Manufacturing and joining technologies for helium cooled divertors
Introduction
The divertor is a key component in fusion power plants which acts as an exhaust for the fusion reaction products and their high energy. Common feature of divertor concepts developed so far is the use of target structures intensively cooled by proper coolant and mostly built from tungsten because of its high melting point, very low sputtering yield, and good thermal conductivity. For cooling pressurized sub-component is thermally joined to target and as structural component is built from suitable structural material fulfilling appropriate requirements particularly for high temperature strength and sufficient toughness in the operational temperature range. Due to the brittleness of tungsten in the low temperature range its use also as a material for the structural component is only possible for the high temperature part of it. Consequently to build a divertor at least another material beside tungsten needs to be selected and technologies for joining with tungsten need to be realized.
One of the most advanced designs for helium cooled (HC) divertor is that developed at the Karlsruhe Institute of Technology (KIT) [1]. The KIT HC divertor has a modular design, in which the divertor target plate is built from several 9-finger-modules each of them consists of 9 one-finger-modules. The one-finger-module is assembled of three parts: tile, thimble and housing (see Fig. 1). The tile is the plasma facing target built from tungsten (W). For cooling the tile it is thermally bonded to the thimble which in turn is joined to the housing. The thimble and the housing are flown and pressurized on their inner side by helium and hence are structural parts. Due to its high temperature strength, low thermal expansion and low activation tungsten lanthanum oxide (WL10) is selected for the thimble while the housing is built from to the reduced activation ferritic martensitic steel EUROFER97 or its oxide dispersion strengthened variant, EUROFER-ODS. With thermal load of 10 MW/m2 and impingement helium cooling the temperatures expected in the tile are in the range between 1200 °C (thimble side) and 2200 °C (plasma side). Hence, the operation temperature of the joint between the tile and the thimble is around 1200 °C which is at the same time the upper operation temperature of the thimble mainly limited by the creep strength and the recrystallization temperature of WL10. The lower operation temperature of the thimble, 650 °C is determined by the ductile to brittle transition temperature (DBTT) of WL10 and proper margin considering its shift due to neutron irradiation. Consequently the joint between thimble and housing is designed at the position where the temperature ranges around 650 °C which is at the same time the upper operation temperature of the steel housing.
To realize the KIT HC divertor described above technologies are required for joining the tile with the thimble, tungsten–tungsten joints, and the thimble with the steel housing, tungsten–steel joints. In addition and in view to the large number of one-finger-modules required (250,000 for one fusion power plant) special manufacturing technologies are necessary for mass producing tiles and thimbles with very high quality from tungsten and WL10, respectively, which however are hard to machine particularly at room temperature. In the following the recent developments in the mentioned joining and manufacturing technologies are presented and discussed.
Section snippets
Tungsten–steel joints
The main challenges in realizing reliable tungsten–steel joints range from mismatch in thermal expansion between tungsten and steel over metallurgical reactions with brittle phase formation to crack stopping ability and in case of brazing excellent surface wetting. These requirements were only met partly and insufficiently in the past, e.g., by casting Cu between tungsten and steel. To fulfil them three types of joints, namely brazed, diffusion bonded and functionally graded tungsten–steel
Tungsten–tungsten joints
Two different brazing methods for joining tungsten components are being developed at KIT. While the first one is based on using commercially available brazing material of up to 50 μm thickness, dictated by the required capillary effect the second method relies on the electro-chemical deposition of desired elements on the tungsten surface in the order of up to 100 μm intermediate layer thickness.
Shaping of tungsten parts
The modular design of the KIT helium cooled divertor requires manufacturing of huge number of tungsten parts, in particular tiles and thimbles. For the divertor of a power plant approximately 250,000, respectively, need to be fabricated with a very high quality, especially the thimbles which are structural parts. Due to brittleness and hardness of tungsten fabrication of tungsten parts by conventional machining such as turning and milling is not only expensive and time consuming but also yields
Conclusions
The hybrid material concept to be selected for the KIT helium cooled divertor with tungsten as armor as well as structural material presents new technical challenges which call for new joining and manufacturing technologies. Various technologies have been pursued and further developed with the aim of producing divertor components with very high quality and sufficient reliability. Thereby very promising routes have been found for which however still R&D works are necessary. These include the
Acknowledgments
This work, supported by the European Communities under the contract of Association between EURATOM and Karlsruhe Institute of Technology (KIT), 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. The development of functionally graded tungsten–steel joints has been carried out in cooperation with the magnetron sputtering group of PD Dr. Sven Ulrich at KIT, IAM-AWP
References (15)
- et al.
J. Nucl. Mater.
(2011) - et al.
Fusion. Eng. Des.
(2011) - et al.
J. Nucl. Mater.
(2011) - et al.
Fusion Eng. Des.
(2011) - et al.
J. Nucl. Mater.
(2012) - et al.
Fusion Eng. Des.
(2011) - et al.
J. Nucl. Mater.
(2013)