suitable working point.
2108
J.-F. Salavy et al. / Fusion Engineering and Design 82 (2007) 2105–2112
A particular effort has been dedicated on the thermohydraulic analysis of the first wall. The cooling of this component is critical due to an extreme one sided heat flux. Unexpected high radial temperature differences were found and asked for better cooling conditions. More details of this assessment can be found in [10]. The flow distribution in the manifolds was investigated both for the individual cooling channels and for the manifolds. It turned out that the mass flow deviations, without any adjustment are within 10–15% of the design values. ◦ Mechanical analyses: in order to validate the last improvements made to the design, several FEM analyses have been performed with ANSYS program on relevant portions of the TBM geometry such a quarter of the TBM, the manifold region, the attachment system, and a complete BU units. The mesh has been generated directly from the CAD drawings. The assessment according to the SDC-IC criteria verifications shows a correct and admissible behaviour of the structure regarding to the loads in operation and fault conditions. Example of this calculation for the first wall can be founded in [10]. The model used for the attachment system covers also the manifolds to allow for formulation of reasonable boundary conditions for the analysis. Dimensioning of the montage gap between the shear keys of the attachment is a crucial issue, to get more insight in this question a detailed investigation is under way with regard to find a compromise in between minimization of the gap to reduce the dynamic loads during a plasma disruption and thermal stresses occurring in the attachment system during operation. So far it was found that the temperature distribution in the manifolds lead to a bowing of the TBM back plate having negative effects on the contact zone of the opposing stub keys. As a general result we found that the thermal stresses in the attachment are of the same order as those due to disruption loads. ◦ Safety and licensing: the Preliminary Safety Analysis Report for the HCPB TBM is under preparation to get the construction permit first and then the operation licence later. In general, the reduced inventory of activation products and tritium associated with the TBM system makes its impact almost negligible to the overall safety risk of ITER. Release of radioactive effluents and radiation exposure of work-
ers connected to the operation of this system are under assessment, but are considered almost uncritical. This consideration can be extended also to the safety assessment of the TBM system related to accidental conditions. Nevertheless, the possibility to jeopardise the ITER safety concept has been analysed in connection to the consequences of specific accident sequences, e.g. the pressurisation of the Vacuum Vessel (VV) due to the He coolant blow-down, the hydrogen production from some breeder/multiplier-steam reactions, the possible interconnection between Port Cell and Vacuum vessel causing air ingress, and the necessity to assure the heat removal in the short and long period. A summary of preliminary studies on accidental sequences can be found in [11].
4. R&D progresses This section only refers to some particular features of the very extensive European R&D program runs these recent years for the two breeding blanket concepts. It concerns progresses in fabrication, tritium control experiments and test facilities development. 4.1. Subcomponent fabrication Each of the main European TBM subcomponents possesses internal cooling channels with a rectangular cross section. This feature makes them difficult to manufacture, even using advanced technologies such like Hot Isostatic Pressing-Diffusion Welding (HIP-DW). Basically, HIP-DW consists in applying a high gas pressure at high temperature to a stack of parts in order to achieve bonding between the surfaces. Gas must be prevented from entering gaps between parts. This is achieved by welding the periphery of the joints or by using a canister. With this process, very good joint properties can be achieved, especially with Eurofer [12]. Complex shapes can be achieved as well, as explained hereafter. A component with round embedded channels can be manufactured by figuring the channels thanks to round tubes, providing the channel bending radii are not too small. In this case, the ends of the tubes are welded to the canister in such a way that the gas pressure enters the channels without entering the joints, which prevents the channels to collapse.
J.-F. Salavy et al. / Fusion Engineering and Design 82 (2007) 2105–2112
Processes based on HIP-DW have been recently reviewed in view of their application to HCLL and HCPB TBMs subcomponents. As a result, prime candidate processes have been defined for each kind of subcomponent, as described in the following: - The first walls have rectangular channels. One proposed manufacturing process is based on the use of round tubes inserted between grooved plates [12–14]. During HIP the tubes expand and fit in the grooves. It appeared that this process is not reliable and not versatile enough because tubes may fail during HIP. One proposed solution is to expand the tubes at room temperature using hydraulic forming, then to insert the tubes in the grooves and apply HIP-DW. The use of thick rectangular tubes inserted between thin plates, as described in [12], is no more considered. - The stiffening plates and cooling plates have small, rectangular channels that must follow sharp turns, which makes impossible to use tubes. Two solutions are envisaged for these components, namely the two steps HIP [12–14] or a new process which consists in closing the top of machined grooves by welding thin strips, then adding a plate by HIP-DW. With this technique, the welds are encapsulated within the structure: the plate acts as a barrier preventing leakage between the coolant side and the breeder side. A 400 mm long HCLL cooling-plate mockup has been manufactured according to this process (Fig. 1). It is currently under thermomechanical testing in CEA/Cadarache. It is worth noting that the
2109
two above processes are applicable to the caps. The welding + HIP process is also, in principle, applicable to the first wall, despite its curvature makes it more difficult. An alternative procedure for the manufacturing of cooling plates is also under investigation. The plates are produced by diffusion welding in a press device, starting from two grooved half plates. The process has been successfully tested in laboratory conditions; at the present the transfer to industrial scale is on-going. 4.2. TBM assembly by welding The assembly of the HCPB and HCLL TBMs imply the development of sophisticated welding technologies and tools. In particular the following issues are a challenge: • Welding of thin (6–8 mm) to thick (35–40 mm) plates. • Access of tools in the internal grid structure (200 mm × 200 mm cells). • Avoid hot cracking (Eurofer like other RAFM steels is sensitive to hot cracking under high energy welding processes). • Limitation of the global deformation. • Optimization of the Post Welding Heat Treatment (PWHT) to retrieve the microstructure and mechanical properties of the joint and Heat Affected Zone (HAZ).
Fig. 1. HCLL cooling-plate mock-up tested in the CEA DIADEMO facility.
2110
J.-F. Salavy et al. / Fusion Engineering and Design 82 (2007) 2105–2112
EU has investigated the capabilities of the TIG and laser welding processes for assembling the TBM structure. Even if the final selection of a reference process will be done later on, the welding capability and strategy for both processes has been established. The final selection will come from optimization of the assembly deformation during fabrication and optimization of cost for tool development. TIG butt welding experiments on Eurofer samples have shown that no cracks are observed for welding plates relevant to the stiffening grid configuration (8 mm thickness). An optimization of the groove shape, process and electrode diameter has allowed avoiding lack of penetration of the root pass and minimization of distortion. Welds up to 150 mm length have been produced. The formation of delta ferrite and possible induced deviation phenomena will be further assessed for ensuring a full control of the process. Standard available laser tools are difficult to use under the geometrical constraints imposed by the TBM grid configuration (in particular for T welding). The standard volume of these tools would lead to perform T-shape welds with a non-negligible angle with regard to the perpendicular direction of the plates. For that purpose, the development of reduced size (14 mm diameter) or even more advanced laser tools are ongoing. Preliminary investigation has shown that standard YAG laser welding parameters applied on 8 mm thick
Eurofer samples result in apparition of hot cracking in the fusion zone of few samples, in particular in the first 50 mm of the weld length. This phenomenon is well known [15] and EDS and SIMS analyses have confirmed that no chemical segregation occurred around the crack, indicating that standard YAG welding parameters induce excessive stress level in the solidification zone (at high temperature, strength of the austenite phase is not sufficient to accommodate stress concentration flow for the this solidification time stage). A development program has been carried-out to eliminate hot cracking trouble. This program was based on a systematic investigation of parameters influencing hot cracking (welding travel speed, beam energy, beam spot profile, pre/post heat treatment, design adjustment, etc.). It was demonstrated that hot cracking can be avoided on Eurofer 8 mm thick plates (Fig. 2). Few pores (<0.5 mm diameter) were observed but experiments have shown that they were never at the origin of crack initiation during destructive examination. For the welding of thick walls (e.g. top/bottom covers, back plates) several processes are under investigation, from industrial processes like the Narrow Gap TIG welding operated in nuclear field to more advanced processes like the Laser MIG Hybrid process developed by CEA for several years in the frame of ITER Vacuum Vessel sector joining development.
Fig. 2. Adapted laser weld shape of Eurofer sample (8 mm thickness). No occurrence of hot cracking, presence of few pores (<0.5 mm).
J.-F. Salavy et al. / Fusion Engineering and Design 82 (2007) 2105–2112
4.3. Tritium control related experiments and test facilities Measurements of the diffusivity and the Sievert’s constants of the hydrogen isotopes protium and deuterium in liquid Pb–15.7 at.% Li, by means of LEDI and SOLE apparatuses [16], have been recently completed (ENEA-Brasimone) using the absorption technique. The hydrogen diffusivity was found to be about two orders of magnitude higher than previously selected reference literature data (Reiter’s data [17] obtained by desorption technique). Also, the values of the Sievert’s constant (a measure for solubility) are about one order of magnitude higher than stated in the literature. These data are very important for the HCLL breeding blanket concept having implications with respect to tritium control, tritium permeation and tritium extraction. If they were true, the tritium permeation problem in case of the HCLL concepts would be less severe as the driving force for permeation is reduced due to the higher solubility in Pb–Li. This is certainly true as long as Pb–Li flow is turbulent and the whole hydrogen inventory is available for permeation. In a strong magnetic field (e.g. in TBMs) where we expect to have laminar flow along the cooling plates, the reduction in permeation could get reversed by the two orders of magnitude higher diffusion coefficient. A repetition of the measurements of hydrogen diffusivity and Sievert’s constant in Pb–Li is currently planned with emphasis on comparison of the data to be obtained by desorption and absorption techniques and using an equipment with walls of extremely low hydrogen solubility. The construction of the TRIEX loop for study of tritium recovery from Pb–Li, to be operated in Brasimone (ENEA), has been completed. Details on loop components, operating parameters and test campaign are given in [18]. This facility will use Hydrogen (tritium) partial pressure sensors for measurement in Pb–Li. Recent developments of such sensors, using Nb and Fe Armco permeation membranes, are being tested in the temperature range of 400–550 ◦ C (Politecnico di Torino, ENEA-Brasimone).
5. Conclusions In order to cope with the ambitious objective to introduce two European TBM systems for day-1 of
2111
ITER operation, EU maintains a large effort in terms of design activities, modelling improvement and R&D development for the HCLL and HCPB blanket modules. Up to now, no killing issue has been identified in these various activities. This effort will be pursued in the framework of a European project whose technical plan has been detailed over the next 10 years [19].
Acknowledgements This work, supported by the European Communities under the contract of Association between EURATOM and CEA, was carried-out within the framework of the European Fusion Development Agreement. The views and opinions expressed herein do not necessarily reflect those of the European Commission.
References [1] L.V. Boccaccini, L. Giancarli, G. Janeschitz, S. Hermsmeyer, Y. Poitevin, A. Cardella, E. Diegele, Materials and design of the European DEMO blankets, J. Nucl. Mater. 329–333 (2004) 148–155. [2] Y. Poitevin, L.V. Boccaccini, A. Cardella, L. Giancarli, R. Meyder, E. Diegele, R. Laesser, G. Benamati, The European Breeding Blankets development and the test strategy in ITER, Fusion Eng. Des. 75–79 (2005) 741–749. [3] L.V. Boccaccini, J.-F. Salavy, R. L¨asser, A. Li Puma, R. Meyder, H. Neuberger, Y. Poitevin, G. Rampal, The European Test Blanket Module systems: design and integration in ITER, invited Paper to the ISFNT-7, Fusion Eng. Des. 81 (2006) 407–414. [4] G. Rampal, D. Gatelet, L. Giancarli, G. Laffont, A. Li-Puma, J.F. Salavy, E. Rigal, Design approach for the main ITER Test Blanket Modules for the EU Helium Cooled Lithium-Lead blankets, Fusion Eng. Des. 81 (2006) 695–700. [5] H. Neuberger, X. Jin, L.V. Boccaccini, B-E. Ghidersa, R. Meyder, Helium loop for the HCPB Test Blanket Module, this conference. [6] G. Aiello, F. Gabriel, L. Giancarli, G. Rampal, J.-F. Salavy, Thermal–hydraulic analysis of the HCLL DEMO Blanket, this conference. [7] G. Rampal, Design and analyses of the HCLL TBM including design of supporting system and instrumentation integration, CEA Report SEMT/BCCR/RT/06-004/A, September 2006. [8] F. Gabriel et al., A 2D finite element modeling of tritium permeation for the HCLL DEMO blanket module, this conference. [9] C. Lacressoniere, O. Gastaldi, TW2-TTBC-005-D02: TBM system detailed safety and licensing, CEA Technical note DTN/STPA/LPC/2005/059, 2005. [10] M.Ili´c, B. Dolensky, L.V. Boccaccini, R. Meyder, C. Polixa, P. Schanz, Thermo-hydraulics and mechanical design of the
2112
[11]
[12]
[13]
[14]
J.-F. Salavy et al. / Fusion Engineering and Design 82 (2007) 2105–2112 First Wall of the European HCPB Test Blanket Module, this conference. L.V. Boccaccini, R. Meyder, X. Jin, Analysis of three LOCAs for the European HCPB Test Blanket System, this conference. A. Cardella, E. Rigal, L. Bedel, P. Bucci, J. Fiek, L. Forest, L.V. Boccaccini, E. Diegele, L. Giancarli, S. Hermsmeyer, G. Janeschitz, R. L¨asser, A. Li Puma, J.D. Lulewicz, A. M¨oslang, Y. Poitevin, E. Rabaglino, The manufacturing technologies of the European breeding blankets, J. Nucl. Mater. 329–333 (2004) 133–140. T. Heider, K. Schleisiek, G. Reimann, Ch Dellis, E. Rigal, Development of manufacturing processes for the ITER heliumcooled pebble bed blanket test module, Fusion Eng. Des. 39–40 (1998) 819–824. E. Rigal, G. Le Marois, T. Lechler, G. Reimann, K. Schleisiek, L. Sch¨afer, P. Weimar, Development of FM steels diffusion
[15]
[16]
[17] [18]
[19]
bonding technologies for blanket manufacturing applications, Fusion Eng. Des. 49–50 (2000) 651–656. G. de Dinechin, C. Chagnot, Ph. Aubert, CEA Weldability Developments on RAFM Steels, Report CEA/DRT/LITEN/ UTIAC-04-RT, May 2004. A. Aiello, A. Ciampichetti, Measurement of Hydrogen Transport Parameters in Lead Lithium Alloy Using LEDI and SOLE devices, ENEA Report, LB-A-R-023, March 2005. F. Reiter, Solubility and diffusivity of hydrogen isotopes in liquid Pb–17Li, Fusion Eng. Des. 14 (1991) 207–211. A. Aiello, A. Ciampichetti, M. Utili, G. Benamati, The TRIEX experimental loop for tritium extraction systems qualification, this conference. Y. Poitevin, L.V. Boccaccini, G. Dell’Orco, E. Diegele, R. L¨asser, J.-F. Salavy, J. Sundstr¨om, M. Zmitko, The Test Blanket Modules project in Europe: from the strategy to the technical plan over next ten years, this conference.