- Email: [email protected]
Simplification of blanket system for SlimCS fusion DEMO reactor
Fusion Engineering and Design 86 (2011) 2269–2272
Contents lists available at ScienceDirect
Fusion Engineering and Design journal homepage: www.elsevier.com/locate/fusengdes
Simplification of blanket system for SlimCS fusion DEMO reactor Youji Someya ∗ , Haruhiko Takase, Hiroyasu Utoh, Kenji Tobita, Changle Liu, Nobuyuki Asakura Japan Atomic Energy Agency, 801-1 Mukoyama, Naka, Ibaraki 311-0193, Japan
a r t i c l e
i n f o
Article history: Available online 2 March 2011 Keywords: Mixed breeder blanket Demo SlimCS Neutronics Thermal analysis
a b s t r a c t Simplification of blanket system is necessary for a fusion DEMO reactor. Although the conceptual design of a tritium-breeding blanket for SlimCS has been studied in the past several years, the structure of the previous blanket seems to be complex and difficult to manufacture from the viewpoint of engineering. In this paper, we proposed simplification of blanket structure without decreasing the net Tritium Breeding Ratio (TBR). In the proposed concept, the blanket is filled with the mixture of Li4 SiO4 pebbles or Li2 O pebbles for the tritium breeding and Be12 Ti pebbles for the neutron multiplication. To confirm the effectiveness of this concept, an ANIHEAT code with the nuclear library FENDL-2.0 was used for calculations of the neutronic and thermal analyses. The result indicated that, under the constraint of the blanket thickness being less than 0.5 m, the mixture of Li2 O pebbles and Be12 Ti ones is the most effective and that the TBR is expected to be greater than 1.05. © 2011 Elsevier B.V. All rights reserved.
1. Introduction SlimCS is the conceptual design of a tokamak DEMO reactor with low aspect ratio. The major parameters of SlimCS are a plasma major and minor radius of 5.5 m and 2.1 m, respectively, an aspect ratio of 2.6 and a fusion power of 2.95 GW [1]. The previous blanket for SlimCS consists of the replaceable and permanent blanket. The conducting shell for plasma positional stability and high beta access is installed in between replaceable and permanent blanket as shown in Fig. 1. Here ap is the plasma minor radius and rwall is the distance between the center of the plasma and the center of conducting shell. The material of conducting shell is reduced-activation martensitic steel (F82H [2]), and the thickness in a radial direction is 0.07 m. The permanent blanket contains F82H and water for the neutron shield, and the ratio of F82H to water is assumed to be 70–30%. The replaceable blanket with neutron multiplier Beplate is designed to be as thin as possible with keeping high Tritium Breeding Ratio (TBR) since it is important for high plasma performance to bring the conducting shell close to plasma as possible. Fig. 2 shows the interior design of the replaceable blanket for SlimCS, which is based on multilayer concept [3]. However, the multilayered concept has engineering difficulties in welding of cooling tubes and the F82H casing for Be-plate, pebbles packing in narrow regions and inspection after fabrication. Therefore, simple structure is important from the viewpoint of engineering feasibility. At the same time, it is necessary to ensure the self-sufficient production of tritium, as well. In this paper, the
∗ Corresponding author. E-mail address: [email protected] (Y. Someya). 0920-3796/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.fusengdes.2011.01.141
simplification of blanket structure was proposed from the viewpoint of mixed breeder blanket. 2. Simplification of blanket structure 2.1. Proposed blanket concept In the previous design, the neutron multiplier (Be) with separated by F82H casing from the lithium ceramic (Li4 SiO4 ) to avoid a reductive degradation of the ceramic. When Be12 Ti is used as the multiplier, the separation between the multiplier and the lithium ceramic can be removed, because Be12 Ti is chemically stable. The blanket is filled with the mixture of Li4 SiO4 pebbles or Li2 O pebbles for the tritium breeding and Be12 Ti pebbles for the neutron multiplication. Fig. 3 shows the interior design of the proposed blanket for SlimCS. Both pebbles are mixed in the blanket without partition. The blanket crate is fabricated by hot isostatic pressing (HIP) using F82H to form the square cooling channel structure (8 mm × 8 mm). The cooling tubes in the blanket are welded on the header in the back. Therefore, there are no welding points in the forward blanket area, which minimizes the risk of irradiation damage of welding lines. A remarkable feature of the proposed blanket is the simple structure in that structural materials in the blanket are cooling tubes and support for them only. As a result, this concept has a possibility of increasing TBR due to a relative increase the fraction of breeding materials. On the other hand, there is a concern that the TBR may decrease because of reducing the Be (n, 2n) reactivity as a result of the slowdown of neutrons (above 1.8 MeV) by Ti of Be12 Ti. Hence, the nuclear characteristics of the proposed blanket need to be understood. In case of proposed blanket, the total neutron and gamma ray fluxes attenuate by 1–2 orders of magni-
2270
Y. Someya et al. / Fusion Engineering and Design 86 (2011) 2269–2272
Fig. 1. Structure of the blanket sector for SlimCS.
tude between inlet and outlet in radial blanket thickness of 0.45 m. When the average neutron wall load (NWL) was 3.0 MW/m2 , the displacement rate per atom (dpa) of the conducting shell was about 0.5 dpa/FPY. Therefore, the conducting shell is a single structure by F82H and is assumed to be tolerant of irradiation effects as well as the blanket structure material. 2.2. Target of design The conducting shell needs to be located at rwall /ap ≤ 1.35 for plasma positional stability and high beta access. In the case of SlimCS with ap = 2.1 m, the thickness of blanket should be 0.5 m or less to meet rwall /ap ≤ 1.35, when the gap between the separatrix and the first wall is 0.15 m as shown in Fig. 1. The design target of this study is to satisfy the net TBR of ≥1.05 using the blanket with the thickness of 0.5 m or less.
Fig. 3. Proposed blanket structure for SlimCS.
3. Calculation conditions 3.1. Calculation model The sizes and arrangement of the cooling tubes for the proposed blanket were changed in accordance with the NWL. Actually, the blanket was approximated by a slab model for the calculations as shown in Fig. 4. The cooling tubes were replaced with slabs having the equivalent cross section. In the 1-D calculations of the neutronic and thermal analysis for the blanket, the ANIHEAT code with the nuclear library FENDL-2.0 [4] was used. The neutronics was calculated on local TBR and nuclear heating in the blanket. The temperature of blanket was evaluated by the 1-D thermal conduction equation. The local TBR is evaluated by changing NWL from 1 to 5 MW/m2 . The heat load is fixed at 0.5 MW/m2 . The material of first wall assumes F82H in this calculation. The thickness of each layer was determined to satisfy the operation temperature of materials, as described in Section 3.3. 3.2. Optimal ratio of breeding material to Be12 Ti Fig. 5 shows the results of the local TBR when the blanket is assumed to be homogeneous. The x-axis is the ratio of breeder,
Fig. 2. Replaceable blanket for SlimCS.
Fig. 4. (a) Proposed blanket structure and (b) 1-D model for mixed breeder blanket.
Y. Someya et al. / Fusion Engineering and Design 86 (2011) 2269–2272
2271
1.5
Li4SiO4 & Be12Ti
1.4
Local TBR
Li2O & Be12Ti 1.3 1.2 1.1 1.0
0
20
40
60
80
100
The ratio of breeder to multiplier, BMR [%] Fig. 5. Local TBR with ratio of mixed breeder.
defined as follows: BMR =
breeder × 100 [%], breeder + multiplier
where the breeder and multiplier are assumed to be Li4 SiO4 or Li2 O, and Be12 Ti, respectively. The calculation conditions were as follows: 6 Li enrichment of 90%, a packing fraction of 80% and the blanket for thickness of 0.80 m. In addition to the breeder and the multiplier, the blanket model contains F82H and water, the fraction of F82H and water was 20%, and the ratio of F82H to water was assumed to be 70–30%. The result indicates that the local TBR becomes a maximum value when the BMR is between 10% and 20%. In the case of the proposed blankets, the maximum TBR is obtained at the BMR of about 15% for both cases. In these conditions, ratio of breeder, multiplier and He gas were 12%, 68% and 20%, respectively. The effective thermal conductivity of the mixed pebbles is defined by the combination of the conductivities of breeder and multiplier pebble beds [5–7] in accordance with the BMR as shown in Fig. 6. 3.3. Conditions of coolant The coolant flows in the toroidal direction. The coolant was assumed to be sub-critical water conditions of 23 MPa and T = 70 ◦ C (290–360 ◦ C). The upper coolant velocity was limited to 6 m/s and the outlet temperature was less than 360 ◦ C. The operation temperature of Li4 SiO4 , Li2 O, Be12 Ti and F82H was limited to 900 ◦ C, 700 ◦ C, 900 ◦ C and 550 ◦ C, respectively.
Fig. 7. MCNP calculation model for SlimCS: (a) vertical cross section in the 3-D model and (b) horizontal cross section at mid-plane.
TBR. This is because the thickness of each layer is dependent on the resulting nuclear heating to meet the operation temperature of blanket materials. In this section, the NWL was calculated for SlimCS by the 3-D Monte Carlo N-particle transport code MCNP-5 [8] with the nuclear data library ENDF/B-VII [9]. Fig. 7 shows the 3-D MCNP calculation model for SlimCS. The model includes the geometrical arrangement of the inboard (IB) and outboard (OB) blanket, the divertor, the central solenoid (CS) and toroidal field (TF) coils. By assuming toroidal axisymmetry, 1/24 sector of the reactor was modeled with reflecting boundaries. The surface area of IB blanket, OB blanket and divertor were 179 m2 , 489 m2 and 326 m2 , respectively. The neutron volume source for plasma emitted neutron with the energy of 14.06 MeV. Fig. 8 shows the layout of the blanket modules along the poloidal direction. The thickness of IB blanket was fixed at 0.3 m. The blanket coverage loss by the divertor is 11.8% and loss by the ports is 1%. In addition, rims, ribs and gap of the structure in the blanket coverage are 7.3%, 3% and 1%, respectively. Therefore, the total coverage of blanket is 75.9%. Fig. 9 shows the poloidal distribution of the NWL. The peak NWL in the IB and OB blanket are 2.93 and 3.82 MW/m2 , respectively. The
3.4. Poloidal distribution of neutron wall load
Thermal Conductivity [W/mK]
Since the NWL changes in the poloidal direction, the thickness of each layer in the blanket needs to be optimized to maximize
3.5
Li2O & Be12Ti
3.0 2.5
Li4SiO4 & Be12Ti
2.0 1.5 1.0 0.5 0.0
0
200
400
600
800
1000
Temperature [ºC] Fig. 6. Thermal conductivities of breeder and multiplier.
Fig. 8. Layout of blanket modules.
2272
Y. Someya et al. / Fusion Engineering and Design 86 (2011) 2269–2272
1.10
Li2O & Be12Ti Net TBR
1.00
0.90
0.80
Previous Blanket
Li4SiO4 & Be12Ti
0
10
20
30
40
50
60
70
80
Thickness of blanket [cm] Fig. 9. Distribution of neutron wall load for SlimCS.
Fig. 11. Dependence of net TBR on blanket thickness.
1.6
Li2O & Be12Ti Local TBR
1.5
Li4SiO4 & Be12Ti
1.4
Previous Blanket
1.3
1.2
0
1
2
3
4
5
6
Fig. 9 and the dependence of local TBR on the NWL. In Fig. 11, the thickness of the OB blanket is changed from 0.3 to 0.6 m and the thickness of the IB blanket is fixed at 0.3 m. Fig. 11 shows that the net TBR of Li2 O&Be12 Ti is higher than those of the other cases. The reason why Li2 O&Be12 Ti shows higher TBR than Li4 SiO4 &Be12 Ti is that Li2 O&Be12 Ti was different from Li4 SiO4 &Be12 Ti in having the high content of 6 Li. The Li2 O&Be12 Ti attained the target of the net TBR (≥1.05) at the blanket thickness of 0.48 m, and the Li2 O&Be12 Ti can keep location for conducting shell (rwall /ap ≤ 1.35). 6. Conclusions
Neutron wall load [MW/m2] Fig. 10. Dependence of local TBR on NWL for blanket of 0.6 m in thickness.
average NWL in the IB and OB blanket are 2.44 and 3.50 MW/m2 , respectively. 4. Dependence of local TBR on NWL The local TBR was evaluated by changing NWL from 1 to 5 MW/m2 as shown in Fig. 10. In Fig. 10, the heat load and the thickness of blanket were fixed at 0.5 MW/m2 and 0.6 m, respectively. The previous blanket is based on the solid breeder (Li4 SiO4 ) with Be-plate, as shown in Fig. 2. In Fig. 10, in all cases, the tendency of local TBR are similar. With decreaseing the NWL from 5 to 1 MW/m2 , the local TBR is improved because of a reduction of the coolant area in the blanket. The local TBR of previous blanket is higher than those of Li4 SiO4 &Be12 Ti and Li2 O&Be12 Ti near the first wall. The local TBR of Li4 SiO4 &Be12 Ti and Li2 O&Be12 Ti are higher than those of the previous blanket in other area. The local TBR of Li4 SiO4 &Be12 Ti is higher than those of the other cases at the 5 MW/m2 . For the improvement of the TBR, the first breeding area near the first wall should be as wide as possible, because an abundance of neutrons (below 100 keV) was effective to tritium production from the neutron scattering in the inside of the first wall. The maximum thickness of first breeding area was Li4 SiO4 &Be12 Ti. Because the maximum operation temperature of the Li4 SiO4 was higher than those of Li2 O.
For not only SlimCS but also nuclear fusion plant, it is favorable that the easiness of manufacture coexists with high TBR. For solving this problem, the simplification of blanket structure was proposed in this paper. The proposed blanket structure is that Li4 SiO4 pebbles or Li2 O pebbles for tritium breeding and Be12 Ti pebbles for neutron multiplication are mixed and these pebbles are filled in the blanket. We confirmed the effectiveness of the proposed blanket structure in the case of mixture of Li2 O and Be12 Ti pebbles by the 1-D calculation based on the slab model with neutronic and thermal analyses. As a result, Li2 O&Be12 Ti attained the target of the net TBR (≥1.05) at the blanket thickness of 0.48 m, and Li2 O&Be12 Ti can satisfy the location required for location for conducting shell (rwall /ap ≤ 1.35). Hence, this mixed breeder blanket is preferable for the SlimCS DEMO reactor. On the other hand, estimation of Loss-Of-CoolantAccident (LOCA) is important for the proposed blanket. Considering these points, the blanket assumed that several ribs were located in the blanket module. Therefore, the analysis of LOCA in the proposed blanket needs to be done in near future and reconfirm the effectiveness. References [1] [2] [3] [4] [5] [6] [7] [8]
5. Dependence of net TBR on blanket thickness Fig. 11 shows the dependence of the net TBR. The net TBR is calculated by considering the poloidal distribution of NWL in
[9]
K. Tobita, et al., Nucl. Fusion 49 (2009) 075029. S. Jitsukawa, et al., J. Nucl. Mater. 307–311 (2002) 179–186. M. Enoeda, et al., Nucl. Fusion 43 (2003) 1837–1844. A.B. Pashshenko, IAEA Report INDC (NDS)-352, 1996. M. Billone, Energy Tech. Division/Fusion Power Program/Engineering Physics Division. ITER Solid Breeder Blanket Materials Database, 1993. O. Krikorian, High Temp. High Press. 17 (1985) 161. M. Uchida Fusion, Eng. Des. 69 (2003) 499–503. X-5 Monte Carlo Team, MCNP-A General Monte Carlo N-Particle Transport Code, Version 5, Vol. II: Users Guide, La-CP-03-0245, Los Alamos National Laboratory, 2003. M.B. Chadwick, P. Oblozinsky, M Herman, et al., ENDF/B-VII.0, Nuclear Data Sheets 107 (2006) 2931–3060.
Contents lists available at ScienceDirect
Fusion Engineering and Design journal homepage: www.elsevier.com/locate/fusengdes
Simplification of blanket system for SlimCS fusion DEMO reactor Youji Someya ∗ , Haruhiko Takase, Hiroyasu Utoh, Kenji Tobita, Changle Liu, Nobuyuki Asakura Japan Atomic Energy Agency, 801-1 Mukoyama, Naka, Ibaraki 311-0193, Japan
a r t i c l e
i n f o
Article history: Available online 2 March 2011 Keywords: Mixed breeder blanket Demo SlimCS Neutronics Thermal analysis
a b s t r a c t Simplification of blanket system is necessary for a fusion DEMO reactor. Although the conceptual design of a tritium-breeding blanket for SlimCS has been studied in the past several years, the structure of the previous blanket seems to be complex and difficult to manufacture from the viewpoint of engineering. In this paper, we proposed simplification of blanket structure without decreasing the net Tritium Breeding Ratio (TBR). In the proposed concept, the blanket is filled with the mixture of Li4 SiO4 pebbles or Li2 O pebbles for the tritium breeding and Be12 Ti pebbles for the neutron multiplication. To confirm the effectiveness of this concept, an ANIHEAT code with the nuclear library FENDL-2.0 was used for calculations of the neutronic and thermal analyses. The result indicated that, under the constraint of the blanket thickness being less than 0.5 m, the mixture of Li2 O pebbles and Be12 Ti ones is the most effective and that the TBR is expected to be greater than 1.05. © 2011 Elsevier B.V. All rights reserved.
1. Introduction SlimCS is the conceptual design of a tokamak DEMO reactor with low aspect ratio. The major parameters of SlimCS are a plasma major and minor radius of 5.5 m and 2.1 m, respectively, an aspect ratio of 2.6 and a fusion power of 2.95 GW [1]. The previous blanket for SlimCS consists of the replaceable and permanent blanket. The conducting shell for plasma positional stability and high beta access is installed in between replaceable and permanent blanket as shown in Fig. 1. Here ap is the plasma minor radius and rwall is the distance between the center of the plasma and the center of conducting shell. The material of conducting shell is reduced-activation martensitic steel (F82H [2]), and the thickness in a radial direction is 0.07 m. The permanent blanket contains F82H and water for the neutron shield, and the ratio of F82H to water is assumed to be 70–30%. The replaceable blanket with neutron multiplier Beplate is designed to be as thin as possible with keeping high Tritium Breeding Ratio (TBR) since it is important for high plasma performance to bring the conducting shell close to plasma as possible. Fig. 2 shows the interior design of the replaceable blanket for SlimCS, which is based on multilayer concept [3]. However, the multilayered concept has engineering difficulties in welding of cooling tubes and the F82H casing for Be-plate, pebbles packing in narrow regions and inspection after fabrication. Therefore, simple structure is important from the viewpoint of engineering feasibility. At the same time, it is necessary to ensure the self-sufficient production of tritium, as well. In this paper, the
∗ Corresponding author. E-mail address: [email protected] (Y. Someya). 0920-3796/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.fusengdes.2011.01.141
simplification of blanket structure was proposed from the viewpoint of mixed breeder blanket. 2. Simplification of blanket structure 2.1. Proposed blanket concept In the previous design, the neutron multiplier (Be) with separated by F82H casing from the lithium ceramic (Li4 SiO4 ) to avoid a reductive degradation of the ceramic. When Be12 Ti is used as the multiplier, the separation between the multiplier and the lithium ceramic can be removed, because Be12 Ti is chemically stable. The blanket is filled with the mixture of Li4 SiO4 pebbles or Li2 O pebbles for the tritium breeding and Be12 Ti pebbles for the neutron multiplication. Fig. 3 shows the interior design of the proposed blanket for SlimCS. Both pebbles are mixed in the blanket without partition. The blanket crate is fabricated by hot isostatic pressing (HIP) using F82H to form the square cooling channel structure (8 mm × 8 mm). The cooling tubes in the blanket are welded on the header in the back. Therefore, there are no welding points in the forward blanket area, which minimizes the risk of irradiation damage of welding lines. A remarkable feature of the proposed blanket is the simple structure in that structural materials in the blanket are cooling tubes and support for them only. As a result, this concept has a possibility of increasing TBR due to a relative increase the fraction of breeding materials. On the other hand, there is a concern that the TBR may decrease because of reducing the Be (n, 2n) reactivity as a result of the slowdown of neutrons (above 1.8 MeV) by Ti of Be12 Ti. Hence, the nuclear characteristics of the proposed blanket need to be understood. In case of proposed blanket, the total neutron and gamma ray fluxes attenuate by 1–2 orders of magni-
2270
Y. Someya et al. / Fusion Engineering and Design 86 (2011) 2269–2272
Fig. 1. Structure of the blanket sector for SlimCS.
tude between inlet and outlet in radial blanket thickness of 0.45 m. When the average neutron wall load (NWL) was 3.0 MW/m2 , the displacement rate per atom (dpa) of the conducting shell was about 0.5 dpa/FPY. Therefore, the conducting shell is a single structure by F82H and is assumed to be tolerant of irradiation effects as well as the blanket structure material. 2.2. Target of design The conducting shell needs to be located at rwall /ap ≤ 1.35 for plasma positional stability and high beta access. In the case of SlimCS with ap = 2.1 m, the thickness of blanket should be 0.5 m or less to meet rwall /ap ≤ 1.35, when the gap between the separatrix and the first wall is 0.15 m as shown in Fig. 1. The design target of this study is to satisfy the net TBR of ≥1.05 using the blanket with the thickness of 0.5 m or less.
Fig. 3. Proposed blanket structure for SlimCS.
3. Calculation conditions 3.1. Calculation model The sizes and arrangement of the cooling tubes for the proposed blanket were changed in accordance with the NWL. Actually, the blanket was approximated by a slab model for the calculations as shown in Fig. 4. The cooling tubes were replaced with slabs having the equivalent cross section. In the 1-D calculations of the neutronic and thermal analysis for the blanket, the ANIHEAT code with the nuclear library FENDL-2.0 [4] was used. The neutronics was calculated on local TBR and nuclear heating in the blanket. The temperature of blanket was evaluated by the 1-D thermal conduction equation. The local TBR is evaluated by changing NWL from 1 to 5 MW/m2 . The heat load is fixed at 0.5 MW/m2 . The material of first wall assumes F82H in this calculation. The thickness of each layer was determined to satisfy the operation temperature of materials, as described in Section 3.3. 3.2. Optimal ratio of breeding material to Be12 Ti Fig. 5 shows the results of the local TBR when the blanket is assumed to be homogeneous. The x-axis is the ratio of breeder,
Fig. 2. Replaceable blanket for SlimCS.
Fig. 4. (a) Proposed blanket structure and (b) 1-D model for mixed breeder blanket.
Y. Someya et al. / Fusion Engineering and Design 86 (2011) 2269–2272
2271
1.5
Li4SiO4 & Be12Ti
1.4
Local TBR
Li2O & Be12Ti 1.3 1.2 1.1 1.0
0
20
40
60
80
100
The ratio of breeder to multiplier, BMR [%] Fig. 5. Local TBR with ratio of mixed breeder.
defined as follows: BMR =
breeder × 100 [%], breeder + multiplier
where the breeder and multiplier are assumed to be Li4 SiO4 or Li2 O, and Be12 Ti, respectively. The calculation conditions were as follows: 6 Li enrichment of 90%, a packing fraction of 80% and the blanket for thickness of 0.80 m. In addition to the breeder and the multiplier, the blanket model contains F82H and water, the fraction of F82H and water was 20%, and the ratio of F82H to water was assumed to be 70–30%. The result indicates that the local TBR becomes a maximum value when the BMR is between 10% and 20%. In the case of the proposed blankets, the maximum TBR is obtained at the BMR of about 15% for both cases. In these conditions, ratio of breeder, multiplier and He gas were 12%, 68% and 20%, respectively. The effective thermal conductivity of the mixed pebbles is defined by the combination of the conductivities of breeder and multiplier pebble beds [5–7] in accordance with the BMR as shown in Fig. 6. 3.3. Conditions of coolant The coolant flows in the toroidal direction. The coolant was assumed to be sub-critical water conditions of 23 MPa and T = 70 ◦ C (290–360 ◦ C). The upper coolant velocity was limited to 6 m/s and the outlet temperature was less than 360 ◦ C. The operation temperature of Li4 SiO4 , Li2 O, Be12 Ti and F82H was limited to 900 ◦ C, 700 ◦ C, 900 ◦ C and 550 ◦ C, respectively.
Fig. 7. MCNP calculation model for SlimCS: (a) vertical cross section in the 3-D model and (b) horizontal cross section at mid-plane.
TBR. This is because the thickness of each layer is dependent on the resulting nuclear heating to meet the operation temperature of blanket materials. In this section, the NWL was calculated for SlimCS by the 3-D Monte Carlo N-particle transport code MCNP-5 [8] with the nuclear data library ENDF/B-VII [9]. Fig. 7 shows the 3-D MCNP calculation model for SlimCS. The model includes the geometrical arrangement of the inboard (IB) and outboard (OB) blanket, the divertor, the central solenoid (CS) and toroidal field (TF) coils. By assuming toroidal axisymmetry, 1/24 sector of the reactor was modeled with reflecting boundaries. The surface area of IB blanket, OB blanket and divertor were 179 m2 , 489 m2 and 326 m2 , respectively. The neutron volume source for plasma emitted neutron with the energy of 14.06 MeV. Fig. 8 shows the layout of the blanket modules along the poloidal direction. The thickness of IB blanket was fixed at 0.3 m. The blanket coverage loss by the divertor is 11.8% and loss by the ports is 1%. In addition, rims, ribs and gap of the structure in the blanket coverage are 7.3%, 3% and 1%, respectively. Therefore, the total coverage of blanket is 75.9%. Fig. 9 shows the poloidal distribution of the NWL. The peak NWL in the IB and OB blanket are 2.93 and 3.82 MW/m2 , respectively. The
3.4. Poloidal distribution of neutron wall load
Thermal Conductivity [W/mK]
Since the NWL changes in the poloidal direction, the thickness of each layer in the blanket needs to be optimized to maximize
3.5
Li2O & Be12Ti
3.0 2.5
Li4SiO4 & Be12Ti
2.0 1.5 1.0 0.5 0.0
0
200
400
600
800
1000
Temperature [ºC] Fig. 6. Thermal conductivities of breeder and multiplier.
Fig. 8. Layout of blanket modules.
2272
Y. Someya et al. / Fusion Engineering and Design 86 (2011) 2269–2272
1.10
Li2O & Be12Ti Net TBR
1.00
0.90
0.80
Previous Blanket
Li4SiO4 & Be12Ti
0
10
20
30
40
50
60
70
80
Thickness of blanket [cm] Fig. 9. Distribution of neutron wall load for SlimCS.
Fig. 11. Dependence of net TBR on blanket thickness.
1.6
Li2O & Be12Ti Local TBR
1.5
Li4SiO4 & Be12Ti
1.4
Previous Blanket
1.3
1.2
0
1
2
3
4
5
6
Fig. 9 and the dependence of local TBR on the NWL. In Fig. 11, the thickness of the OB blanket is changed from 0.3 to 0.6 m and the thickness of the IB blanket is fixed at 0.3 m. Fig. 11 shows that the net TBR of Li2 O&Be12 Ti is higher than those of the other cases. The reason why Li2 O&Be12 Ti shows higher TBR than Li4 SiO4 &Be12 Ti is that Li2 O&Be12 Ti was different from Li4 SiO4 &Be12 Ti in having the high content of 6 Li. The Li2 O&Be12 Ti attained the target of the net TBR (≥1.05) at the blanket thickness of 0.48 m, and the Li2 O&Be12 Ti can keep location for conducting shell (rwall /ap ≤ 1.35). 6. Conclusions
Neutron wall load [MW/m2] Fig. 10. Dependence of local TBR on NWL for blanket of 0.6 m in thickness.
average NWL in the IB and OB blanket are 2.44 and 3.50 MW/m2 , respectively. 4. Dependence of local TBR on NWL The local TBR was evaluated by changing NWL from 1 to 5 MW/m2 as shown in Fig. 10. In Fig. 10, the heat load and the thickness of blanket were fixed at 0.5 MW/m2 and 0.6 m, respectively. The previous blanket is based on the solid breeder (Li4 SiO4 ) with Be-plate, as shown in Fig. 2. In Fig. 10, in all cases, the tendency of local TBR are similar. With decreaseing the NWL from 5 to 1 MW/m2 , the local TBR is improved because of a reduction of the coolant area in the blanket. The local TBR of previous blanket is higher than those of Li4 SiO4 &Be12 Ti and Li2 O&Be12 Ti near the first wall. The local TBR of Li4 SiO4 &Be12 Ti and Li2 O&Be12 Ti are higher than those of the previous blanket in other area. The local TBR of Li4 SiO4 &Be12 Ti is higher than those of the other cases at the 5 MW/m2 . For the improvement of the TBR, the first breeding area near the first wall should be as wide as possible, because an abundance of neutrons (below 100 keV) was effective to tritium production from the neutron scattering in the inside of the first wall. The maximum thickness of first breeding area was Li4 SiO4 &Be12 Ti. Because the maximum operation temperature of the Li4 SiO4 was higher than those of Li2 O.
For not only SlimCS but also nuclear fusion plant, it is favorable that the easiness of manufacture coexists with high TBR. For solving this problem, the simplification of blanket structure was proposed in this paper. The proposed blanket structure is that Li4 SiO4 pebbles or Li2 O pebbles for tritium breeding and Be12 Ti pebbles for neutron multiplication are mixed and these pebbles are filled in the blanket. We confirmed the effectiveness of the proposed blanket structure in the case of mixture of Li2 O and Be12 Ti pebbles by the 1-D calculation based on the slab model with neutronic and thermal analyses. As a result, Li2 O&Be12 Ti attained the target of the net TBR (≥1.05) at the blanket thickness of 0.48 m, and Li2 O&Be12 Ti can satisfy the location required for location for conducting shell (rwall /ap ≤ 1.35). Hence, this mixed breeder blanket is preferable for the SlimCS DEMO reactor. On the other hand, estimation of Loss-Of-CoolantAccident (LOCA) is important for the proposed blanket. Considering these points, the blanket assumed that several ribs were located in the blanket module. Therefore, the analysis of LOCA in the proposed blanket needs to be done in near future and reconfirm the effectiveness. References [1] [2] [3] [4] [5] [6] [7] [8]
5. Dependence of net TBR on blanket thickness Fig. 11 shows the dependence of the net TBR. The net TBR is calculated by considering the poloidal distribution of NWL in
[9]
K. Tobita, et al., Nucl. Fusion 49 (2009) 075029. S. Jitsukawa, et al., J. Nucl. Mater. 307–311 (2002) 179–186. M. Enoeda, et al., Nucl. Fusion 43 (2003) 1837–1844. A.B. Pashshenko, IAEA Report INDC (NDS)-352, 1996. M. Billone, Energy Tech. Division/Fusion Power Program/Engineering Physics Division. ITER Solid Breeder Blanket Materials Database, 1993. O. Krikorian, High Temp. High Press. 17 (1985) 161. M. Uchida Fusion, Eng. Des. 69 (2003) 499–503. X-5 Monte Carlo Team, MCNP-A General Monte Carlo N-Particle Transport Code, Version 5, Vol. II: Users Guide, La-CP-03-0245, Los Alamos National Laboratory, 2003. M.B. Chadwick, P. Oblozinsky, M Herman, et al., ENDF/B-VII.0, Nuclear Data Sheets 107 (2006) 2931–3060.