HAL Id: jpa-00229353
https://hal.archives-ouvertes.fr/jpa-00229353
Submitted on 1 Jan 1989
HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
HEAVY ION INERTIAL FUSION
R. Bock
To cite this version:
R. Bock. HEAVY ION INERTIAL FUSION. Journal de Physique Colloques, 1989, 50 (C1), pp.C1-
489-C1-504. �10.1051/jphyscol:1989154�. �jpa-00229353�
JOURNAL DE PHYSIQUE
Colloque Cl, supplement au nol, Tome 50, janvier 1989
HEAVY ION INERTIAL FUSION
R. BOCK
Gesellschaft fiir Schwerionenforschung (GSI), 0-6100
Darmstadt,
F.R.G.Abstract
The prospects of Heavy Ion Inertial Fusion for energy production are discussed and current programs are reviewed.
1. Introduction
Thermonuclear fusion by inertial confinement (ICF) is a potential route to electric power generation i n the next century. In our present understanding it is the most promising alternative among the various fusion concepts.
Concerning the choice of the driver for an ICF-based power plant, research accomplished during the past dec- ade has strongly increased our confidence that the heavy ion accelerator is the best
-
if not the only-
driver candidate for an economic power generation.It is, therefore, mandatory to study the feasibility of such a concept seriously and to investigate, whether a heavy ion accelerator can fulfil1 the necessary conditions and can be built at a reasonable cost level.
What are these conditions? Experiments with high intensity laser beams carried out i n particular in the U.S. and systematic theoretical studies have provided the necessary data base for the most crucial issue of inertial con- finement fusion, the implosion of the pellet. With respect to the conditions for ignition and burn, experiments with laser beams have now nearly reached breakeven. Also the more spectacular experimental results of pellet implosions achieved by underground nuclear explosions which were reported recently1, confirm the theoretical predictions for the necessary pulse energy of about 5 MJ. Unfortunately, laser facilities as drivers for a power pfant have serious shortcomings, i n particular very low efficiency and an extremely low repetition rate, and it is hard t o imagine how the necessary orders of magnitude could b e overcome to reach the requirements.
For heavy ion accelerators both the efficiency and the repetition rate is no critical issue at all and other re- quirements such as the final focusing and the beam-pellet coupling look satisfying. The enormous progress in accelerator technology and the experience i n operational reliability with extended and complex accelerator fa- cilities open favourable perspectives for a heavy ion driver. The necessary beam intensities of such a driver, however, exceed those of existing accelerators considerably. A number of open problems, in particular prob- lems of space-charge dominated beam dynamics and beam handling needs to be further investigated before a final judgement can be made.
In our feasibility study HIBALL*, started in 1980 and upgraded 1984, a fusion reactor scenario based on an rf- linaclstorage ring driver concept has been investigated. HIBALL was the first conceptual design study for a heavy ion driven fusion reactor, i n which the key issues for such a driver scenario have been identified sys- tematically. Other system studies followed in Japan and in the U.S. and have confirmed the favourable prospects
Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1989154
Cl-490
JOURNAL DE PHYSIQUE
for an accelerator driven ICF power plant. Work on HIBALL was carried out in the framework of a government- funded program in Germany and included also the development of some specific accelerator components such as high-brilliance ion sources and RFQ structures, as well as ICF relevant theoretical studies. Some of these issues have been investigated experimentally, but crucial problems were not accessible to experimental in- vestigation because of lack of appropriate facilities.
Consequently, ideas for a more dedicated facility for such investigations were developed in the framework of our program. These ideas, combined with the needs of the nuclear physics community finally lead to the con- cept of a synchrotron and storage/cooler ring facility SlSlESR which at present is under construction at GSI and will enable in a year from now many investigations on key problems of heavy ion inertial fusion, both in the field of accelerators and in beam target interaction.
The present report. will concentrate on ongoing experimental work and on research to be investigated with this facility. Starting with a short reminder about basic facts and some useful numbers for a heavy ion inertial fusion scenario, I will briefly review the results and achievements of HIBALL and other subsequent systems studies carried out recently in the U.S. Then the SlSlESR facility and its parameters characteristic to heavy ion fusion will be described. Finally results of the present activities and the perspectives for the future program will be presented and discussed.
2. Some Facts and Numbers of Heavy Ion Inertial Fusion
A DT-mixture of a few milligrams is confined in a pellet of up to 10 mm diameter, which is heated and com- pressed by ablation up to ignition conditions. The energy is supplied by laser or heavy ion beams. The inertial forces of the implosion keep the pellet together for a few nanoseconds, long enough to burn a considerable fraction of the fuel. At an ignition temperature of about 5 keV a compression of about 103 (200g/cm3) has to be reached in order to fulfil1 the Lawson Criterium
(n is the number of D and T per cm3 and 7, the confinement time) which characterizes the breakeven condition.
A similar relation exists between the fuel density p and the fuel radius R at ignition
which relates the amount of fuel in a pellet necessary for substantial burn (which means about 30 %) with the necessary fuel compression. It tells us that e.g. for an amount of fuel which can be handled in a reactor chamber (up to a few milligrams), a compression of about 1000 times liquid density is needed.
An ICF power plant consists of three major facilities: (1) The driver, (2) the reactor cavity (including the con- ventional electrical installations) and (3) the pellet fabrication facility. One advantage of the ICF scenario as compared to magnetic confinement is the local separation of the driver and the reactor cavity which facilitates its construction and maintainance as well as the extraction of the generated energy. Others are the simple ge- ometry of the cavity and the absence of magnetic fields. Most of the technical problems of both concepts, mag- netic and inertial confinement, are very different from each other and will therefore not be discussed in comparison. For ICF many concepts and their technical implications have been investigated in recent years, but not much is published yet.
In trying to evaluate the situation of heavy ion inertial fusion, three basic questions have to be asked:
1) Does the pellet work as predicted by theory?
2) Can the overall concept be realized (technologically, economically, environmentally)?
3) Can the necessary beam conditions be reached by a heavy ion accelerator?
Concerning Question (1) as the most crucial issue of the ICF concept, the results with laser beams have pro- vided the essential facts for our present understanding. Some recent data shall be quoted in order to give an impression where we stand. Question (2) concerns the systems studies and I will try to summarize the latest results of studies accomplished in the U.S. on various driverlreactor concepts. Question (3) is the heart of the matter for our own research and will be covered in more detail.
Results with Laser Beams and Driver Evaluation
The trend for the laser facilities during the last years was to shorter wave length, aiming at a better coupling of the laser beam to the pellet, and to indirect-driven (X-ray driven) pellets, in order to achieve more symmetrical and homogeneous illumination. In the U.S., after SHWA (10 kJ) the new facility NOVA (100 kJ, 100 TW, 10 beams) is in operation since 1986. It is based on Nd-Glass (1.08 pm) and can also be operated at 112 or 113 wave length.
Doubling and tripling of the frequency can be achieved with high efficiency. As to recent reports3 1.7 keV and n T, = 1 . 5 ~ 10'4 slcm3 at a compression of nearly 100 times have been reached in X-ray driven implosions. The maximum ion temperature obtained in a pellet was about 10 keV. When operatedat its final design parameters, NOVA is supposed to reach a compression of 103 and will come near to breakeven. In Japan a similar facility (GEKKO, 40 kJ, 12 beams) is in operation. Because there is no classification many important results on the physics of pellet dynamics have been reported during the past year. The Lawson diagram of Fig.1 includes inertial as well as magnetic confinement facilities and shows that both confinement concepts are approaching burn conditions4
Concerning the future, a new facility "Laboratory for Microfusion (LMF)" is considered as a next step in the U.S..
Design aims are beam pulses of 10 MJ at 1000 TW, which would deliver an energy output of 1 GJ per pellet. If approved it could be completed until the end of the century at a cost level of about 850 M$.
For a comparison between various driver candidates, Table 1 shows a list of the essential requirements and its qualitative evaluation for electric power production and for an engineering test facility.5 This list; prepared by the LLNL laser group, shows a clear preference for heavy ion accelerators. Most important are the excellent prospects for its repetition rate, the beam-target coupling and its efficiency (about 25%).
Parameters for a heavy ion driver
As to our present understanding for the ignition of a pe:iet, a heavy ion pulse of 5 MJ has to be delivered within about 20 ns. Consequently the pulse power is 250 TW. This number is a product of the heavy ion kinetic energy and the beam current. Since the kinetic energy of the ions is limited to about 50 MeVlnucleon by their range in the pellet (10 GeV for the heaviest ions) the necessary pulse current must reach 25 kA. The corresponding figure
C 1-492 JOURNAL
DE
PHYSIQUEfor a proton beam, for example, would be 25 MA because the stopping power is lower by a factor of 1000 as compared to a Bismuth beam. Based on a 5 MJ pulse and a pellet gain of 100 (which due to theoretical results is a realistic assumption) the output energy of a single pulse would be 500 MJ, and assuming a pulse rate of 20 pulses per second a thermal power of 10 GW would be obtained which is equivalent to an electrical power of about 3 GW (Table 2). Some more problems, connected to the effective beamlpellet interaction, e.g. the pulse shaping, the range shortening in a plasma, all kind of pellet design questions, can not be discussed here.
3. Concepts for Driver Accelerators and System Studies
After the design parameters are given (Table 2), the question is whether an accelerator can meet these re- quirements. Among the various accelerator combination~ investigated, two concepts have survived the early discussions and are both considered as suitable drivers:
(1) the induction linac and (2) the RF-linac with storage rings
lnspite of some common features, the transport and manipulation of space-charge dominated beams, the tech- nical problems of both concepts are quite different. In particular, they are very different with respect to the pulse structure.
(1) The pulse sequence of the induction linac can be directly accommodated to the needs of the reactor right from the beginning. No transformation of the pulse structure is needed. Consequently, extremely high in- tensity from a pulsed ion source is necessary and a pulse compression during acceleration by many orders of magnitude.
(2) A high-current RF-linac may deliver a heavy ion beam continuously up to several 100 mA. In order to create the final pulse structure and the final pulse currents required, accumulator and buncher rings are needed for current amplification and storage rings to hold the pulses for several milliseconds before feeding the reactor.
Induction Linac Scenario
Research on heavy ion induction linacs has been carried out by Keefe et aL6 at LBL in Berkeley. The basic idea is to inject a long beam bunch and to achieve current amplification by ramping the inductive acceleration fields as the bunch passes. By this procedure the pulses have to be compressed from 20 ps at injection down to 10 ns at the target, the current being increased from amperes to kiloamperes. One of the important conceptual improvements was the splitting of a single high-intensity beam into a large number of parallel beamlets, each of them being separately focussed inside the same accelerating structure. This concept has improved focussing because of smaller emittance and has shown to be cost effective if the number of beamlets is in the range of 8 to 16. The present concept for a driver starts with 64 beamlets at injection which later, when space-charge forces decrease, are combined, each 4 beams into 1. For a 3 + charge state of bismuth, the whole length of the ac- celerator is about 5 km (Fig. 2).
Present work concentrates on a prototype experiment (MBE-4) for the multi-beam c ~ n c e ~ t . ~ l t consists of 4 beamlets. In spite of its limitation t o the front-end only it allows many critical issues to be investigated because the front-end with the initial pulse formation is the most critical part. These issues are in particular:
- t h e creation of long high-intensity pulses - t h e principle of current amplification
-
handling of multiple beams-
longitudinal and vertical beam dynamicsAs a next step an "Induction Linac System Experiment (ILSE)" has been proposed which is intended t o address all key issues of a full scale driver, including
-
transport of space-charge dominated beams-
combining and bending of beams-
compression and pulse-shaping-
final focussingILSE i s a 30 M $ project and is expected to be funded next year. Construction should be finished within 4-5 years and it might be necessary t o have another step in between before a full-scale driver can b e designed and con- structed.8 Table 3 shows a comparison of the parameters for the three facilities, MBE-4, ILSE and a full scale Induction Linac Driver.
RF-Linac Scenario
The RF-linaclstorage ring concept looks more conventional than the induction linac, because nearly all accel- erator and beam handling elements are known from operating facilities. No experience, however, exists about their operation in a regime of space-charge dominated non-relativistic beam dynamics, where various types of instabilities may occur and where the beam transport and any beam manipulations may be connected with se- vere emittance growth and intensity loss.
In order to identify all the critical issues we made a conceptual design study,
HI BALL^,
which was followed by similar studies in theu.s.~,
the USSR'O and in ~ a ~ a n l ' . I will first briefly describe our own concept (upgraded 1984, HIBALL [l1*) as far as the accelerator is concerned and focus on the problems identified and then present some results of the systems studies in general.HIBALL 11: Concept and critical issues. A 5 km RF-linac provides acceleration at constant current using RFQ, Wideroe and Alvarez structures (Fig. 3). Because of space-charge limitations the charge-state of the ions is chosen 1
+.
Bismuth was chosen as a beam, because it is mono-isotopic, but other heavy elements may be used as well. The front end starts with 8 parallel channels because of space-charge limits at low velocity. By funneling connected with frequency doubling the parallel channels are combined stepwise over a length of about 500 m. To achieve the 165 mA design value of the linac current, sufficient Bi' intensity with a normalized emittance of 0.2 10% rad was already obtained by sources developed at GSII~. One of the remaining problems in this area is the question of emittance growth by funneling. The rest of the accelerator, transfer, storage and buncher rings is devoted to current multiplication and to provide the adequate pulse frequency andCl-494 JOURNAL
DE
PHYSIQUEpulse shape for the reactor. The current multiplication factor is 8000 and the maximum emittance growth is al- lowed to be 30 at maximum. The maximum storage time of 4 ms is a compromise between the longitudinal microwave instability in the storage rings and the beam intensity provided by the linac. If the microwave insta- bility would turn out to b e not so serious the beam intensity of the linac could be further decreased. On the other hand, because of the short storage time the driver can provide a much higher pulse frequency (- 100 Hz) than one reactor chamber could use (5 Hz), the same accelerator therefore can serve a number of reactor chambers.
The HIBALL study assumes 4 cavities but even more would be possible, thus further reducing the cost of elec- tricity.
For the ring system and the injection into the reactor the problems listed in Table 4 have been identified. In this area riot much experience is available, so ideas and new techniques may improve the situation. Phase space stacking in the synchrotron, for example, might be greatly improved by changing the charge state of ~ i l + at injection by laser beams, as suggested by ~ u b b i a ' ~ . A method for the suppression of microwave instabilities by pulse shaping was proposed by ~ o f m a n n ' ~ .
Cost of Electricity. The breakdown of costs for the HIBALL II power plant can be characterized by the following nu mbers:
Linac (including building
+
installation) 20%Ring system and beam transport 39%
Reactor (4 cavities) 23%
Conventional facilities etc. 18%
The total cost of a 3.7 GW, power plant was 6.2 G$ (cost level 1985). Under these assumptions the cost of electricity would be about 4.8 cents/kWh.
During the last 5 years systematic studies about cost estimates and cost optimization have been accomplished in the
u.s.~.
In Table 5 comparisons are made between various concepts including Tokamak reactors. The comparison looks good for ICF. The variation of plant parameters (pulse rate, driver cost, size of the plant etc.) shows clear tendencies concerning the cost of electricity ( c o E ) ~ ~ . Strong dependence for the COE is observed from the net electric power of a plant and the number of reactor units per plant. For a reference design of a 1 GWe power plant an increase by 50 % would reduce the COE by 25 %. A 25 % change in the driver cost results in a 13 % change in COE. Since the driver cost is relatively high, increasing the number of cavities per plant also reduces drastically the cost of electricity (Table 6).4. Future Research at GSI
A government-funded program on heavy ion inertial fusion in W. ~ e r m a n ~ l ~ started with an exploratory phase in the first half of the past decade (1979-86). It was devoted mainly to the feasibility study HIBALL and to re- search on some key issues such as development of high-brilliance sources, RFQ and other accelerator relevant components, as well as to theoretical investigations on accelerator and target issues. GSI, KfK Karlsruhe, the Max-Planck Institute in Garching and a number of German universities participated in this program. The accel- erator development program concentrated on the front end of an rf-linac driver, the major achievements were a full scale high-brilliance ion source CHORDIS which meets already the driver ~ ~ e c i f i c a t i o n s ' ~ , as well as a low-frequency RFQ prototype injector18 for high-current injection into the existing UNILAC accelerator.
As a result of the HIBALL study it became obvious that for a final judgement on many critical issues the exper- imental data base is too narrow. Consequently in the second half of this period, starting 1983, a concept for an experimental facility was developed. It was partially based on the existing plans of the nuclear physics com- munity for a heavy ion synchrotron. Two additional devices were suggested:
- a high-current injectorq8 for high-intensity operation of the synchrotron and
-
a storage ring1' for beam dynamics studies at space-charge limits with a beam cooling facility for gener- ating low-emittance beams in order to create high energy density in a target.This accelerator facility S I S I E S R ~ O is now under construction at GSI in Darmstadt and will be finished in 1989190 (Fig. 4). With the high-current injector in operation, about 10" ions will be injected.
With the present funding period (1987
-
1991) we were entering into an experimental phase, predominantly de- voted to the preparation and performance of experiments with this machine including associated theoretical studies. With this accelerator facility nearly all driver relevant issues of ICF should be accessible to exper- imental investigation. Beam experiments will address the following problems:- instabilities in space-charge dominated beams
-
phase-space dilution in multi-turn stacking-
bunching and debunching, resonance crossing - bunch mergingIn an early stage before the ESR will be in operation, a bunch merging experiment is planned as depicted in Fig.
5, with the following steps: (1) Single-turn injection and debunching of the linac beam, (2) acceleration on the 4th harmonic, (3) bunch merging 4 + 2 bunches, (4) merging of 2 into 1 bunch and finally (5) fast bunch com- pression (factor of 10) and extraction ( c 100 ns) in order to use it for a beamltarget interaction experiment.
In Fig. 6, a comparison is made between the rf-linaclstorage ring driver scenario of HIBALL and the GSI facility.
Except the emittance growth by funneling, the final focussing and the final bunching, all kinds of beam manipu- lations and beam dynamics can be investigated experimentally.
In addition to the study of accelerator issues, a new area of research is opened by the beam cooling facility in the ESR. A beam of high phase-space density is produced which can be used for heating small samples of matter to temperatures up to about 20 eV or more (Table 7). The generation of these plasma samples of solid state density, well defined in time and geometry enables a new kind of spectroscopic investigations. This field of heavy ion plasma physics covers a broad spectrum of problems of beam plasma interaction, dynamics of hot dense plasmas, pumping of short wave length lasers, etc. A target station is being designed which will enable a number of proposed experiments in this field (Fig. 7). Concerning the ICF relevant issues categories of problems are indicated in Fig. 8. The category called "ion deposition and equation of state physics" will be the domain of investigations at SISIESR, studies on "radiation physics" need higher temperatures which might be reached by special target arrangements. In this field many key issues will remain open.
Finally I would like to flash on some highlights of ongoing experiments. An experiment on beam target inter- action is going to be carried out in collaboration with a French group from 0 r s a ~ 1 0 r l e a n s ~ ~ . Its aim is to measure the stopping power for heavy ions in a plasma, which is predicted to be enhanced as compared to cold matter. The experimental set-up as it is used at the 1.4 MeVlu beam of the UNILAC is shown in Fig. 9. The result exhibited in Fig. 10 clearly shows the drastic enhancement of the stopping power in a plasma, a result which can be reproduced quantitatively by calculations (Peter and ~ e ~ e r - t e r - ~ e h n ~ ~ ) . In another experiment at the
Cl-496
JOURNAL DE PHYSIQUERFQ injector (50 keV/u) the first heavy ion created plasma has been observed recently26. The temperature achieved is about 1 eV and the experiments are devoted to optical spectroscopy of heavy ion generated plas- mas.
Another class of experiments concerns the cross sections for ion-ion collisions, with respect to charge-changing collisions in the storage rings of a driver accelerator. Here for the first time a heavy system was investigated.
Salzborn et al.27 measured the system Bi' on Bi' in the energy range of 10-100 keV (cm. energies) which corresponds to the relative motion of the beam projectiles in the bunch (Fig. 11). A rough calculation gives a loss rate in the percent region for a HIBALL scenario.
5. Concluding Remarks
In the last International Symposium on Heavy Ion lnertial Fusion held some months ago in Darmstadt the progress on the conventional concepts of drivers was demonstrated and new ideas have been presented which may help to overcome existing difficulties. The increase of experience with the new large accelerator facilities being in operation or under constructing may have a considerable impact on driver scenarios14.
The following statements may characterize the present situation:
1) Heavy Ion Inertial Fusion is considered as a realistic route towards an economic energy production 2) Two accelerator concepts have been developed which may satisfy the driver requirements for a power
plant.
3) New facilities planned (ILSE) or under construction (SIS/ESR) will promote research on key issues of both concepts and the results obtained will help to define the future direction more clearly.
4) SISIESR will open, in addition, research on many exciting basic physics issues in areas of accelerator technology and beam dynamics as well as of beam plasma interaction.
References
1) W. J. Broad, New York Times, March 22, 1988, p. A I
2) B. Badger et al., HIBALL
-
A Conceptual Design Study...
Report KfK - 3202 (1981) 2 Vols.3) 1986 Laser Program Annual Report, LLNL Report UCRL 50021-86
Ei.
M. Campbell, p. 3-14) S. Witkowski, Nuclear Fusion
3
(1985) 227 5) Ref. 3, W. J. Hogan p. 8-646) D. Keefe et al. LBL-Report 23168 (1987)
7) T. Fessenden, Proc. of the 4'h Int. Symposium on Heavy Ion lnertial Fusion, Darmstadt June 1988, Eids. R. Bock, I. Hofmann and J. Meyer-ter-Vehn;
to be publ. as a special issue of Nucl. Inst. and Methods, Section A, 1988 8) C). Keefe, Ref. 7, Invited Paper
9) W. R. Meier et al., AIP Proc. of the Conf. on Heavy Ion lnertial Fusion
Washington D. C. May 1986, Eds. M. Reiser, T. Godlove and R. Bangerter, New York 1986, p. 515 C). J. Dudziak et al. same Proceedings, p. 111
10) 1. M. Kapchinski et al. Ref. 9, p. 49
C). G. Koshkarev et al., Ref. 7, lpvited Paper
11) I-. Yamaki, Proc. of the Int. Symposium on Heavy Ion Accelerators and their Applications to Inertial Fusion, INS Tokyo 1984, p. 141
Y. Hirao, Ref. 9, p. 39
12 HIBALL 11, Report KfK-3840 (1985)
13) R. Kel'ler et al., Proc. Int. Ion Engin. Congr. Kyoto 1983, p. 25 14) C. Rubbia, Ref. 7, lnvited Paper
15) 1. Hofmann, Ref. 9, p. 74 16) W. R. Meier et al., Ref. 3, p. 8-27
17) R. Bock, Ref. 9, p. 23 and Plasma Physics and Controlled Fusion
3
(1984) 237 18) R. W. Miiller et al., Ref. 9, p. 156 and Report GSI-86-1319) I. Hofmann, Internal GSI-Report, GSI-SIT-1~~183, Oct. 1983 20) P. Kienle, Nucl. Physics
A478
(1988) 847D. Bohne, Ref. 7, lnvited Talk (1988) 21) I. Hofrnann, Internal GSI-Report, Dec. 1987 22) D.H.H. Hoffmann, private communication
23) R. Arnold and J. Meyer-ter-Vehn, Report on Progress in Physics
50
(1987) 559 24) C. Deutsch et al., Europhysics Letters, to be published 1988D. H. H. Hoffmann, Ref. 7, lnvited Talk (1988) C. Deutsch Ref. 7, lnvited Talk (1988) K. Weyrich, Thesis 1988
25) T. Peter and J. Meyer-ter-Vehn, Ref. 7, lnvited Talk (1988) 26) J. Jacoby et al. to be published
27) E. Salzborn and A. Melchert, J. Phys. B, to be published E. Sa17hnrn Rof 7 In\ritorl Talk i 1 9 A A \
ci T
FTRC @-~hiva
d
Asdex
PLT
10"
-
Magnetic Confinement O Inertial Confinement
Temperature
t
keV l ----cFigure 1
Lawson diagram showing the product of density n and the confinement time z, versus the temperature achieved .in some magnetic and inertial fusion devices (witkowski3).
JOURNAL DE PHYSIQUE
10 MeV 100 MeV 10 GeV 10 GeV
50 A 250 A 10 kA 100 k A
L1 AGNETIC
XRI 865-1965
Figure 2
Induction Linac Driver (A
--
200, q = 3). The length of the accelerator is about 5 km.IKII<
3!
320 Htk7
,X
\ 2 ~ 1 0 u rStorage times
turns ps
A R 1.2 3 i8 W R 1 3 L8 W R Z 6 96 WR 3 12 192
Multiturn 7 lnject~on u ~ a s t Kicker W RF Cavity --.L Switch + Combmw
Septum array (static) S = Stacking factors D = Dilution factors E
=
transverse emittance H=
horizontalV
=
verticalAplp= relative momentum deviation I = peak beam current
Figure 3
RF-LinacIStorage Ring scenario as a driver for a power plant'2 ( H I B A L L 11). The storage time is 4 ms.
SIS
Figure 4
The SlSlESR accelerator facility at GSI~'. SIS is a fast
cycling heavy Ion synchrotron for energies up t o 2 Figure 5
GeVlnucleon for light projectiles and up t o 1.3 Bunch manipulation experiment at high intensity GeVlnucleon for heavy projectiles. The SIS diameter as described in the text2'. Beam parameters for i s 65 m. The "Experimental Storage Ring" ESR has the various stages (Phase I without ESR, Phase II half the circumference and half the bending power. with ESR) are given in Table 7.
S ~ a r a l l e l B ~ a r a l l e l A ? a r 3 1 1 ~ 1
channelr channels cliannelr 2 . 1 channel
Problems t o b e i n v e r t i g a t e d
DebunChing L o n g i t v n d i n a l
o f :he l i n a ~ beam and t r a n s v e r s a l
41BALL X u l t i - t u r n i n j e c t i o n m'crow*dve
a t space charge l i m t r i"lLab"ity F a s t bunching.
Resonance c r o s s i n g
Bunch merging Beam c o o l i n g
r e l e v a n t ( e l e c r r o n and
I t O C h a l t l C )
Figure 6
HIBALL relevant beam dynamics issues t o be investigated at SISIESR. The figures for emittance and intensities given for the GSI facility are experimental results.
JOURNAL
DE
PHYSIQUEFigure 7
Design of the experimental setup for investigations in heavy ion plasma physics22. The beam is focussed on a small spot (0.2 mm) by a special quadrupol lens system (QD, QT). Optical spectrometers with Streak Cameras (SC) measure the time evolution of the plasma target.
ION DEPOSITION AND EOS PHYSICS
H I B A L L
R F Q
BEAM 515- 18 R E 4 C T O R STUDYI l I I I
_
10-' 1 10 10' 10' 10
'
SPECIFIC DEPOSITION P O W E R p ( T W I q l Figure 8
Temperature reachedby energy deposition of a heavy ion beam in solid material as a function of the specific power*3. K - - - - - a
I I
/' I
/ /
Plasma
Capacitor Spark gap Figure 9
Experimental setup for heavy ionlplasma interaction. The plasma is produced by a Z-pinch arrangement, the stopping power is measured by a time-of-flight method.
I I I I 1 I I I
238 U
SH
plasma. gas-
Target density n,,Hr10'7
Figure 10
Stopping power for uranium ions of 1.4 MeVIu in a hydrogen plasma as a function of the plasma density.
Figure 11
Cross sections for Bi+
+
Bi++Bio+
BiZ+(cr,), Bi++
Bi+-+Bif+
BiZ++
e(~,) and 0, = 2(20,+
0;) measured in crossed-beamCl-502 JOURNAL DE PHYSIQUE
Table 1
Evaluation of requirements for various inertial fusion driver cancidates (W.J. iiogan4).
KrF SS FEL LID HID Requirements for electric power production
Pulse characteristics
++ ++
Pulse rate (5 to 10 Hz)
++
??Efficiency (> 10%) ? ?
Reliability ? ?
Durability (101° to 1011 pulses) ?? ?
Maintainability
+ +
Cbst
-
?Requirements for a n engineering test facility Sustained high pulse rate
+
? Ability to drive multiple chambers++ ++
Illumination geometry flexibility
++ ++
Diagnostic access
++ ++
Experiment geometry flexibility
++ ++
Cost ?? ??
++
= great advantage KrF = krypton fluoride laser+
= advantage SSL = solid state laser (Nd:glass)? = uncertainty or neutral FEL = free-electron laser
?? = great uncertainly LID = light-ion driver - = disadvantage HID = heavy-ion driver
Table 2
Beam and plant requirements for a heavy ion driver.
Pulse Energy (needed f o r p e l l e t i g n i t i o n ) 5 MJ
Pulse Duration (implosion time) 20 ns
Pulse Power 250 TW
P r o j e c t i l e Energy ( f o r B i o r Pb)
50 MeVInucl eon 10 GeV
Pulse Current 25 kA
Pulse Current per beam ( f o r 20 beams) 1.25 kA
R e p e t i t i o n Rate ( f o r thermal power 10 GW) 20 Hz (assumed g a i n 100; e n e r g y l p e l l e t 500 MJ)
Table 3
Beam Parameters of the prototype experiment MBE-4, the Induction Linac System Experiment ILSE, and a full s c a l e driver.
Parameter
MBE-4 ILSEDriver
Ion
CS+
c+(AL++) 200+*Final Volt. (MV) 1 .O Final CurrenUA) 0.
l
Beam Energy (J) 0.1 AccelGrad.
(MV/m) 0.07#
Beams
4Pulse Width 2- 1
Charge
(~COU~.) 0.1 Init.bunch
length (m) 1.1Table 4
Key issues to be investigated for the storage and buncher ring scenario of a power plant and its effect on per- formance.
Issue Consequence
- debunching of the linac beam
-
beam combining by septa - longitudinal instability-
beam transport over long distances -final bunch compression-
prepulse formation, pulse shaping - final focussing-
phase space dilution in multiturn stacking-
ion-ion charge exchange-
technological issuesmomentum spread emittance growth
holding time in storage rings emittance growth
pellet gain pellet gain intensity loss emittance growth intensity loss
Table 5
Comparison of the cost of various reactor concepts and the effect on cost of electricity (COE) for two plant sizes, 1.2 and 3.8 GWe (1985 Dollars). STARFIRE and MARS are magnetic fusion facilities, HlBALL and HlFlCAS heavy ion.driven plants. HlFlCAS is based on an induction linac (charge state 3 + ) and the reactor concept CASCADE of LLNL (W.R. Meier et al.').
N e t Power
-
1200 MWe N e t Power = 3784 MW, STARFIRE MARS HIFlCAS HIBALL-I1 HIFlCASD i r e c t c o s t (G$) 2.07 2.24 1.83 4.80 3.43
T o t a l c o s t (G$) 3.78 4.10 3.34 8.78 6.30
COE ($/kWeh) 6.13 6.33 5:14 4.59 3.07
Table 6
Changes in the cost of electricity (COE) for a heavy-ion fusion power plant resulting from changes in the refer- ence case parameters (given brackets). The reference case COE is 5.84 centslkWeh (W.R. Meier, W.J. Hogan and R.O. ~ a n ~ e r t e r " ) .
COE COE
Parameter (*/kWe- h) Parameter (VkWe.h)
Fulse rate Conversion efficiency
[5 HzIa W%]
2 Hz 6.5 ( + l l%)b 45% 6.2 ( + 6%)
10 Hz 5.7 ( - 2%) 35% 6.7 ( + 15%)
Driver cost Target gain curve
ICdI 0.75 Cd 1.25 Cd Reactor cost
lGl 0.75 C, 1.25 C, Reactor scaling
[0.49]
0.70 0.90
Target factory scaling [OI
0.6
5.1 (-13%) G,,
6.6 ( + 13%) Net electric power 11.0 GW.1
i'
3.9 ( - 33%)5.9 ( + 2%) 3 3.3 ( - 43%)
6.0 (+3%)
a The reference case parameters are given in brackets.
Fractional changes from the reference case 6.4 ( + 1°%) are given in parentheses.
Cl-504 JOURNAL
DE
PHYSIQUETable 7
Parameters for some SlSlESR beam scenarios and the corresponding temperatures obtained in a target by stopping the fine-focussed beam.
Phase I Phase 11
Ne"
.A+ Xeb1990 1991 1991
