- Email: [email protected]
Research and development status on fusion DEMO reactor design under the Broader Approach
G Model
ARTICLE IN PRESS
FUSION-7350; No. of Pages 5
Fusion Engineering and Design xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Fusion Engineering and Design journal homepage: www.elsevier.com/locate/fusengdes
Research and development status on fusion DEMO reactor design under the Broader Approach Kenji Tobita a,∗ , Gianfranco Federici b , Kunihiko Okano c , The BA DEMO Design Activity Unit a
Japan Atomic Energy Agency (JAEA), Rokkasho, Japan Fusion for Energy, Garching, Germany c International Fusion Energy Research Centre (IFERC), Rokkasho, Japan b
h i g h l i g h t s • • • •
Latest status of the DEMO design activity under the Broader Approach. Start points of DEMO parameter scoping study. DEMO divertor study aiming at detached plasma with impurity injection and magnetic flux tube expansion by advanced divertor configurations. Assessment of possible remote maintenance schemes for DEMO.
a r t i c l e
i n f o
Article history: Received 9 September 2013 Received in revised form 23 February 2014 Accepted 24 February 2014 Available online xxx Keywords: DEMO Fusion reactor Divertor Remote maintenance Safety
a b s t r a c t The main objective of DEMO design activity under the Broader Approach is to develop pre-conceptual design of DEMO options by addressing key design issues on physics, technology and system engineering. This paper describes the latest results of the design activity, including DEMO parameter study, divertor and remote maintenance. DEMO parameter study has recently started with “pulsed” DEMO having a major radius (Rp ) of 9 m, and “steady state” DEMO of Rp = 8.2 m or more. Divertor design study has focused on a computer simulation of fully detached plasma under DEMO divertor conditions and the assessment of advanced divertor configuration such as super-X. Comparative study of various maintenance schemes for DEMO and narrowing down the schemes is in progress. © 2014 Elsevier B.V. All rights reserved.
1. Introduction In the DEMO design activity under the Broader Approach (BA), the first task was to crystalize the joint design work to be implemented by EU and Japan (JA). After three-year preparatory activity (2007–2010) in which various views on DEMO were shared by EU and JA researchers in workshops, the details of the joint design work were agreed on and the joint work started in 2011. Previous DEMO design study carried out before the BA [1–3] revealed critical design issues on huge power removal in the divertor, tritium self-sufficiency and power extraction, remote maintenance, integrated design including plasma control and so on. All these issues are not only difficult themselves but each one
∗ Corresponding author. Tel.: +81 175716670. E-mail addresses: [email protected], [email protected] (K. Tobita).
is closely related with the other. Therefore, in the design integration, it is required to understand trade-off relations and intertwined constraints underlying in these issues and resolve all of them systematically. For the purpose, the BA DEMO design has concentrated on assessment of critical design issues, which includes the consolidation of the knowledge base so far achieved and needed for the design, and analysis of key design issues and options. In the process, every possible option of each component or technology is investigated and assessed for narrowing down to feasible options. The assessment will contribute to developing pre-conceptual DEMO options and planning future technology development programs toward DEMO. In parallel with the study on critical design issues, DEMO design parameters need to be investigated to envisage possible DEMO options. Benchmark of systems codes independently developed by CCFE (Culham Centre for Fusion Energy) and JAEA (Japan Atomic Energy Agency) has been carried out, confirming good agreement
http://dx.doi.org/10.1016/j.fusengdes.2014.02.077 0920-3796/© 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: K. Tobita, et al., Research and development status on fusion DEMO reactor design under the Broader Approach, Fusion Eng. Des. (2014), http://dx.doi.org/10.1016/j.fusengdes.2014.02.077
G Model FUSION-7350; No. of Pages 5
ARTICLE IN PRESS
2
K. Tobita et al. / Fusion Engineering and Design xxx (2014) xxx–xxx
except several issues such as impurity radiation power, fast ion beta and bootstrap fraction [4,5]. This paper describes the main areas of the activity, this is, progress in DEMO design parameters, divertor and remote maintenance in the following sections. The plan for safety research for DEMO is briefly summarized in Section 5. 2. Parameter study of DEMO The basic approach of the BA DEMO design is to develop possible DEMO options that are foreseen from ITER, other on-going and planned R&D programmes. In this context, projections of ITER and the present knowledge to DEMO should be based on a certain broadening dependent on conservative and optimistic views. Another aspect to note is that EU and JA have different requirements for DEMO. Consequently, multiple DEMO concepts are likely to be developed in the BA. EU has explored pulsed and steady state power plants and actually a pulsed option was characterized as the most conservative model in the European Power Plant Conceptual Study (PPCS) [2]. In the BA activity, EU puts emphasis on a pulsed DEMO concept assuming small extension of ITER technology. Preliminary scoping study using an EU systems code is presently pursuing a pulsed DEMO generating net electricity (Pe ) of 0.5 GWe and fusion power (Pfus ) of around 2 GW at a major radius (Rp ) of 9 m, aspect ratio of about 4 and pulse length of as short as about 2 h. Previously, JA reactor study had focused on steady state DEMO concepts that were relatively compact and capable of producing plant-level electricity (∼1 GWe) [1,3]. The previous reactor study reveals that such compact reactors have problems in power removal in the divertor, removal of residual heat of the in-vessel components in a loss-of-coolant accident (LOCA) and tokamak operation capability due to insufficient volt-second supply. Based on the result, JA currently considers a relatively large and lower power steady state concept with sufficient volt-second supply for operational development from pulsed (∼0.5 h) to steady state, having parameters of Rp ∼ 8.2 m or more, Pfus = 1.3–1.5 GW and Pe = 0.2–0.3 GWe. In the scoping study, it is commonly recognized that a vertically stable plasma elongation under constraints of DEMO needs to be investigated, which means no use of in-vessel coils for vertical stabilization and the installation of stabilizing shells away from breeding blanket. 3. Divertor Divertor is one of the most critical design issues giving a significant impact on DEMO design parameters. The problem is how to handle a huge power exhausted from the main plasma being several times as high as that in ITER and eventually to maintain the divertor heat flux at a tolerable level well below 10 MW/m2 . In order to find a possible solution for the problem, two approaches have been progressing. The first approach is to pursue a fully detached plasma condition based on divertor simulation [6,7], and the second is to seek for the possibility of modifying divertor configuration such as super-X and snowflake [8]. 3.1. Divertor simulation Admitting that existing divertor simulation codes have not satisfactorily been validated by experiments, computer simulation is practically the only approach to foresee DEMO because the DEMO conditions are substantially far away from existing experiments. For the JA divertor design study, a divertor simulation code SONIC was upgraded to appropriately deal with thermal and friction forces
of impurity ions along the magnetic field lines. In the simulation, impurity transport and the resulting plasma detachment were simulated self-consistently for different detachment scenarios. 3.1.1. Impurity injection Impurity seeding using different species (Ne, Ar and Kr) is anticipated to change radiation distribution in the scrape-off layer (SOL) and the divertor because of the Z-dependence of radiation loss rate coefficient. The divertor simulation indicates that selecting the impurity species can control the distribution of radiation power in that, with seeding higher-Z impurity such as Kr, a wider plasma detachment region and lower divertor heat flux were obtained at the outer divertor [7]. On the other hand, higher-Z impurity tends to be transported to the main plasma and to cause higher radiation in the main edge, which may deteriorate plasma confinement. 3.1.2. Long leg divertor Previously, it was confirmed that the divertor geometry contributes to reducing divertor heat flux and that V-shaped corner is efficient to enhance particle recycling near the strike point and to produce detached plasma [9]. In order to reduce the divertor heat flux further, longer leg divertor was proposed and investigated for SlimCS [6]. The role of the long leg divertor is to decrease the divertor plasma temperature because of extended connection length to the target plate and to increase the heat receiving area of the divertor. Recent simulation [7] indicates that, when Ar is seeded to the divertor region of the plasma with a fusion output (Pfus ) of 3 GW, full detachment over the outer divertor is attained as shown in Fig. 1(a). Then, the radiation power fraction reaches 80% of the power exhausted to the SOL from the main plasma (500 MW) and the divertor electron and ion temperatures decrease to 2 eV. In the long leg divertor, the ionization front moves upstream (Fig. 1(b)), contributing to reducing the peak heat load on the divertor plate. 3.2. Alternative divertor configurations Another approach for reducing the divertor heat load is a flux tube expansion in the divertor with “super-X” or “snowflake” configuration. Application of such configurations has been investigated by EU [10] and JA [8]. Fig. 2 depicts the arrangement of poloidal field (PF) coils for the super-X configuration and the required coil currents, where the original plasma configuration is based on SlimCS with a plasma current of 16.8 MA. The super-X configuration is numerically obtained using external PF coils. However, the reality of coil current required for Coil #9 of 180 MA seems unlikely because the cross section for the current is anticipated to be 4–6 m2 when the conductor current is 60–100 kA. Development of high current conductor of 150–200 kA can allow the super-X with external PF coils but the leak field produced by such a high current coil may restrict the installation of peripheral equipment such as diagnostics, solenoid valves and pumps. By winding some of the PF coils in an inter-linked way against the toroidal field (TF) coils, the required coil currents are dramatically reduced as shown in Fig. 2(b). Inter-linked winding was originally proposed as a novel concept of central solenoid (CS) allowing a compact CS with a sufficient volt-second supply [11,12]. The winding needs to be carried out in situ in the middle of the TF coil installation and the vacuum vessel one. Although the engineering feasibility of the inter-linked winding must be verified from various aspects, the required coil currents have reality. Further study is under way from the points of view of the compatibility of the inter-linked coils and the neutron and thermal shield. Regarding the super-X divertor, it is noteworthy that the connection length along a magnetic field line near the super-X null point is extended. The connection length of the magnetic field line
Please cite this article in press as: K. Tobita, et al., Research and development status on fusion DEMO reactor design under the Broader Approach, Fusion Eng. Des. (2014), http://dx.doi.org/10.1016/j.fusengdes.2014.02.077
G Model FUSION-7350; No. of Pages 5
ARTICLE IN PRESS K. Tobita et al. / Fusion Engineering and Design xxx (2014) xxx–xxx
Fig. 1. Distribution of radiation power density in (a) the standard divertor with V-shaped corner, and (b) the long leg divertor.
of 0.001 m-away from the separatrix on the outer mid-plane is as long as 60 m, being much longer than those of the standard (36 m) and the long leg divertor (41 m) [8]. This may ease the positional control of the ionization front of detached plasma, which will be investigated using the SONIC code in the coming years. In contrast, the snowflake is likely to be ruled out of the possible divertor configurations for DEMO. Defects of the configuration is (1) extremely high coil currents in part of CS [8], (2) lower plasma volume for the same major radius due to the higher elongation, (3) concern on the vertical stability of such an elongated plasma [13]. 3.3. Divertor target plate Structure and materials of divertor target plates are another divertor issue to be resolved. Originally, JA considered a concept composed of tungsten mono-blocks and a cooling tube made of reduced activation ferritic martensitic (RAFM) steel using pressurized water as coolant [14]. On the other hand, EU had applied helium-cooled divertor in most of earlier EU reactor designs such as PPCS [15]. Despite such historical background, EU presently considers water-cooled divertor as a conservative option for DEMO because of its effective heat removal capability and technical maturity. Preliminary survey by EU suggested that copper (Cu)-based alloy with tungsten armor seems favorable as the water-cooled divertor. Of course, material development is indispensable for the purpose because no existing Cu-alloys withstand neutron irradiation of even as low as 1 dpa. Recently, JA assessed heat removal
3
Fig. 2. Super-X divertor configuration produced by (a) external poloidal field coils and (b) inter-linked poloidal field coils. Upper poloidal field coils are omitted.
capability of the RAFM-based water-cooled divertor, concluding that the allowable heat flux is equivalent to as low as 5 MW/m2 when the tungsten mono-block armors are tapered on the surface to avoid leading edge problem. Based on the result, JA is considering a combination of RAFM-based and Cu-alloy-based divertor for using both materials as the condition demands: RAFM-based for lower heat load and higher dpa zone such as the divertor dome, and Cu-alloy-based for higher heat load and lower dpa zone such as divertor wetted areas. The problem of the combined use of the materials is how to mitigate thermal stress in the arrangement of both in that the coolant temperature for RAFM and Cu-alloy should be different; typically 300 ◦ C for RAFM and 200–240 ◦ C for Cu-alloy. Since the design life of the water-cooled divertor with Cu-alloy is anticipated to be 1 year, its application will be limited to the early operation phase of DEMO. For the later phase, an alternative divertor concept needs to be developed. 4. Remote maintenance Maintenance is a critical design issue because remote maintenance (RM) on DEMO would be far beyond the present technology level. ITER chooses an in-vessel maintenance scheme where small blanket modules are replaced in the vacuum vessel using maintenance vehicles. The scheme is reasonable when a small number of damaged components are replaced. However, because of maintenance time, the scheme is not suitable for periodic replacement of the whole set of blanket modules as required for DEMO. In particular, the most time-consuming processes such as in situ cutting, re-welding and inspection need to be minimized to reduce the maintenance time. For this reason, it is necessary to develop a new
Please cite this article in press as: K. Tobita, et al., Research and development status on fusion DEMO reactor design under the Broader Approach, Fusion Eng. Des. (2014), http://dx.doi.org/10.1016/j.fusengdes.2014.02.077
G Model FUSION-7350; No. of Pages 5
ARTICLE IN PRESS
4
K. Tobita et al. / Fusion Engineering and Design xxx (2014) xxx–xxx
Fig. 3. (a) Sector transport using a limited number of horizontal maintenance port (S-LHP), and (b) MMS-based vertical port maintenance.
RM scheme which would be reliable and potentially allow us to attain favorable plant availability. Since RM scheme is closely coupled with the reactor configuration, it needs to be determined in early time of the DEMO conceptual design. 4.1. Maintenance schemes Recent design study suggests that RM on DEMO requires a major shift from the ITER scheme because of in-vessel dose rate (10–100 times higher than that of ITER) and required plant availability. This shift will also have a lot of influence on reactor configuration, reactor building, hot cell and RM equipment. JA has mainly investigated sector transport maintenance in which sectors assembling blanket modules and divertor cassettes are removed from the reactor and spare sectors are installed anew. In the RM scheme, most parts of cutting, re-welding and inspection of piping are carried out in the hot cell. Comparing three kinds of sector maintenance schemes of (i) sector transport using all horizontal maintenance ports (S-AHP), (ii) sector transport using a limited number of horizontal maintenance ports (S-LHP), (iii) sector transport using limited number of vertical maintenance ports (S-LVP) [16], it is concluded that S-LHP (Fig. 3(a)) will be less challenging than the other two. S-AHP has a serious problem in supporting the turnover force of TF coils, and S-LVP needs to resolve a difficult question on how to transport sectors in the toroidal direction after inserting them in the vacuum vessel. After a free thinking range of maintenance schemes including sector and Multi-Module Segments (MMS) maintenance in 2011, EU selected MMS-based vertical port maintenance (Fig. 3(b)) in which blanket modules are integrated on a banana-shaped inboard and outboard segment and each MMS is transported through a small vertical maintenance port [17,18]. Divertor cassettes are replaced like ITER using a small port located in the lower outboard. The design studies by JA and EU commonly pointed out that the hot cell facility of DEMO seems to be huge. EU’s assessment indicates the total floor area of 23,500 m2 for RM, component processing, storage, waste and recycling, being 6 times as large as the hot cell of ITER (4000 m2 ). JA’s assessment indicates the floor areas of 17,100 m2 (5300 m2 for hot cell and 11,800 m2 for interim storage). The evaluated areas will be updated in future investigation. The point is that waste management strategy associated with the maintenance scenario needs to be defined in the early stage of the conceptual design for the layout of peripheral equipment and facilities.
is to score various criteria including maintenance time, reliability, waste volume and contamination control for each RM scheme as proposed for ARIES-AT [19]. The problem is the weighting for each criterion. Fig. 4 illustrates a relation diagram depicting how the plant availability is determined. When a DEMO plant is operated in a scheduled manner, the maintenance period will be key to the availability. However, the plant is forced to stop operation due to a trouble or an accident, the resulting unscheduled shutdown could result in a serious deterioration of the availability. In this sense, how to improve the reliability of in-vessel components (IVC) would be key to attaining a reasonable availability and a considerably high weighting factor needs to given to the IVC reliability in the scoring approach in the selection of the most suitable RM scheme for DEMO. Installation of brand-new MMS or sectors in every replacement, which improves IVC reliability because of fabrication and inspection by hand in non-controlled areas, should be effective to attain a desirable availability. Then, remote work in radiologically controlled area is limited to cutting, re-welding and inspection of coolant manifolds of MMS or sectors. On the other hand, installation of brand-new MMS or sectors cause a problem of increase in radioactive waste, and the sector maintenance seems to produce a larger amount of radioactive waste than MMS. The discussion described above is an aspect of maintenance. For narrowing down the RM schemes, it is necessary to insightfully analyze maintenance requirements for DEMO as well as engineering feasibility. 5. Safety study Safety studies carried out in past decades have shown favorable features of fusion and indicated that the safety characteristics are strongly dependent on design choices including the combination
4.2. Viewpoint to narrow down the RM schemes The most important design issue on maintenance is methodology of how to select one of many possible options. An approach
Fig. 4. Schematic diagram indicating the impact of RM scheme on the plant availability.
Please cite this article in press as: K. Tobita, et al., Research and development status on fusion DEMO reactor design under the Broader Approach, Fusion Eng. Des. (2014), http://dx.doi.org/10.1016/j.fusengdes.2014.02.077
G Model FUSION-7350; No. of Pages 5
ARTICLE IN PRESS K. Tobita et al. / Fusion Engineering and Design xxx (2014) xxx–xxx
of materials such as structural materials, breeder, multiplier and coolant, and the design parameters of a reactor. In this context, it is still of great significance to investigate safety characteristics of DEMO assuming the most likely design parameters and materials. Safety study in the BA has been conducted in parallel with DEMO design since 2013. The purpose of the safety study is (1) to investigate the safety characteristics of DEMO through safety analysis on hypothetical accidents, (2) to define safety systems and their specifications required for DEMO to reduce or prevent radiological risks to the public, and (3) to formulate DEMO safety guidelines.
[5] [6]
[7] [8]
[9]
6. Summary [10]
The BA DEMO design activity has mainly addressed key design issues, options and DEMO parameters from a broad perspective. After assessing these issues, options and parameters, they will be narrowed down to develop pre-conceptual DEMO options. The activity will contribute to pointing a reasonable direction of DEMO that is foreseeable around 2030 s.
[11]
[12]
[13]
Acknowledgments [14]
This work was carried out within the framework of the Broader Approach DEMO Design Activity. Contribution by all members of JA and EU Home Teams for BA is greatly appreciated.
[15]
[16]
References [1] K. Tobita, S. Nishio, M. Enoeda, H. Kawashima, G. Kurita, H. Tanigawa, et al., Compact DEMO, SlimCS: design progress and issues, Nucl. Fus. 49 (2009) 075029. [2] D. Maisonnier, D. Campbell, I. Cook, L. Di Pace, L. Giancarli, J. Hayward, et al., Power plant conceptual studies in Europe, Nucl. Fus. 47 (2007) 1524. [3] R. Hiwatari, K. Okano, Y. Asaoka, K. Shinya, Y. Ogawa, Demonstration of tokamak fusion power plant for early realization of net electric power generation, Nucl. Fus. 45 (2005) 96. [4] R. Kemp, M. Nakamura, D.J. Ward, K. Tobita, G. Federici, Benchmarking reactor systems studies by comparison of EU and Japanese system code
[17]
[18]
[19]
5
results for different DEMO concepts, in: Proc. of 24th IAEA Fusion Energy Conference, FTP/P7-29 San Diego, 2012, http://www.naweb.iaea.org/ napc/physics/FEC/FEC2012/.papers/24 FTPP729.pdf K. Okano, G. Federici, K. Tobita, DEMO design activities in the Broader Approach under Japan/EU collaboration, Fus. Eng. Des. (2014). K. Hoshino, K. Shimizu, N. Asakura, T. Takizuka, M. Nakamura, K. Tobita, Simulation study of an extended divertor leg for heat control in the SlimCS DEMO reactor, Contr. Plasma Phys. 52 (2012) 550. N. Asakura, et al., A simulation study of large power handling in the divertor DEMO reactor, Nucl. Fus. 53 (2013) 123013. N. Asakura, K. Shinya, K. Tobita, K. Hoshino, Investigation of advanced divertor configuration for DEMO tokamak reactor, Trans. Fus. Sci. Technol. 63 (2013) 70. H. Kawashima, K. Shimizu, T. Takizuka, K. Tobita, S. Nishio, S. Sakurai, et al., Simulation study for divertor design to handle huge exhaust power in the SlimCS DEMO rector, Nucl. Fus. 49 (2009) 065007. J.W. Coenen, Liquid metals as alternative solution for the power exhaust of future fusion devices: status and perspective, J. Nucl. Mater. (2014) (still in print). H. Utoh, N. Asakura, K. Tobita, Y. Someya, Y. Sakamoto, K. Hoshino, M. Nakamura, Application of inter-linked superconducting coils for central solenoid and advanced divertor configuration of DEMO, Fus. Eng. Des. (2014). M. Nakamura, K. Tobita, H. Utoh, Y. Sakamoto, Impact of an ‘interlinked’ central solenoid configuration on the size of tokamak DEMO, Fus. Eng. Des. 88 (2013) 1146. G. Federici, R. Kemp, D. Ward, C. Bachmann, T. Franke, S. Gonzalez, et al., Overview of EU DEMO design and R&D activities, Fus. Eng. Des. (2014) http://dx.doi.org/10.1016/j.fusengdes.2014.01.070 K. Tobita, S. Nishio, M. Enoeda, M. Sato, T. Isono, S. Sakurai, et al., Design study of fusion DEMO plant at JAERI, Fus. Eng. Des. 81 (2006) 1151. P. Norajitra, R. Giniyatulin, N. Holstein, T. Ihli, K. Krauss, R. Kruessmann, et al., Status of He-cooled diverter development for DEMO, Fus. Eng. Des. 75–79 (2005) 307. H. Utoh, Y. Somaya, K. Tobita, N. Asakura, K. Hoshino, M. Nakamura, Critical design factors for sector transport maintenance in DEMO, Nucl. Fus. 53 (2013) 23005. A. Loving, O. Crofts, N. Sykes, D. Iglesias, M. Coleman, J. Thomas, et al., Preconceptual design assessment of DEMO remote maintenance, Fus. Eng. Des. (2014). M. Coleman, N. Sykes, D. Cooper, D. Iglesias, R. Bastow, A. Loving, J. Harman, Concept for a vertical maintenance remote handling system for multi module blanket segments in DEMO, Fus. Eng. Des. (2014) http://dx.doi.org/ 10.1016/j.fusengdes.2014.02.047 L.M. Waganer, The ARIES Team, ARIES-AT maintenance system definition and analysis, Fus. Eng. Des. 80 (2006) 161.
Please cite this article in press as: K. Tobita, et al., Research and development status on fusion DEMO reactor design under the Broader Approach, Fusion Eng. Des. (2014), http://dx.doi.org/10.1016/j.fusengdes.2014.02.077
ARTICLE IN PRESS
FUSION-7350; No. of Pages 5
Fusion Engineering and Design xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Fusion Engineering and Design journal homepage: www.elsevier.com/locate/fusengdes
Research and development status on fusion DEMO reactor design under the Broader Approach Kenji Tobita a,∗ , Gianfranco Federici b , Kunihiko Okano c , The BA DEMO Design Activity Unit a
Japan Atomic Energy Agency (JAEA), Rokkasho, Japan Fusion for Energy, Garching, Germany c International Fusion Energy Research Centre (IFERC), Rokkasho, Japan b
h i g h l i g h t s • • • •
Latest status of the DEMO design activity under the Broader Approach. Start points of DEMO parameter scoping study. DEMO divertor study aiming at detached plasma with impurity injection and magnetic flux tube expansion by advanced divertor configurations. Assessment of possible remote maintenance schemes for DEMO.
a r t i c l e
i n f o
Article history: Received 9 September 2013 Received in revised form 23 February 2014 Accepted 24 February 2014 Available online xxx Keywords: DEMO Fusion reactor Divertor Remote maintenance Safety
a b s t r a c t The main objective of DEMO design activity under the Broader Approach is to develop pre-conceptual design of DEMO options by addressing key design issues on physics, technology and system engineering. This paper describes the latest results of the design activity, including DEMO parameter study, divertor and remote maintenance. DEMO parameter study has recently started with “pulsed” DEMO having a major radius (Rp ) of 9 m, and “steady state” DEMO of Rp = 8.2 m or more. Divertor design study has focused on a computer simulation of fully detached plasma under DEMO divertor conditions and the assessment of advanced divertor configuration such as super-X. Comparative study of various maintenance schemes for DEMO and narrowing down the schemes is in progress. © 2014 Elsevier B.V. All rights reserved.
1. Introduction In the DEMO design activity under the Broader Approach (BA), the first task was to crystalize the joint design work to be implemented by EU and Japan (JA). After three-year preparatory activity (2007–2010) in which various views on DEMO were shared by EU and JA researchers in workshops, the details of the joint design work were agreed on and the joint work started in 2011. Previous DEMO design study carried out before the BA [1–3] revealed critical design issues on huge power removal in the divertor, tritium self-sufficiency and power extraction, remote maintenance, integrated design including plasma control and so on. All these issues are not only difficult themselves but each one
∗ Corresponding author. Tel.: +81 175716670. E-mail addresses: [email protected], [email protected] (K. Tobita).
is closely related with the other. Therefore, in the design integration, it is required to understand trade-off relations and intertwined constraints underlying in these issues and resolve all of them systematically. For the purpose, the BA DEMO design has concentrated on assessment of critical design issues, which includes the consolidation of the knowledge base so far achieved and needed for the design, and analysis of key design issues and options. In the process, every possible option of each component or technology is investigated and assessed for narrowing down to feasible options. The assessment will contribute to developing pre-conceptual DEMO options and planning future technology development programs toward DEMO. In parallel with the study on critical design issues, DEMO design parameters need to be investigated to envisage possible DEMO options. Benchmark of systems codes independently developed by CCFE (Culham Centre for Fusion Energy) and JAEA (Japan Atomic Energy Agency) has been carried out, confirming good agreement
http://dx.doi.org/10.1016/j.fusengdes.2014.02.077 0920-3796/© 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: K. Tobita, et al., Research and development status on fusion DEMO reactor design under the Broader Approach, Fusion Eng. Des. (2014), http://dx.doi.org/10.1016/j.fusengdes.2014.02.077
G Model FUSION-7350; No. of Pages 5
ARTICLE IN PRESS
2
K. Tobita et al. / Fusion Engineering and Design xxx (2014) xxx–xxx
except several issues such as impurity radiation power, fast ion beta and bootstrap fraction [4,5]. This paper describes the main areas of the activity, this is, progress in DEMO design parameters, divertor and remote maintenance in the following sections. The plan for safety research for DEMO is briefly summarized in Section 5. 2. Parameter study of DEMO The basic approach of the BA DEMO design is to develop possible DEMO options that are foreseen from ITER, other on-going and planned R&D programmes. In this context, projections of ITER and the present knowledge to DEMO should be based on a certain broadening dependent on conservative and optimistic views. Another aspect to note is that EU and JA have different requirements for DEMO. Consequently, multiple DEMO concepts are likely to be developed in the BA. EU has explored pulsed and steady state power plants and actually a pulsed option was characterized as the most conservative model in the European Power Plant Conceptual Study (PPCS) [2]. In the BA activity, EU puts emphasis on a pulsed DEMO concept assuming small extension of ITER technology. Preliminary scoping study using an EU systems code is presently pursuing a pulsed DEMO generating net electricity (Pe ) of 0.5 GWe and fusion power (Pfus ) of around 2 GW at a major radius (Rp ) of 9 m, aspect ratio of about 4 and pulse length of as short as about 2 h. Previously, JA reactor study had focused on steady state DEMO concepts that were relatively compact and capable of producing plant-level electricity (∼1 GWe) [1,3]. The previous reactor study reveals that such compact reactors have problems in power removal in the divertor, removal of residual heat of the in-vessel components in a loss-of-coolant accident (LOCA) and tokamak operation capability due to insufficient volt-second supply. Based on the result, JA currently considers a relatively large and lower power steady state concept with sufficient volt-second supply for operational development from pulsed (∼0.5 h) to steady state, having parameters of Rp ∼ 8.2 m or more, Pfus = 1.3–1.5 GW and Pe = 0.2–0.3 GWe. In the scoping study, it is commonly recognized that a vertically stable plasma elongation under constraints of DEMO needs to be investigated, which means no use of in-vessel coils for vertical stabilization and the installation of stabilizing shells away from breeding blanket. 3. Divertor Divertor is one of the most critical design issues giving a significant impact on DEMO design parameters. The problem is how to handle a huge power exhausted from the main plasma being several times as high as that in ITER and eventually to maintain the divertor heat flux at a tolerable level well below 10 MW/m2 . In order to find a possible solution for the problem, two approaches have been progressing. The first approach is to pursue a fully detached plasma condition based on divertor simulation [6,7], and the second is to seek for the possibility of modifying divertor configuration such as super-X and snowflake [8]. 3.1. Divertor simulation Admitting that existing divertor simulation codes have not satisfactorily been validated by experiments, computer simulation is practically the only approach to foresee DEMO because the DEMO conditions are substantially far away from existing experiments. For the JA divertor design study, a divertor simulation code SONIC was upgraded to appropriately deal with thermal and friction forces
of impurity ions along the magnetic field lines. In the simulation, impurity transport and the resulting plasma detachment were simulated self-consistently for different detachment scenarios. 3.1.1. Impurity injection Impurity seeding using different species (Ne, Ar and Kr) is anticipated to change radiation distribution in the scrape-off layer (SOL) and the divertor because of the Z-dependence of radiation loss rate coefficient. The divertor simulation indicates that selecting the impurity species can control the distribution of radiation power in that, with seeding higher-Z impurity such as Kr, a wider plasma detachment region and lower divertor heat flux were obtained at the outer divertor [7]. On the other hand, higher-Z impurity tends to be transported to the main plasma and to cause higher radiation in the main edge, which may deteriorate plasma confinement. 3.1.2. Long leg divertor Previously, it was confirmed that the divertor geometry contributes to reducing divertor heat flux and that V-shaped corner is efficient to enhance particle recycling near the strike point and to produce detached plasma [9]. In order to reduce the divertor heat flux further, longer leg divertor was proposed and investigated for SlimCS [6]. The role of the long leg divertor is to decrease the divertor plasma temperature because of extended connection length to the target plate and to increase the heat receiving area of the divertor. Recent simulation [7] indicates that, when Ar is seeded to the divertor region of the plasma with a fusion output (Pfus ) of 3 GW, full detachment over the outer divertor is attained as shown in Fig. 1(a). Then, the radiation power fraction reaches 80% of the power exhausted to the SOL from the main plasma (500 MW) and the divertor electron and ion temperatures decrease to 2 eV. In the long leg divertor, the ionization front moves upstream (Fig. 1(b)), contributing to reducing the peak heat load on the divertor plate. 3.2. Alternative divertor configurations Another approach for reducing the divertor heat load is a flux tube expansion in the divertor with “super-X” or “snowflake” configuration. Application of such configurations has been investigated by EU [10] and JA [8]. Fig. 2 depicts the arrangement of poloidal field (PF) coils for the super-X configuration and the required coil currents, where the original plasma configuration is based on SlimCS with a plasma current of 16.8 MA. The super-X configuration is numerically obtained using external PF coils. However, the reality of coil current required for Coil #9 of 180 MA seems unlikely because the cross section for the current is anticipated to be 4–6 m2 when the conductor current is 60–100 kA. Development of high current conductor of 150–200 kA can allow the super-X with external PF coils but the leak field produced by such a high current coil may restrict the installation of peripheral equipment such as diagnostics, solenoid valves and pumps. By winding some of the PF coils in an inter-linked way against the toroidal field (TF) coils, the required coil currents are dramatically reduced as shown in Fig. 2(b). Inter-linked winding was originally proposed as a novel concept of central solenoid (CS) allowing a compact CS with a sufficient volt-second supply [11,12]. The winding needs to be carried out in situ in the middle of the TF coil installation and the vacuum vessel one. Although the engineering feasibility of the inter-linked winding must be verified from various aspects, the required coil currents have reality. Further study is under way from the points of view of the compatibility of the inter-linked coils and the neutron and thermal shield. Regarding the super-X divertor, it is noteworthy that the connection length along a magnetic field line near the super-X null point is extended. The connection length of the magnetic field line
Please cite this article in press as: K. Tobita, et al., Research and development status on fusion DEMO reactor design under the Broader Approach, Fusion Eng. Des. (2014), http://dx.doi.org/10.1016/j.fusengdes.2014.02.077
G Model FUSION-7350; No. of Pages 5
ARTICLE IN PRESS K. Tobita et al. / Fusion Engineering and Design xxx (2014) xxx–xxx
Fig. 1. Distribution of radiation power density in (a) the standard divertor with V-shaped corner, and (b) the long leg divertor.
of 0.001 m-away from the separatrix on the outer mid-plane is as long as 60 m, being much longer than those of the standard (36 m) and the long leg divertor (41 m) [8]. This may ease the positional control of the ionization front of detached plasma, which will be investigated using the SONIC code in the coming years. In contrast, the snowflake is likely to be ruled out of the possible divertor configurations for DEMO. Defects of the configuration is (1) extremely high coil currents in part of CS [8], (2) lower plasma volume for the same major radius due to the higher elongation, (3) concern on the vertical stability of such an elongated plasma [13]. 3.3. Divertor target plate Structure and materials of divertor target plates are another divertor issue to be resolved. Originally, JA considered a concept composed of tungsten mono-blocks and a cooling tube made of reduced activation ferritic martensitic (RAFM) steel using pressurized water as coolant [14]. On the other hand, EU had applied helium-cooled divertor in most of earlier EU reactor designs such as PPCS [15]. Despite such historical background, EU presently considers water-cooled divertor as a conservative option for DEMO because of its effective heat removal capability and technical maturity. Preliminary survey by EU suggested that copper (Cu)-based alloy with tungsten armor seems favorable as the water-cooled divertor. Of course, material development is indispensable for the purpose because no existing Cu-alloys withstand neutron irradiation of even as low as 1 dpa. Recently, JA assessed heat removal
3
Fig. 2. Super-X divertor configuration produced by (a) external poloidal field coils and (b) inter-linked poloidal field coils. Upper poloidal field coils are omitted.
capability of the RAFM-based water-cooled divertor, concluding that the allowable heat flux is equivalent to as low as 5 MW/m2 when the tungsten mono-block armors are tapered on the surface to avoid leading edge problem. Based on the result, JA is considering a combination of RAFM-based and Cu-alloy-based divertor for using both materials as the condition demands: RAFM-based for lower heat load and higher dpa zone such as the divertor dome, and Cu-alloy-based for higher heat load and lower dpa zone such as divertor wetted areas. The problem of the combined use of the materials is how to mitigate thermal stress in the arrangement of both in that the coolant temperature for RAFM and Cu-alloy should be different; typically 300 ◦ C for RAFM and 200–240 ◦ C for Cu-alloy. Since the design life of the water-cooled divertor with Cu-alloy is anticipated to be 1 year, its application will be limited to the early operation phase of DEMO. For the later phase, an alternative divertor concept needs to be developed. 4. Remote maintenance Maintenance is a critical design issue because remote maintenance (RM) on DEMO would be far beyond the present technology level. ITER chooses an in-vessel maintenance scheme where small blanket modules are replaced in the vacuum vessel using maintenance vehicles. The scheme is reasonable when a small number of damaged components are replaced. However, because of maintenance time, the scheme is not suitable for periodic replacement of the whole set of blanket modules as required for DEMO. In particular, the most time-consuming processes such as in situ cutting, re-welding and inspection need to be minimized to reduce the maintenance time. For this reason, it is necessary to develop a new
Please cite this article in press as: K. Tobita, et al., Research and development status on fusion DEMO reactor design under the Broader Approach, Fusion Eng. Des. (2014), http://dx.doi.org/10.1016/j.fusengdes.2014.02.077
G Model FUSION-7350; No. of Pages 5
ARTICLE IN PRESS
4
K. Tobita et al. / Fusion Engineering and Design xxx (2014) xxx–xxx
Fig. 3. (a) Sector transport using a limited number of horizontal maintenance port (S-LHP), and (b) MMS-based vertical port maintenance.
RM scheme which would be reliable and potentially allow us to attain favorable plant availability. Since RM scheme is closely coupled with the reactor configuration, it needs to be determined in early time of the DEMO conceptual design. 4.1. Maintenance schemes Recent design study suggests that RM on DEMO requires a major shift from the ITER scheme because of in-vessel dose rate (10–100 times higher than that of ITER) and required plant availability. This shift will also have a lot of influence on reactor configuration, reactor building, hot cell and RM equipment. JA has mainly investigated sector transport maintenance in which sectors assembling blanket modules and divertor cassettes are removed from the reactor and spare sectors are installed anew. In the RM scheme, most parts of cutting, re-welding and inspection of piping are carried out in the hot cell. Comparing three kinds of sector maintenance schemes of (i) sector transport using all horizontal maintenance ports (S-AHP), (ii) sector transport using a limited number of horizontal maintenance ports (S-LHP), (iii) sector transport using limited number of vertical maintenance ports (S-LVP) [16], it is concluded that S-LHP (Fig. 3(a)) will be less challenging than the other two. S-AHP has a serious problem in supporting the turnover force of TF coils, and S-LVP needs to resolve a difficult question on how to transport sectors in the toroidal direction after inserting them in the vacuum vessel. After a free thinking range of maintenance schemes including sector and Multi-Module Segments (MMS) maintenance in 2011, EU selected MMS-based vertical port maintenance (Fig. 3(b)) in which blanket modules are integrated on a banana-shaped inboard and outboard segment and each MMS is transported through a small vertical maintenance port [17,18]. Divertor cassettes are replaced like ITER using a small port located in the lower outboard. The design studies by JA and EU commonly pointed out that the hot cell facility of DEMO seems to be huge. EU’s assessment indicates the total floor area of 23,500 m2 for RM, component processing, storage, waste and recycling, being 6 times as large as the hot cell of ITER (4000 m2 ). JA’s assessment indicates the floor areas of 17,100 m2 (5300 m2 for hot cell and 11,800 m2 for interim storage). The evaluated areas will be updated in future investigation. The point is that waste management strategy associated with the maintenance scenario needs to be defined in the early stage of the conceptual design for the layout of peripheral equipment and facilities.
is to score various criteria including maintenance time, reliability, waste volume and contamination control for each RM scheme as proposed for ARIES-AT [19]. The problem is the weighting for each criterion. Fig. 4 illustrates a relation diagram depicting how the plant availability is determined. When a DEMO plant is operated in a scheduled manner, the maintenance period will be key to the availability. However, the plant is forced to stop operation due to a trouble or an accident, the resulting unscheduled shutdown could result in a serious deterioration of the availability. In this sense, how to improve the reliability of in-vessel components (IVC) would be key to attaining a reasonable availability and a considerably high weighting factor needs to given to the IVC reliability in the scoring approach in the selection of the most suitable RM scheme for DEMO. Installation of brand-new MMS or sectors in every replacement, which improves IVC reliability because of fabrication and inspection by hand in non-controlled areas, should be effective to attain a desirable availability. Then, remote work in radiologically controlled area is limited to cutting, re-welding and inspection of coolant manifolds of MMS or sectors. On the other hand, installation of brand-new MMS or sectors cause a problem of increase in radioactive waste, and the sector maintenance seems to produce a larger amount of radioactive waste than MMS. The discussion described above is an aspect of maintenance. For narrowing down the RM schemes, it is necessary to insightfully analyze maintenance requirements for DEMO as well as engineering feasibility. 5. Safety study Safety studies carried out in past decades have shown favorable features of fusion and indicated that the safety characteristics are strongly dependent on design choices including the combination
4.2. Viewpoint to narrow down the RM schemes The most important design issue on maintenance is methodology of how to select one of many possible options. An approach
Fig. 4. Schematic diagram indicating the impact of RM scheme on the plant availability.
Please cite this article in press as: K. Tobita, et al., Research and development status on fusion DEMO reactor design under the Broader Approach, Fusion Eng. Des. (2014), http://dx.doi.org/10.1016/j.fusengdes.2014.02.077
G Model FUSION-7350; No. of Pages 5
ARTICLE IN PRESS K. Tobita et al. / Fusion Engineering and Design xxx (2014) xxx–xxx
of materials such as structural materials, breeder, multiplier and coolant, and the design parameters of a reactor. In this context, it is still of great significance to investigate safety characteristics of DEMO assuming the most likely design parameters and materials. Safety study in the BA has been conducted in parallel with DEMO design since 2013. The purpose of the safety study is (1) to investigate the safety characteristics of DEMO through safety analysis on hypothetical accidents, (2) to define safety systems and their specifications required for DEMO to reduce or prevent radiological risks to the public, and (3) to formulate DEMO safety guidelines.
[5] [6]
[7] [8]
[9]
6. Summary [10]
The BA DEMO design activity has mainly addressed key design issues, options and DEMO parameters from a broad perspective. After assessing these issues, options and parameters, they will be narrowed down to develop pre-conceptual DEMO options. The activity will contribute to pointing a reasonable direction of DEMO that is foreseeable around 2030 s.
[11]
[12]
[13]
Acknowledgments [14]
This work was carried out within the framework of the Broader Approach DEMO Design Activity. Contribution by all members of JA and EU Home Teams for BA is greatly appreciated.
[15]
[16]
References [1] K. Tobita, S. Nishio, M. Enoeda, H. Kawashima, G. Kurita, H. Tanigawa, et al., Compact DEMO, SlimCS: design progress and issues, Nucl. Fus. 49 (2009) 075029. [2] D. Maisonnier, D. Campbell, I. Cook, L. Di Pace, L. Giancarli, J. Hayward, et al., Power plant conceptual studies in Europe, Nucl. Fus. 47 (2007) 1524. [3] R. Hiwatari, K. Okano, Y. Asaoka, K. Shinya, Y. Ogawa, Demonstration of tokamak fusion power plant for early realization of net electric power generation, Nucl. Fus. 45 (2005) 96. [4] R. Kemp, M. Nakamura, D.J. Ward, K. Tobita, G. Federici, Benchmarking reactor systems studies by comparison of EU and Japanese system code
[17]
[18]
[19]
5
results for different DEMO concepts, in: Proc. of 24th IAEA Fusion Energy Conference, FTP/P7-29 San Diego, 2012, http://www.naweb.iaea.org/ napc/physics/FEC/FEC2012/.papers/24 FTPP729.pdf K. Okano, G. Federici, K. Tobita, DEMO design activities in the Broader Approach under Japan/EU collaboration, Fus. Eng. Des. (2014). K. Hoshino, K. Shimizu, N. Asakura, T. Takizuka, M. Nakamura, K. Tobita, Simulation study of an extended divertor leg for heat control in the SlimCS DEMO reactor, Contr. Plasma Phys. 52 (2012) 550. N. Asakura, et al., A simulation study of large power handling in the divertor DEMO reactor, Nucl. Fus. 53 (2013) 123013. N. Asakura, K. Shinya, K. Tobita, K. Hoshino, Investigation of advanced divertor configuration for DEMO tokamak reactor, Trans. Fus. Sci. Technol. 63 (2013) 70. H. Kawashima, K. Shimizu, T. Takizuka, K. Tobita, S. Nishio, S. Sakurai, et al., Simulation study for divertor design to handle huge exhaust power in the SlimCS DEMO rector, Nucl. Fus. 49 (2009) 065007. J.W. Coenen, Liquid metals as alternative solution for the power exhaust of future fusion devices: status and perspective, J. Nucl. Mater. (2014) (still in print). H. Utoh, N. Asakura, K. Tobita, Y. Someya, Y. Sakamoto, K. Hoshino, M. Nakamura, Application of inter-linked superconducting coils for central solenoid and advanced divertor configuration of DEMO, Fus. Eng. Des. (2014). M. Nakamura, K. Tobita, H. Utoh, Y. Sakamoto, Impact of an ‘interlinked’ central solenoid configuration on the size of tokamak DEMO, Fus. Eng. Des. 88 (2013) 1146. G. Federici, R. Kemp, D. Ward, C. Bachmann, T. Franke, S. Gonzalez, et al., Overview of EU DEMO design and R&D activities, Fus. Eng. Des. (2014) http://dx.doi.org/10.1016/j.fusengdes.2014.01.070 K. Tobita, S. Nishio, M. Enoeda, M. Sato, T. Isono, S. Sakurai, et al., Design study of fusion DEMO plant at JAERI, Fus. Eng. Des. 81 (2006) 1151. P. Norajitra, R. Giniyatulin, N. Holstein, T. Ihli, K. Krauss, R. Kruessmann, et al., Status of He-cooled diverter development for DEMO, Fus. Eng. Des. 75–79 (2005) 307. H. Utoh, Y. Somaya, K. Tobita, N. Asakura, K. Hoshino, M. Nakamura, Critical design factors for sector transport maintenance in DEMO, Nucl. Fus. 53 (2013) 23005. A. Loving, O. Crofts, N. Sykes, D. Iglesias, M. Coleman, J. Thomas, et al., Preconceptual design assessment of DEMO remote maintenance, Fus. Eng. Des. (2014). M. Coleman, N. Sykes, D. Cooper, D. Iglesias, R. Bastow, A. Loving, J. Harman, Concept for a vertical maintenance remote handling system for multi module blanket segments in DEMO, Fus. Eng. Des. (2014) http://dx.doi.org/ 10.1016/j.fusengdes.2014.02.047 L.M. Waganer, The ARIES Team, ARIES-AT maintenance system definition and analysis, Fus. Eng. Des. 80 (2006) 161.
Please cite this article in press as: K. Tobita, et al., Research and development status on fusion DEMO reactor design under the Broader Approach, Fusion Eng. Des. (2014), http://dx.doi.org/10.1016/j.fusengdes.2014.02.077