Alternatives for upgrading the EU DCLL breeding blanket from MMS to SMS

https://doi.org/10.1016/j.fusengdes.2021.112380Get rights and content

Highlights

  • Different alternatives to upgrade the EU DCLL MMS design to a SMS architecture have been evaluated.

  • An electrically resistive PbLi-compatible ceramic is proposed for the BZ, enclosed by a steel case including the FW.

  • CO2 is a fairly good alternative to He, although possible radiological issues must be clarified.

  • The preliminary analyses of two proposed supporting systems point out towards the mechanical feasibility of the design.

  • The double wall compromises the tritium self-sufficiency. Some variants of the baseline are promising to enhance the TBR.

Abstract

In the last years, CIEMAT worked in the development of a low temperature DCLL breeding blanket design aimed at using conventional materials and technologies. It followed the multi-module segment (MMS) approach, which allows operating within the thermal range tolerated by EUROFER while keeping a low bulk velocity of the self-cooled liquid breeder (PbLi) in the breeding zone. Thinking in the commercial exploitation of fusion, it must be taken into account that there is risk that the foreseen availability goal in fusion reactors is not achieved. In that case, the limited net efficiency of low temperature blankets could not be sufficient to obtain a competitive cost of the electricity. For that reason, more advanced blanket solutions are being explored with the objective of developing a simpler, more reliable and more efficient design based on a single-module segment (SMS) architecture and the use of self-cooled incompressible liquid breeder. In this work, different strategies to solve interrelated aspects like the topology and electrical insulation of the breeder circuits, the first wall integration and the main neutronic responses are analysed and discussed.

Introduction

CIEMAT worked along several years in the design of a breeding blanket (BB) based on the dual-coolant lithium-lead (DCLL) concept [[1], [2], [3]], in which PbLi acts as main coolant, tritium breeder, tritium carrier and neutron multiplier while helium cools the first wall and other structures. The European DEMO Programme encouraged the development of a “low temperature” version of the DCLL BB to allow using conventional materials and technologies. This approach led to adopt the multi-module segment architecture (MMS), in which the blanket is split into a number of vertical segments, and each segment is composed, in turn, by a series of modules attached to a common back supporting structure (BSS) which also accomplishes functions of manifold and shield (Fig. 1). The arrangement of individual fluid circuits in parallel made possible to operate within the thermal range tolerated by EUROFER (300−550 °C) while keeping the bulk velocity of the self-cooled liquid breeder as low as ∼2 cm/s in the breeding zone, which involves important benefits from the point of view of magnetohydrodynamics (MHD) and corrosion. The design also included flow channel inserts (FCI) to electrically decouple the liquid metal bulk flow from the steel walls.

Paving the way from demonstration reactors towards power plants, economic studies consider that the cost of electricity generated by fusion will be dominated by capital construction cost rather than operating cost, so that plant availability will be a more important factor than in other generation technologies [4]. The efficiency of the power conversion cycle will be another decisive parameter to establish the cost of electricity. What is not proven is that the assumed power levels, conversion efficiency and most notably availability can be achieved in practice [5]. Therefore, there is a risk that the limited net efficiency of low temperature blankets is not sufficient to obtain a competitive cost of electricity generated by fusion. For that reason, more advanced blanket solutions are being explored with the objective of developing simpler, more reliable and more efficient designs. Key aspects to achieve such goal are the use of a self-cooled incompressible liquid breeder, which is a vector to high efficiencies, and segments with single module segment architecture (SMS), which can help in providing more simplicity and reliability.

The strategy proposed here consists in a deep evolution of the DCLL concept in aspects like structural material, electrical insulation method, topology of the fluid circuits and first wall (FW) integration. Different alternatives for each of those aspects have been considered and certain design or technology options have been preliminarily selected for further assessment. The reasons which justify such choices are discussed in Section 2. Additionally, Sections 3–8 are devoted to describe some specific calculations (MHD, neutronics, thermomechanics, etc.) in support of them.

Section snippets

Structural material and electrical insulation method

Option 1: Advanced steel + coating. Oxide dispersion strengthened (ODS) EUROFER is perhaps the most attractive option as structural material for a moderate increase of the operational temperature because of its good tensile and creep properties, which would allow operating at T∼650 °C [6]. However, its corrosion behaviour is not much better than the EUROFER97 one. This can be solved by applying a coating on the surface of the PbLi channels which acts as corrosion barrier and electrical

Coolability of the FW

The irregular spatial distribution of heat flux on the FW due to radiation from the plasma and charged particles is undoubtedly challenging for the design of the FW cooling system. Indeed, a conservative oversizing of the coolant mass flow rate can have a severe impact on the blanket thermal efficiency and the consumption of the auxiliary systems, since He is employed –together with PbLi- as thermal source for the power conversion system (it represents ∼30 % of the power extracted from the

CO2 as alternative FW coolant

3D CFD steady-state analyses with conjugate heat transfer have been carried out in ANSYS Fluent to compare the performance of He and CO2 as coolant for the FW. The model takes advantage of the linear periodic symmetry of the OBC FW cooling system along the poloidal direction. Therefore, it only includes two coolant streams in counterflow and the corresponding W and EUROFER parts. Smooth channels of 12.5 × 12.5 mm2 have been considered, with a pitch of 5 mm and a front wall thickness of 2 mm.

PbLi poloidal channels vs B-oriented channels

This Section explains the methodology to compare the 3D pressure drop associated to the bends in the B-oriented configuration and the annulled 2D fraction, as introduced in Subsection 2.4. In order to establish a comparison criterion, it must be taken into account that the increase of temperature per unit of PbLi volume and poloidal length should be the same in both schemes. This implies that the ratio of mean velocities in the classical (poloidal channels) and alternative (B-oriented channels)

Performance of a supporting system based on shear keys

The functioning of the shear keys system (option 2 in Subsection 2.5) can be illustrated by means of a quite simple 2D steady-state FEA. The model geometry (Fig. 9), adapted from the equatorial section of the OBC segment, is composed by the ceramic box (grid of 6 parallel circuits formed by poloidal channels), the steel case and the back wall (both solid). All the walls have a thickness of 25 mm excepting the back wall (250 mm). In this case, alumina has been considered as testing material for

Pressurization of the ceramic box in case of in-box LOCA

The inert gas which fills the gap between the ceramic box and the steel case should allow detecting and then relieving the overpressure produced by the accidental break of the FW cooling channels. But high pressure zones could appear on the ceramic surface in short transients before detection. For that reason, 2D and 3D CFD transient analyses have been performed in ANSYS Fluent to simulate the pressure evolution at the surface of the wall opposite to the channel break, considering this as an

Tritium breeding performance

One of the main concerns of this design is the impact of the double wall (steel-ceramic) on tritium self-sufficiency. Hence, a heterogenized neutronic model with SiC as material for the ceramic box has been created, given the results of the radiological assessment commented in Subsection 2.1. The model has been prepared using the tool SuperMC MCAM 3.2 [35]. The particle transport calculations have been performed with MCNP5, using the JEFF 3.2 nuclear data library [36]. The direct simulation

Conclusions

Different alternatives to upgrade the EU DCLL MMS design to a SMS architecture have been evaluated. It has been proposed to use an electrically resistive and PbLi-compatible ceramic as structural material for the breeding zone, which allows doing without the FCIs and suppresses the limit in the operational temperature imposed by creep in the case of reduced activation steels. The breeding zone is enclosed by a RAFM steel envelope which includes a continuous FW panel cooled by He. CO2 has been

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work has been carried out within the framework of the EUROfusion Consortium and has received funding from the Euratom research and training programme 2014-2018 and 2019-2020 under grant agreement No 633053. The views and opinions expressed herein do not necessarily reflect those of the European Commission. This work has been partially supported by the computing facilities of Extremadura Research Centre for Advanced Technologies (CETA-CIEMAT), funded by the European Regional Development Fund

References (38)

  • Y. Wu et al.

    Design status and development strategy of China liquid lithium-lead blankets and related material technology

    J. Nucl. Mater.

    (2007)
  • F. Maviglia

    Optimization of DEMO geometry and disruption location prediction

    Fusion Eng. Des.

    (2019)
  • J.I. Linares

    Supercritical CO2 Brayton power cycles for DEMO fusion reactor based on helium cooled Lithium Lead blanket

    Appl. Therm. Eng.

    (2015)
  • D. Martelli

    Literature review of lead-lithium thermophysical properties

    Fusion Eng. Des

    (2019)
  • R. Bermejo et al.

    Mechanical characterization of ceramics: designing with brittle materials

    (2014)
  • Y. Wu et al.

    CAD-based interface programs for fusion neutron transport simulation

    Fusion Eng. Des.

    (2009)
  • J.L. Duchateau

    Conceptual design for the superconducting magnet system of a pulsed DEMO reactor

    Fusion Eng. Des.

    (2013)
  • D. Rapisarda

    Conceptual design of the EU-DEMO dual coolant lithium lead equatorial module

    IEEE Trans. Plasma Sci. IEEE Nucl. Plasma Sci. Soc.

    (2016)
  • F.R. Urgorri

    Magnetohydrodynamic and thermal analysis of PbLi flows in poloidal channels with flow channel insert for the EU-DCLL blanket

    Nucl. Fusion

    (2018)
  • Cited by (16)

    • Comparative study of the mechanical properties after proton irradiation of Al<inf>2</inf>O<inf>3</inf> and Si-SiC ceramics for fusion breeding blanket applications

      2022, Fusion Engineering and Design
      Citation Excerpt :

      So far, designs are considering either the location of floating ceramic tubes inside the BB channels (so-called flow channel inserts, FCI) [2] or the coating of the inner channel surface with a thin insulating film [3] to decouple the flow of PbLi from the magnetic field with adding an electric insulating component [4]. Nevertheless, recent designs are considering the viability of insulating ceramic channels as single-module segments (SMS) [5] in which the bulk ceramic will also provide certain structural functionality. During the operation of the fusion reactor, the high-energy neutrons (14 MeV) will interact with the structural materials, generating transmutations and atomic displacements [6].

    • The microstructural in-sight to the mechanical behavior of alumina after high energy proton irradiation

      2022, Nuclear Materials and Energy
      Citation Excerpt :

      The transmutted hydrogen stimulates nucleation and growth of cavities, which would trigger internal stresses in the ceramic material, thus changing its initial mechanical properties [6]. Therefore, the characterization of the mechanical behaviour in irradiated ceramics is essential since the material would have to fulfill some structural functionality in the most recent DCLL design, based on a single-module segment (SMS) architecture [7,8]. This work aimed to study the mechanical behaviour under compression and micro-indentation tests of high purity alumina material, comparing the as-received condition and high-energy proton material irradiated conditions, accompanying the likely radiation-induced radiation-induced tests microstructural changes with SEM and TEM observations.

    View all citing articles on Scopus
    View full text