Transcript US-J WS on Reactor Design

Present activities on LHD-type
Reactor Designs FFHR
Akio SAGARA,
National Institute for Fusion Science, Japan
Contributed with
S. Imagawa, O. Mitarai*, T. Dolan, T. Tanaka,
Y. Kubota, K. Yamazaki, K. Y. Watanabe, N. Mizuguchi,
T. Muroga, N. Noda, O. Kaneko, H. Yamada, N. Ohyabu,
T. Uda, A. Komori, S. Sudo, and O. Motojima
Japan-US Workshop on
Fusion Power Plants and Related Advanced Technologies
with participation of EU
January 11-13, 2005 at Tokyo, JAPAN
Presentation Outline
Direction of Compactness
Magnetic field Bax (T)
15
FFHR2
Ap ~ 8
10
FFHR2m1
Ap ~ 8
LHD
Ap ~ 6
5
This
work
FFHR2m2
Ap ~ 6
0
0
5
10
15
20
Coil major radius R (m )
1. Introduction of FFHR
2. New design approach
 The size is increased
 Why ?
3. Conclusions
Presented in 20th IAEA
Fusion Energy Conference
1 - 6 November 2004
Vilamoura, Portugal
Reactor design collaborations in NIFS
Helical core plasma
K.Yamazaki
(NIFS )
Ignition access
& heat flux
O.Mitarai
(Kyusyu Tokai
Univ.)
Thermo-mechanical analysis
H.Matsui
(Tohoku Univ.)
Structural Materials
Safety
& Cost
Blanket
Core
Plasma
Pr otectio n wall
2
Ph =0.2 MW/m
2
Pn =1.5 MW/m
Nd =450 dp a/30y
Radiation s hield
r edu ction
> 5 or ders
Th er mal shield
Self- coo le d
T br eeder
TB Rloca l> 1.2
Be Liqui d
/ Gas
14MeV neutron
shi el ding
mat eri als.
Vacuu mv es sel
&
T bo undary
SC
magnet
Advanced
first wall
T.Norimatsu
(Osaka Univ.)
Chemistry
Device system code
H.Hashizume
(Tohoku Univ.)
HX
Tritium
T storage
Purifier
Tank
Advanced
thermofluid
T.Kunugi
(Kyoto Univ.)
Pump
Pump
Thermofluid
FFHR
Helical reactor design
/ Sytem Integration
A.Sagara
(NIFS)
T-dise ngager
In- vessel
C omponents
1993〜
20°C
high temp.
surface heat fl ux
Blanket system
S,Tanaka
(Univ. of Tokyo )
Thermofluid
Thermofluid
MHD
system
S.Satake
(Tokyo Univ. Sci.) K.Yuki
(Tohoku Univ.)
Turbine
T-disengager system
S.Fukada, M.Nishikawa
(Kyusyu Univ.)
Heat exchanger & gas turbine system
A.Shimizu
(Kyusyu Univ.)
June 6. 2003, A.Sagara
Tritium recovery systems
Kyushyu Univ.: S.Fukada
Permeation leak through the recovery system
is a crucial problem
 Small amount of Flibe or
He gas flow in the
double tube are good as
permeation barrier to
reduce < 10Ci/day.
 The most serious
problem is permeation
leak of ~34 kCi/day
through the heat
exchanger to the He
loop
Energy conversion systems
for Flibe in/out temperature of 450˚C and 550˚C
Kyushyu Univ.: A.Shimizu
Three-stage compression-expansion
He-GT system was newly proposed
 However, max decreases
rapidly with the increase
of pressure drop.
 Therefore the layout of
energy conversion
system is a key design
issue.
I. C.・
I. C.・
RH. ・
H.
RH. ・
～
C.・
C.・
E.・
C.・
E.・
E.・
Rec.
0.4
Pr.C.
Thermal efficiency
 max ~ 37% for
compression ratio of 1.5,
th
0.3
0.2
0.1
0
0
0.1
0.2
P
 Po
o
0.3
0.4
0.5
Relative pressure drop
innovative free surface wall* design
Kyoto Univ.: T.Kunugi
*KSF wall (Kunugi-Sagara type Free surface wall)
1 m/sec
Liq.Temp.
°C
g
1. 2
2. 4 mm
(48cell)
1. 2
0. 2
Solution domain
14 cell
0.1 MW/m2
t = 3.016
5000
4000
・
2
K]
6000
Steady State
540
3000
Twall
2000
520
Tbulk
1000
0
560
3.01
3.02
Time [sec]
500
3.03
Temperature [C]
 which enhances heat
transfer efficiency
about one order.
1 m/sec
20 mm
(98cell)
[W/m
 Numerical simulation
has explored the
formation of a pair of
symmetrical spiral
flow,
g
h local
 Micro grooves are
made on the first wall
to use capillary force
to withstand the
gravity force
TNT loop ~ 0.1m3, < 600°C
Thermofluid R&D activities
Tohoku Univ.: H.Hashizume
for enhancing heat-transfer in such
high Prandtl-number fluid as Flibe
 “TNT loop” (Tohoku-NIFS
Thermofluid loop) has been
operated using HTS
(Heat Transfer Salt, Tm= 142°C)
 Results are converted into Flibe
case at the same Pr=28.5
(Tin=200oC for HTS and 536oC for Flibe)
 Same performance as turbulent
flow is obtained at one order
lower flow rate.
 This is a big advantage for MHD
effects and the pumping power.
QuickTimeý Dz
TIFF (LZW) êLí£É vÉçÉOÉ âÉÄ
ǙDZÇÃÉsÉN É`É ÉǾå©Ç ÈÇ…ÇÕïKóvÇ­Ç•
ÅB
Bird’s eye view of TNT loop
Control
Room
Upper Tank
Air Cooler
Test
Section
Main Pump
Dump Tank
Self-cooled Be-free Li/V blanket
NIFS : T.Tanaka, T.Muroga
based on R&D progress on
in-situ MHD coatings and
high purity V fabrication
 Simple models are evaluated as
alternatives for FFHR2 blanket..
 Balance of TBR and the
shielding performance is
examined, because shielding is
poor w/o Be.
 TBR of Li/V is higher than 1.3 at
about 50 cm with an acceptable
shielding efficiency for superconducting magnets.
B reeding coolant (~83 vol. % ):
Liquid lithium or Flibe
Structural m aterial (~17 vol. %):
V-4C r-4Ti or JLF-1
W arm or First w all
(5m m)
(5m m)
JLF-1
(5cm)
Plasm a
Supercondu cting
m agnet
JLF-1 (70 vol. % )
+ B 4C (30 vol. %)
Self-cooled
tritium breeder
channels
x cm
0
Radiation shield
(120 - x) cm
120 cm
Modeling to Evaluate MHD pressure drop
is established for self-cooled lithium blanket.
Tohoku Univ. : H.Hashizume
10m
5 mm
10-25 m
100 mm
Insulator
 Three-layered wall is proposed,
where the inner thin metal layer
protects permeation of lithium
into the crack of coated layer
Channel wall (V)
Inner layer(V)
B field
 Extremely good agreement
between FEM and theory has
been obtained
1.00E+06
10 μm ( numerical)
dp /d z(Pa/m)
 The performance required to the
insulator is evaluated to be
1.00E+07
15 μm(n ume ric al)
20 μm（nu me rical)
1.00E+05
25 μm(n ume ric al)
10 μm(an alytical)
25 μm(an alytical)
1.00E+04
 insulator
108 109
V
1.00E+03
1.00E-12 1.00E-10 1.00E-08 1.00E-06 1.00E-04 1.00E-02 1.00E+00
σ(insulator) /σ(V)
Present R & D activities on Flibe blanket in Japan
Presented by A.Sagara(NIFS) , Feb.’04
FY1993
Helical reactor
FFHR design
with R&D
1997
2001
2004
2007
$
R&D/LHD
TNT loop
Ultrahigh HT
$
JUPITER-II
$
ITER-TBM
Frame?
Resouce?
2015
Flibe/RAF blanket, Li/V blanket, SB-He/SiC Blanket
R&D in Japan-US joint project JUPITER-II(FY’01~’06)
INEEL
1-1-A: FLiBe Handling/Tritium.Chemistry
1-1-B: FLiBe Thermofluid Flow Simulation
2-2 : SiC System Thermomechanics
Japan
3-1: Design-based Integration Modeling
3-2: Materials Systems Modeling
UCLA
ORNL
1-2-B: V Alloy Capsule Irradiation
2-1 : SiC Fundamental Issues, Fabrication,
and Materials Supply
2-3 : SiC Capsule Irradiation
ANL
(2001)
1-2-A: Coatings for MHD Reduction
http://jupiter2.iae.kyoto-u.ac.jp/index-j.html
LHD-type D-T Reactor FFHR
m a c 
Selection of lower   
l R 
To reduce mag. foop force
To expand blanket space
Many advantages :
 current-less
 Steady state
 no current drive power
 Intrinsic divertor
LHD operation: 1998 ~
=tan
1993 FFHR-1 (l=3, m=18 )
R=20, Bt=12T, b=0.7%
1995 FFHR-2 (l=2, m=10 )
R=10, Bt=10T, b=1.8%
2
W
0.3
< f a > / (B0 IH )
Z (m)
coil cross-section


1
NIFS-960812 / S.I.
0.4
Rax =3.6m, B q=100%, 
LHD
0
LHD
H
0.2
ac
ac /H=4
FFHR-2
2
0.1
4
2
0
FFHR-1
=2
-0.1
=3
-1
-0.2
@W/H=2
a c /H W /H < f a > (MN/m)
LHD
3.4
FFHR-1 2.3
FFHR-2 3.0
1.74 10.7
2.0
96.7
2.0 124.5
-0.3
0.5
-2
2
3
4
R (m)
5
6
0.6 0.7 0.8 0.9 1
1.1 1.2
Coil pitch parameter  c =(m/ )( ac /R)
1.3
However, direction of compact
design has engineering issues
 Insufficient tritium breeding
ratio (TBR)
 Insufficient nuclear shielding
for superconducting (SC)
magnets,
 Replacement of blanket due to
high neutron wall loading
 Narrowed maintenance ports
due to the support structure for
high magnetic field
Blanket space
limitation
Replacement
difficulty
New design approach is proposed
to overcome all these issues
 Introducing a long–life &
thicker breeder blanket
 Increasing the reactor size
with decreasing the magnetic
field
 Improving the coils-support
structure
Blanket
space
Replacement
Proposal and Optimization of STB
(Spectral-shifter and Tritium breeder Blanket)
1400
°C
 Lifetime of Flibe/RAFS liquid
blanket in FFHR ~ 15MWa/m2
Fast
neutron
First wall
Breeder
this work
1000
Temperature
factor
 Neutron wall loading
3
in FFHR2 in 30 years
1.5MW/m2 x 30y = 45MWa/m2
Carbon
1200
Thermal creep (1% creep strain
under applied stress sy/3 )
800
Void swelling (>1%)
600
He effect (?)
operation
400
200
0
0
He effect (?)
DBTT increase by irradiation
2
4
6
8
10 12 14 16 18 20
Neutron wall loading MWa/m
2
Radiation shield
Vacuum vessel
&
&
Thermal shield
T boundary
Self-cooled
T breeder
First wall
Tiles
mechanical
joint
wall
Carbon Be First
coolant out
Breeder
Plasma
C
Flibe
6Li
C
Fast Be
Be C
neutron4
LCFS
16
10
0
1
17
SC
magnet
JLF-1
+ B4 C
(30 vol.%)
40%
2
1
ISSEC : by Kulchinski, ‘75
Reduced Activation Ferritic Steel
TBR
0.3 7.7 2
53
JLF-1
Thermal
insulation
5 10
JLF-1
(60 vol.%)
STB
JLF-1 &Beoptimization
(Redox)
107 cm
STB
Tritium breeding
18
1.1
0
20
40
60
80
100
120
Thickness of Be C (mm)
Local TBR
1.8
1.6
STB
1.4
1.2
1
0.8
0
5
10
15
20
Thickness of the first wall JLF-1 (mm)
Neutron flux (n/m2 s/lethargy)
2
design point
6
1.3
at the SC magnet 5
1.2
4
1.1
1
at the first wall
(4.2 w/o STB)
Be C=70mm
3
2
2
0.9
70
1
80
90
100 110 120 130
Thickness of 2nd carbon (mm)
/m2s)
1.0
C
120 mm
1.4
7
13
C Be C
2
10
First wall
10 mm
breeder
& shield
Li-6 : 40%
Fast Neutron Flux ( x 10
Local TBR
100 mm
STB
2
2nd C =80 mm
design point
1.5
Fast Neutron Flux ( x 10
1.2
/m s)
Shielding efficiency
1020
1019
Original FFHR2 design
FFHR2 with STB
1018
1017
1016 -7 -6 -5 -4 -3 -2 -1 0 1 2
10 10 10 10 10 10 10 10 10 10
Neutron energy (MeV)
Results and Key R&D Issues
Results :
 Fast neutron flux at the first
wall is factor 3 reduced.
 Local TBR > 1.2 is possible.
 Fast neutron fluence to SC is
reduced to 5x1022n/m2
(Tc/Tco>90% in 30y)
 Surface temperature <
2000°C(~mPa of C) is
Keypossible
R&D issues :
(1) Impurity shielding in edge
plasma
(2) Neutron irradiation effects
on tiles at high temperature
(3) Heat transfer enhancement
in Flibe flow
Radiation shield
Vacuum vessel
&
&
Thermal shield
T boundary
Self-cooled
T breeder
First wall
Tiles
mechanical
joint
coolant out
Plasma
C
Flibe
6Li
C
Be
Be 2 C
40%
4
LCFS
1
16
10
0
JLF-1
1
17
SC
magnet
JLF-1
+ B4 C
(30 vol.%)
0.3 7.7 2
Be (60 vol.%)
(Redox)
Thermal
insulation
53
JLF-1
STB
5 10
JLF-1
107 cm
Required conditions :
 Neutron wall loading Design
parameter
2
< 1.5MW/m
 Blanket thickness
> 1100 mm
  for C-Be-C tiles
> 100W/mK
 Super-G sheet for tiles joint
> 6kW/m2K
 Heat removal by Flibe
> 1MW/m2
Improved design parameters
Design parameters
Improved
input
Required
Blanket
space
Required
Wall loading
ISS95  0.26P
0.59
ne
0.51
B
0.83
R
a 2/ 3
0.65 2.21
0.4
LHD
Polarity
l
2
Field periods
m
10
Coil pitch parameter

1.25
Coil major Radius
Rc m
3.9
Coil minor radius
ac m
0.98
Plasma major radius
Rp m
3.75
Plasma radius
ap m
0.61
Blanket space
 m
0.12
Magnetic field
B0 T
4
Max. field on coils
Bmax T
9.2
Coil current density
j MA/m2
53
Weight of support
ton
400
Magnetic energy
GJ
1.64
Fusion power
PF GW
Neutron wall load
MW/m2
External heating power
Pext MW
 heating efficiency

Density lim.improvement
H factor of ISS95
Effective ion charge
Zeff
Electron density
ne(0) 10^19 m-3
Temperature
Ti(0) keV
<b>=2*n*T/(B^2/ ) (parabolic distribution)
COE
Yen/kWh
FFHR2 FFHR2m1 FFHR2m2
2
10
1.15
10
2.3
10
1.2
0.7
10
13
25
2880
147
1.77
2.8
100
0.7
1
2.53
1.32
28.0
27
1.8
21.00
2
10
1.15
14.0
3.22
14.0
1.73
1.2
6.18
13.3
26.6
3020
120
1.9
1.5
80
0.9
1.5
1.92
1.34
26.7
15.8
3.0
14.00
2
10
1.25
17.3
4.33
16.0
2.80
1.1
4.43
13
32.8
3210
142
3.0
1.3
100
0.9
1.5
1.68
1.35
19.0
16.1
4.1
9.00
Self-ignition access in FFHR2m1
50MW
ext
10MW
e
ex t
=0)
10
,
15
20
25
30
Temperature (keV)
Zero-dimensional analysis
 H ISS95 = 1.2 x 1.6 Already
achieved in
 * / E = 3 ( < 7)
LHD
 parabolic profiles
 ne< 1.5 x Sudo limit
  heating efficiency = 0.9
O.Mitarai et al., Fusion Eng. Design 70 (‘04) 247.
3
ne (1020 /m3)
5
T (10keV)
P (GW)
e
F
4
2
3
2
1
S
1
DT
(1019 /m3s)
0
0
50
100
150
Time (s)
0
200
F
5
b (%)
7
P, b
0
4
6
f
Operation path
0
(10MW)
e
n limit (P
EXT
T,
1
8
DT
P =
P
S
20MW
FFHR2m1
9
ext
20
30MW
5
e
b=
5%
3GW
4%
2GW
3%
1GW
2%
P =0.5GW
3
P
-
40MW
2
10
FFHR2m1
n,
3
Density (x 10 m )
4
Self-ignition access in FFHR2m2
11
5
FFHR2m2
-
FFHR2m2
10
30MW
20MW
ex t
10MW Operation path
5
10
Pf =0.5GW
15
20
25
P (GW)
7
F
30
Temperature (keV)
Zero-dimensional analysis
 H ISS95 = 1.1 x 1.6 Already
achieved in
  * /  E = 3 ( < 7 )
LHD
 parabolic profiles
 ne< 1.5 x Sudo limit
  heating efficiency = 0.9
O.Mitarai et al., Fusion Eng. Design 70 (‘04) 247.
3
6
ne (1020 /m3)
5
4
3
F
0
b=5 %
4%
3%
2
T (10keV)
e
2
S
1
DT
(1019 /m3s)
0
0
50
100
150
Time (s)
P, b
0
4
b (%)
e
P =
(10MW)
T,
1
EXT
8
DT
3GW
P
9
e
50MW
40MW
=0)
,
2
ext
S
e
ext
n limit (P
P
20
3
n,
3
Density (x 10 m )
4
1
0
200
Improved Design of
Coil-supporting Structure (1/2)
 Cylindrical supporting structure
under reduced magnetic force,
which facilitates expansion of the
maintenance ports.
 Helical coils supported at inner,
outer and bottom only.
17.4
RIV =9.5m, ZOV =3.6 m
17.3
120
design
point
119
118
3.2
Radius of OV Coil (m)
 W/H = 2 & H/ac determined by
 Bmax ~ 13 T for such as Nb3Sn
or Nb3Al.
 Then J=25~35A/mm2
 Poloidal coils layout as for
stored magnetic energy and
stray field
Stored Energy (GJ)
121
3.3
3.4
3.5
3.6
Z of IV Coil (m)
3.7
17.2
17.1
3.8
Improved Design of
Coil-supporting Structure (2/2)
Coil current (MA)
HC 43.257
OV -21.725
IV -22.100
 The maximum stress can be
reduced less than 1000 MPa
(<1.5Sm)
FFHR2m1
 This value is allowable for
strengthened stainless steel.
outer
top
100
inner
bottom
outer
50
0
-50
Fa (minor-radius)
Fb (overturning)
0
12
24
36
48
60
Toroidal Angle, Phi (deg)
72
Electromagnetic Force (MN/m)
Electromagnetic Force (MN/m)
Electromagnetic forces on
a helical coil of FFHR2m1.
80
60
40
20
FZ of OV
FR of OV
0
FZ of IV
FR of IV
-20
-40
0
12
24
36
48
60
Toroidal Angle, Phi (deg)
72
on poloidal coils
Replacement of In-Vessel
Components (1/2)
 Large size maintenance ports at
top, bottom, outer and inner sides.
 The vacuum boundary located
just inside of the helical coils and
supporting structure.
 Blanket units supported on the
permanent shielding structures,
which are mainly supported at
their helical bottom position.
Replacement of In-Vessel
Components (2/2)
Proposal of the “Screw coaster” concept
to replace STB armor tiles
 Replacement of bolted tiles
during the planned inspection
period.
 Using the merit of helical
structure, where the normal
cross section of blanket is
constant.
 Toroidal effects can be
adjusted with flexible
actuators.
Cost Estimation
Using PEC code developed in NIFS.
Calibrated with ARIES-AT, ARIES-SPPS, resulting in good agreement within 5 %.
(T.J.Dolan,K.Yamazaki, A.Sagara, in press in Fusion Science & Tech.)
 COE’s for FFHR2, FFHR2m1 and
FFHR2m2 decreases with increasing
the reactor size, because the fusion output
R 2,
increases in ~
while the weight of coil
supporting structure increases in~ R 0.4 ( not ~R3).
COE, Yen/kWh
11
favail = 0.85 - favail *1.4
10
 When the blanket lifetime ~ 30y, the
COE decreases ~20% due to higher
8
7
6
0
5
10
Hiss=2
2 GW
25
Hiss=1.5
FFHR2
20
1
beta=1%
15
2%
10
4 GW
5
3%
Pf = 1 GW
30
HISS=1.5
2 GW
25
4 GW
FFHR2m1
7 GW
0
8
10 12 14 16 18
Rp, m
beta=1%
2%
10
4%
5
6%
4%
7 GW
HISS=2
=1.15
Rp/<a>=8.09
=1.2 m
20
15
0
25
10 12 14 16 18 20 22 24
HISS=1.5
30
35
40
=1.25
=1.1 m
Rp/<a>=5.71
35
Pf=1 GW
30
25
2 GW
20
beta=1%
4 GW
15
beta=2%
7 GW
10
beta=4%
5
beta=7%
FFHR2m2
0
Major radius Rp, m
20
40
COE, Yen/kWh
COE, Yen/kWh
30
15
Blanket Lifetime, years
HISS=2
35
 = 1.15
Rp/<a>=8.33
 = 0.7 m
COE, Yen/kWh
Pf=1 GW
*tm / Wtb
9
availability and lower cost for replacement.
35
w
14
16
18
20
Major radius Rp, m
22
24
Conclusions
Design studies on FFHR have focused on new design
approaches to solve the key engineering issues of
blanket space limitation and replacement difficulty.
(1) The combination of improved support structure
and long–life breeder blanket STB is quite
successful.
(2) The “screw coaster” concept is advantageous in
heliotron reactors to replace in-vessel
components.
(3) The COE can be largely reduced by those
improved designs.
(4) The key R&D issues to develop the STB concept
are elucidated.