NET

Fusion Engineering and Design 87 (1991) 87-94 North-Holland

87

The Karlsruhe solid breeder blanket and the test module to be irradiated in I T E R / N E T M. Dalle Donne

1, E. B o j a r s k y , U . F i s c h e r , M . K i i c h l e , P. N o r a j i t r a , G . R e i m a n n ,

H. Reiser

Kernforschungszentrum Karlsruhe, Postfaeh 3640, W-7500 Karlsruhe, Germany H . D . B a s c h e k a n d E. B o g u s c h

lnteratom, Postfach, W-5060 Bergisch-Gladbach 1, Germany

The blanket for the DEMO reactor should operate at an average neutron flux of 2.2 M W / m 2 for 20000 h. This requires the use of a structural material which can withstand high neutron fluences without swelling. The ferritic steel Manet was chosen for this purpose. The breeder material is in the form of Li4SiO 4 pebbles of 0.35 to 0.6 mm diameter. The 6 mm thick beds of pebbles are placed between beryllium plates which are cooled by high pressure helium flowing inside steel tubes. Breeder material and beryllium are contained in radial canisters, placed inside boxes. The coolant helium enters the blanket at 250 ° C, cools first the box walls and then the breeder and multiplier, and leaves the blanket at 450 ° C. The maximum temperature in the first wall steel is 550 ° C, while the minimum and maximum temperatures in the breeder are 380 and 820 ° C, respectively. The resulting total tritium inventory in the breeder is only 10 g, and the real tridimensional tritium breeding ratio is 1.11. The conceptual design of the test module, of its extraction system and of the required out-of-reactor ancillary systems has allowed an estimate of the time constants of the various components and thus allowed an assessment of the requirements given by the testing of the modules on the N E T / I T E R machine.

I. Introduction In the f r a m e w o r k of t h e E u r o p e a n Fusion T e c h n o l ogy P r o g r a m ( E F T P ) t h e K a r l s r u h e Nuclear R e s e a r c h C e n t e r (KfK) is p e r f o r m i n g design a n d r e l a t e d R a n d D work for the B r e e d e r O u t - o f - T u b e ( B O T ) H e l i u m Cooled Solid B r e e d e r B l a n k e t ( H C S B B ) for the Demonstration Reactor (DEMO). D E M O is the next step after N E T / I T E R . Its blanket should d e m o n s t r a t e t h e capability of p r o d u c i n g tritium in the b l a n k e t with a b r e e d i n g ratio h i g h e r t h a n one. F u r t h e r m o r e , the b l a n k e t coolant t e m p e r a t u r e should b e sufficiently high so that a p l a n t efficiency of at least 30% for the h e a t extracted from the b l a n k e t should b e possible [1]. Beside the B O T - H C S B B , t h r e e o t h e r D E M O relevant b l a n k e t types are b e i n g investig a t e d within t h e E F T P : o n e is the B r e e d e r Inside T u b e ( B I T ) - H C S B B , the o t h e r two are with a liquid metal b r e e d e r (liquid m e t a l or w a t e r cooled, respectively). C o m m o n specifications for the D E M O reactor and the

Delegated from Euratom to Kernforschungszentrum Karlsruhe. 0920-3796/91/$03.50

D E M O r e l e v a n t b l a n k e t have b e e n p r o p o s e d by the E u r o p e a n Test B l a n k e t Advisory G r o u p [1]. T a b l e 1 shows the most i m p o r t a n t of these specifications. T h e main objective of t h e E u r o p e a n R a n d D work for the D E M O relevant b l a n k e t s is to choose two b l a n k e t types (one with solid a n d the o t h e r with a liquid

Table 1 DEMO specifications Geometry:

Double null Number of sectors: 16 Number of segments: 48 outboard, 32 inboard Plasma major radius: 6.3 m Blanket thickness: 0.85 m outboard, 0.55 m inboard

Operation:

Neutron wall load: 2.2 M W / m 2 average Surface heat flux: 0.4 M W / m 2 average 0.5 M W / m 2 peak Operation time of segment: 20000 h Tritium Breeding Ratio (TBR) > 1.0 Tritium residence time in blanket < 10 d Structural material: Manet

© 1991 - E l s e v i e r S c i e n c e P u b l i s h e r s B . V . A l l r i g h t s r e s e r v e d

88

M. Dalle Donne et al. / The Karlsruhe solid breeder blanket

breeder) and to prepare the related test modules to bc irradiated in N E T / I T E R . One of the main items of this work is the design of the D E M O relevant blanket and of the test modules. In the present paper only a short description of this design work for the BOTHCSBB will be given, more detailed information will be published later [2].

2. The BOT DEMO blanket The present design is a development of a previous design which was made for the NET reactor [3]. The design is based on the following principles: (a) The use of lithium orthosilicate (Li4SiO 4) as breeder to have fast tritium release (low tritium inventory) and low lithium partial pressure at high temperatures (low lithium transport). (b) The use of small Li4SiO 4 pebbles to avoid thermal stress problems and to provide a well defined path for the helium purge flow. (c) The use of a helium purge flow at a pressure below atmospheric to reduce the amount and probability of tritium losses and to reduce the mass flow rate of the purge gas system. (d) The use of the Breeder Out-of-Tube solution to keep the thickness of the coolant pressure tubes within reasonable limits. (e) Cool the first wall with cold (inlet) helium; keep the minimum temperature of the breeder above 380 ° C to reduce the tritium inventory; keep the beryllium temperature as low as possible (reduce beryllium swelling) and at the same time keep beryllium and breeder well mixed (increase tritium breeding ratio). (f) Use of radial canisters, which allow a good filling with breeder and multiplier of the space available in the blanket region, and at the same time allow a subdivision of the blanket in relatively small submodales; this reduces the thermal stresses and the stresses caused by plasma disruptions, makes precise construction easier and gives the possibility of making significant tests starting from the smallest submodules. (g) Use of a redundant convective cooling system and of a double containment against tritium losses for safety improvement.

2.1. The blanket configuration for the DEMO reactor Figure 1 shows a vertical cross section of the right side of the torus with the proposed blanket. As in the case of the outboard blanket design for NET, the

canisters are contained in a segment box, however, due to the greater space available in the radial direction, the blankets are thicker and a canister solution is now available for the inboard blanket as well. Blanket canisters are placed also behind the divertors, and the coolant tubes for thc lower divertor and canisters are coming from below. Due to the small place available at the neck in the upper region of the inboard boxes, thc diameter of the coolant helium feeding tubes is relatively small (126 mm o.d.), thus the helium pressure has to be higher than in the outboard blanket ( 10 MPa against 8 MPa) to keep the pressurc drops within reasonable limits. The vacuum vessel and the magnet are protected against an excessive neutron flux by helium cooled steel shields. At the inboard side thc shields should contain about 10% zirconium hydride to reduce the neutron fluence in the vacuum vessel below the allowable limit. Figure 2 shows a radial toroidal cross section of a segment of the outboard blanket. It looks similar to the NET design, however, there are some significant differences mainly due to the choice of the structural material. Due to the higher neutron fluence in thc D E M O reactor, the martensitic steel Manet has been chosen rather than austenitic steel like for the NET blanket. This dictates the choice of the coolant helium temperature of 250 ° C at the blanket inlet to keep the Manet at temperatures above the DBTT (Ductile-Brittle Transition Temperature) level. The reduced difference between outlet and inlet helium temperature, and the better thermal conductivity of Manet allow to have the inlet and outlet helium feeding tubes welded together (fig. 2). Thus they make the back wall of the box, allowing a considerable simplification at the back of the box with respect to the NET solution. Figure 3 shows fin isometric view of the outboard segment. Figure 4 shows a radial-toroidal cross section of a segment of the inboard blanket. The canister is similar to the outboard canister, however, the helium coolant coils lie in radial-toroidal planes rather than in radialpoloidal ones like in the outboard blanket. The power distribution, the local tritium production and the tritium breeding ratio have been evaluated by means of three-dimensional Monte Carlo calculations using the code MCNP. These calculations account also for the presence of ten ports placed in the equatorial plane of ten outboard blanket segments [4]. The blanket temperatures and stresses in the structural material (first wall and boxes) have been evaluated by tridimensional calculations using the computer code ABAQUS. The stresses are within acceptable values. Table 2 shows thc main results of these calculations.

89

M. Dalle Donne et al. / The Karlsruhe solid breeder blanket

Purge flow

Shield coolant helium

" iI

I

I'~MIL._d

Upper divertor \

Torus axis

~1, m~ l

i-~--.- Blanket

coo,ant

; i~1',~_

helium

PF coils

V

Outboard ,blanket

i Inboard i blanket

/

h

/

i

\

\ \

i

i

i

t___~

'\ i Shield~ I I I I I I

i

i

/////

\/ v

/\

',/' 7 Breeder canisters

V Vacuum vessel TF coils ~Lower// divertor

Fig. 1. Vertical cross section of the BOT Helium Cooled Solid Breeder Blanket for the DEMO Reactor.

90

M. Dalle Donne et al. / The Karlsruhe solid breeder blanket

He CoolingSystem1 EB Welded He Cooling System2

600 °C [7]. These data might not be completely relevant to the present blanket as the beryllium of ref. [7] was irradiated at low temperatures ( < 7 5 ° C ) , was probably considerably anisotropic and contained a relatively large amount of BeO. The tritium purge system data appear quite fcasible. The tritium losses by leakage are neglibible. The tritium losses by permeation from the purge flow system to the main helium system are considerably higher than in the case of NET. This is due to the fact that the permeability of tritium through Manet is considerably higher than through austenitic steels [6].

Coolant /

Purge Gas Supply

18,5

'Segment Box with First W a l l - - ~

1975 ~' 13,l

Purge Gas

~d24L..-

Fig. 2. Radial-toroidal cross section of a segment of the outboard blanket (dimensions in mm).

Outle Helium Coolant Inlet2.2. Tritium inventories and control

The evaluation of the tritium inventories and the tritium control in the blanket are particularly important as the DEMO blanket has to show the capability of producing sufficient tritium for a continuous plasma operation (TBR > 1). Similarly to the blanket design for NET, the tritium extraction and control are based on a tritium purge flow system using helium plus 0.1% hydrogen at subatmospheric pressure to extract the major fraction of the tritium produced in the blanket. Furthermore, 0.1% of the helium mass flow is continuously extracted from the main helium coolant circuit and sent to a helium purification plant for the extraction of the impurities and of the tritium resulting from permeation from the purge flow or directly injected from the plasma. The assessment of these tritium quantities has been performed with the methods illustrated in refs. [3] and [5]. The permeation data for Manet are based on the experimental results of ref. [6]. Table 3 shows the main results of these calculations. The highest tritium inventory is in beryllium. This has been calculated on the assumption that all the tritium produced in the beryllium is trapped in it. Recent experimental information shows that beryllium irradiated to a total fluence of 5 × 1022 cm 2 (E >/ 1 MeV) releases 99% of tritium only at temperatures above

Breeder Coolant

Tubes

Pebble Bed Breeder Canister

Copper Saddle

First Wall-

Support Structure and Shield--

Fig. 3. Isometric view of the outboard segment.

M. Dalle Donne et aL / The Karlsruhe solid breeder blanket

91

Table 2 Main characteristics of BOT Helium Cooled Solid Breeder Blanket for the DEMO reactor 0.35-0.6 mm Li4SiO 4 pebbles (90% 6Li enrichment), total amount = 60 ton Beryllium, total amount = 228 ton 2500 MW ( + 300 MW in the divertors) inlet = 250 ° C outlet = 450 ° C 8 MPa outboard, 10 MPa inboard

Breeding material: Multiplier: Total blanket power: Coolant helium temperature: Coolant helium pressure: Coolant helium pressure drop (first wall, blanket, feeding tubes): First wall maximum steel temperature: Max. temp. in beryllium: Max. temp. in pebble bed: Min. temp. in pebble bed: Peak thermal and pressure load:

0.26 MPa outboard, 0.4 MPa inboard 550°C 600 ° C 820 ° C 380°C 437 MPa (Von Mises, primary plus secondary stress, at first wall, outboard equatorial plane, T = 481 ° C)

Real tridimensional tritium breeding ratio (assuming ten 3 x 1 m ports on outboard blanket for heating systems and others): Tritium production rate:

1.11 390 g / d

T h e tritium inventory in t h e first wall a n d the tritium direct losses f r o m t h e p l a s m a to t h e m a i n h e l i u m cooling system are k n o w n only very roughly. They have b e e n calculated with the code D I F F U S E for a similar b l a n k e t [8]. T h e a m o u n t s of tritium coming directly from the p l a s m a by p e r m e a t i o n are very large (estim a t e d at 1 - 1 0 0 g / d ) . Only with a n effective oxidizing Helium Coolant Channel

a t m o s p h e r e in the h e l i u m m a i n cooling system a n d / o r p e r m e a t i o n r e d u c i n g coatings o n t h e p e r m e a t i n g surfaces could it b e possible to r e d u c e the tritium losses to

Table 3 Tritium inventories and control Tritium inventories: Tritium inventory in Li4SiO 4 pebbles: 10 g - Tritium in first wall: 3 to 300 g - Tritium in beryllium at the end of blanket life: 2080 g Tritium in solution in blanket structural material: 0.15 g

Shield

-

-

Tritium purge system: - Total purge helium mass flow: 0.67 kg/s Average helium pressure: 0.08 MPa - Purge helium velocity in the bed: 0.3 m / s - Pressure drops in the bed: 0.011 MPa outboard, 0.006 MPa inboard HT partial pressure in purge helium: 0.85 Pa outboard, 0.44 Pa inboard - H / H T ratio: 94 outboard, 182 inboard -

o0

-

Beryllium

First W a l l

Coolant Tubes 14/12mm

Purge Gas Inlet

Fig. 4. Radial-toroidal cross section of the in board blanket segment and relative shield (dimensions in mm).

Tritium losses (neglecting the direct losses from plasma to main helium cooling system) by permeation from purge system to main helium coolant system: 2.9 g / d by permeation from the main helium coolant system to water/steam circuit: reducing atmosphere in He-system: 175 C i / d (6.5 T B q / d ) oxidizing atmosphere in He-system: < 10 C i / d (0.37 T B q / d )

92

M. Dalle Donne et al. / The Karlsruhe solid breeder blanket

the environment to the desired value of less than L0 C i / d (0.37 T B q / d ) .

behind the standard water cooled N E T / I T E R first wall. Then, a 316 L module should be placed with its helium cooled wall directly facing the plasma. Finally, a Manet module facing the plasma should bc tested. Figure 5 shows a vertical and horizontal cross section of the test module placed behind the first wall. Thc module composed of a box containing six canisters, the neutron shields and the various pipes, is contained in an adapter box. The adapter box is attached to the vacuum vessel by electrically insulated bolts. The test module and the shield are mounted on rails (also electrically insulated from the box) and can be withdrawn and brought to a manipulator box by opening two horizontally sliding doors. The module inside the manipulator box can be transported to the hot cells. The conceptual design of the test module, of its

3. The test module

O n e of the main objectives of N E T / I T E R is to test D E M O relevant blankets. Presently the N E T / I T E R testing program foresees first module or submodule tests in ports placed in the equatorial plane of outboard blanket segments and later tests of whole segments or sectors. The testing philosophy for the D E M O relevant B O T - H C S B B foresees that the module testing should be made in three stages in order to reduce the risk of malfunction. First the module should be made of austenitic steel 316 L, rather than Manet, and placed

ELECTRICII'ISULATION

REETANGLLAR

i

W

lUrER P~RTI "

N~III~ -()x ~"'i'mmL._L_

,.a

,

-$

-a W,:~--

,

__

] .

.

.

,,~z,,.-~::/< .

-e

RRST WALL

~ l \ l ~ l R l ~ i

c*a.~su,Pumut"~,~ /'

Iv~,~m.E

FIRSTWALL TESTBLANKET

• tt~ 0

1

i

Y ,< /

/ /'

q~lJE~

/

tmlu~a~t

b

t

r

I ...............................

i t Fig. 5. Vertical (upper picture) and horizontal (lower picture) cross section of the test module placed behind the first wall m a horizontal port of the outboard region. The figure shows also the shield. Test module and shield are shown also inside the manipulator box (dotted lines) in the position used for transportation to the hot cells (dimensions in mm).

93

M. Dalle Donne et al. / The Karlsruhe solid breeder blanket

extraction system and of the required out-of-reactor ancillary systems (two helium cooling systems, one purification system for the cooling helium and one purge gas system) has allowed to estimate the time constants of the various components and thus allowed to assess fhe requirements given by the testing of the modules on the N E T / I T E R machine (burn time, duty cycle and duration of continuous operation). Table 4 shows the minimum requirements posed on the N E T / I T E R operation by the testing of the BOTHCSBB module. Due to the rapid release of tritium from the LiaSiO 4 and the relatively high temperature in the pebble bed, a burn shot of 1000 s is sufficient to achieve 50% of the equilibrium tritium release. The time to reach equilibrium of tritium permeation to the main helium coolant system and steady-state conditions in the tritum extraction system is considerably long~-r ( ~ 1 d),_ however quick preliminary tests (shakedown tests) can be performed during the early technology phase of the N E T / I T E R machine in 10 to 20 h. At the end of the technology phase longer tests are required, especially for the validation of the computer codes previously developed on the base of outof-pile tests, tests in fission reactors and tests with 14.1 MeV neutron sources. These tests pose most severe requirements on continuous operation time ( = 5 d) and duty cycle (>/50%) for the N E T / I T E R machine.

4.

C o n c l u s i o n s

a n d

future

w o r k

The design and R and D work performed for the Karlsruhe Helium Cooled Solid Breeder Blanket for the DEMO Reactor have shown so far no feasibility problems. There are, however, still a number of major technical issues to be solved, which are listed below: (a) LiaSiO4 pebbles: behaviour at high neutron fluences (10 at% total lithium burnup) in the temperature range 380-800 (900) ° C, namely: tritium transport, lithium transport, mechanical stability, - thermal conductivity of the pebble bed, - compatibility with beryllium. (b) Beryllium: behaviour at high neutron fluences (2.2× 10 22 n / c m e, E > 1 MeV) and temperatures (250-600 ° C), namely: - swelling, embrittlement, compatibility with Manet, - behaviour of the interface between beryllium and cooling tubes under thermal cycling, tritium retention. These properties should be optimized by changing the beryllium structure and its oxygen content. (c) Segment box and canisters: general behaviour of

Table 4 Minimum requirements posed on NET/ITER operation by the testing of the BOT-HCSBB module Time to reach: [100 s in blanket front - steady state temp. in the breeder a: I 300 s in blanket back 50% with 10005 burn time shot length tritium inventory (tritum release) fraction in the breeder: 67% with 30005 burn time shot length equilibrium tritium permeation to the coolant: = 1 d steady state in tritium extraction system: = 6 h (time constant of the purge system)+ test. Duration of continuous operation a - during early technology phase: 10 to 20 h - at the end of the technology phase: 5 d -

Recommended cycle parameters during continuous operation b. burn time shot length: >/1000 s - duty cycle c: >/50% - off burn time: ~<1000 s

--

a Coolant flow and inlet coolant temperature control during off-burn-time. b Presently foreseen during hybrid operation: burn time = 2290 s, off burn time = 350 s, duty cycle = 85%. c Duty cycle = (burn time/(burn time + off-burn time))average.

94

M. Dalle Donne et al. / The Karlsruhe solid breeder blanket

these structures under neutron irradiation (70 dpa peak in Manet), stationary and cycling thermal stresses, and other stresses. Behaviour under plasma disruptions. The conceptual design of the test module, of its extraction system and of the required out-of-reactor ancillary system (two helium cooling systems, one purification system for the cooling helium and one purge gas system) has allowed to estimate thc time constants of the various components and thus allowed to assess the requirements given by the testing of the modules on the N E T / I T E R machine (burn time, duty cycle and duration of continuous operation).

Acknowledgement This work was performed in the framework of the K ~ Nuclear Fusion Project and is supported by the E u r o p e a n Communities within the E u r o p e a n Fusion Technology Program.

References [1] Minutes of the 4th Meeting of the Test Blanket Advisory Group, CCR Ispra, March 14th, 1990, unpublished.

[2] M. Dalle Donne, ed. BOT Helium Cooled Solid Breeder Blanket Status Report, KfK contribution to the development of a DEMO relevant Test blanket for NET/ITER, KfK Report 4929, Kernforschungszentrum Karlsruhe. [3] M. Dalle Donne, U. Fischer, M. Kfichle, G. Schumacher. G. Sordon, E. Bojarsky, P. Norajitra, H. Reiser, H.D. Baschek and E. Bogusch. Pebble-bed canister: The Karlsruhe ceramic breeder blanket design for the Next European Torus, Fusion Technol. 14 (1988) 1357-1388. [4] U. Fischer, Impact of ports on the breeding performance of liquid metal and solid breeder blankets for the DEMONET configuration, this conference. [5] M. Dalle Donne and S. Dorner, Tritium control in a helium cooled ceramic blanket for a fusion reactor, Fusion Technol. 9 (1986) 484-491. [6] K. Forcey, D. Ross, I. Simpson and D. Evans, Hydrogen transport and solubility in 316 L and 1.4914 steels for fusion reactor applications, J. Nucl. Mater. 1611 (1988) 117-124. [7] D. Baldwin, D. Gelles and O. Slagle, Tritium release from irradiated beryllium at elevated temperatures, J. Nucl. Mater. 179-181 (1991). [8] E. Proust, CEA, April 1990. unpublished.