.- W. M. CA 94550/67531/metadc... · allow high Pn at high bootstrap current fractions with good alignment between the desired total plasma current density and the bootstrap current

UCRL JC-118673 PREPRINT

Advanced Tokamak Operating Models in TPX and ITER

W. M. Nevins Lawrence Livermore National Laboratory

Livermore, CA 94550

.-

This paper was submitted to thq Tokamak Concept Improvement Workshop

Varenna, Italy August 29 - September 3, 1994

August 1994

This is a preprint of a paper intended for publication in a journal or proceedings. Since changes may be made before publication, this preprisit is made available with the understanding that it will not be cited or reproduced without the permission of the author.

DISCLAIMER

Portions of this document may be illegible in electronic image products. Images are produced from the best available original document.

Advanced Tokamak Operating Modes in TPX and ITER W.M. Nevins

LLNL, Livermore CA 94551 USA -.

quantify this improvement by scaling p* to the normalized current, defining

* P* Pn =(Ip/aB,)

Abstract

A program is described to develop the advanced tokamak physics required for an economic steady-state fusion reactor on existing (short-pulse) tokamak experiments; to extend these operating modes to long-pulse on TPX; and finally to demonstrate them in a long-pulse D-T plasma on ITER.

I. Introduction

Tokamaks have proved to be the most successful device for the magnetic confinement of plasmas. They have performed so well that the world fusion programs are planning to construct a large tokamak reactor (ITER) that, based on the tokamak physics data base, we confidently expect will produce 1.5 GW of fusion power. Given these successes, we must begin by asking "why do we need Advanced Physics in tokamaks?

From a programmatic perspective the answer to this question is related to the goal the US, EC, RF, and Japanese fusion programs-which is to develop commercially viable electric power plants. The many successes of tokamaks in the magnetic confinement of plasmas give us confidence that a tokamak fusion reactor will be capable of generating net power. However, what is necessary for a commercially viable power plant is that it generate net revenue for its owners. We are faced with the problem that the (projected) cost of electricity from a fusion reactor based on long-pulse, inductive tokamaks is marginal (2 100 mills/kW-hr, see Fig. 1). Hence, We need to improve tokamak performance. That is, we need to develop means of increasing both the plasma pressure as measured by P*=2h<p*>1/2/%2, and the energy confinement time, 7,. We can

while scaling ZE to the ITER '89 power law scaling by defining

A long-pulse inductive tokamak reactor with today's physics database) with

would be expect to operate (consistent Pn*s 3, and H 5 2. A substantial

improvement in the cost of electricity (a reduction by about a factor of 2) could be obtained if we could instead operate with pn*5 6, and H ,< 3. This is illustrated by the "TPX reactor" df Fig. 1, which assumed 5, and Z E / Z ~ R - P = 2.5.

250

s 0 C .- .- E 2 100 w

0

Figure 1. Projected cost of electricity for fusion power plants (figure courtesy of J. Perkins).

From a physics perspective, the shift in focus toward advanced physics results from the success of the tokamak physics program. The major physics issue for the plasma core of a long-pulse, inductive tokamak has been energy confinement. This issue is now nearly resolved-we are reasonably confident

-2-

that the energy confinement required for ignition in a tokamak can be achieved by increasing the size and plasma current by about a factor of 3 beyond what has been realized in experiments to date. This increase in the scale of tokamaks is largely a job for engineers. However, operation at high normalized P and 0 with enhanced energy confinement provides a clear challenge to the ability of tokamak physicists to understanding and control the core plasma. Hence, I expect that Advanced Physics will become the dominate issue for the world tokamak physics program.

We must next ask what "Advanced Physics" in needed to improve the economics of a tokamak fusion power reactor? Clearly, there are two key issues-energy confinement and P-limi ts. We begin by considering energy confinement the advantages of improved (beyond H-mode) energy confinement. The surprising answer that we get from our colleagues in the reactor-studies area is that better energy confinement alone will not greatly improve the economics of tokamak power reactors! Improved energy confinement will allow ignition in smaller tokamaks. While this could greatly reduce the cost of ignition experiments (grid the cost of the program to develop commercial fusion power), it does not improve the reactor economics if these reactors are constrained to present P-limits because the power density will be too low in a smaller tokamak. As a result the fusion power is reduced, while the size of the blanket, shield, and magnet set (which must surround the blanket and shield) cannot be correspondingly reduced because the shield thickness depends only logarithmically on the fusion power. Hence, the cost of electricity does not improve greatly with improvements in energy confinement alone. However, if we can achieve higher Pn* and higher H simultaneously, we can expect substantial improvements in tokamak reactor economics. This is because higher Pn* will allow higher power density at fixed magnetic field, leading to more fusion power for same investment; alternatively, we might "spend" any additional margin in Pn* by reducing the size of the tokamak. In combination with improved energy confinement, this would allow a smaller (and less costly) tokamak reactor to generate the same power. An additional advantage to operation at higher Pn* is that will allow higher bootstrap current fractions, thereby reducing the recirculating power requirements for a steady-state tokamak fusion power reactor.

-3-

2. How to Achieve "Advanced" Tokamak Operation

How will we get advanced performance from tokamaks? It is important to note that our goals (f3; - 6, and H - 3) have been achieved transiintly in present experiments. What is required is to extend such high-performance operation to steady-state. Clearly, the first step is to reproduce those conditions that have demonstrated to improve tokamak performance in present experiments, and then provide the additional control required to extend this regime to steady-state. In particular, we will require control over the particle recycling and the pl'asma shape because this has proved critical to enhanced performance in present tokamaks; and over the current profile and the plasma rotation because our present understanding of f3-limits and transport suggests that this will allow us to extend enhanced tokamak operation to steady-state.

Control of particle recycling has proved to be crucial to obtaining advanced confinement modes (like H-mode and VH-mode). In addition, density control is essintial for efficient non-inductive current drive because the current drive efficiency scales as l/ne. Hence, an inability to control the plasma density will interfere with steady-state tokamak operation and current profile control. In the present generation of tokamaks the key experimenta! tool in particle recycling control has been wall conditioning. In future steady-state tokamaks the key tool will be a pumped divertor. While the database is now small, present indications are that divertor pumping can successfully control particle recycling and the plasma density.*

Control of the plasma shape allowed the DIII-D tokamak to achieve high beta and enhanced energy confinement simultaneously.2 In particular, what is required is a relatively high plasma elongation, K~ 2 1.7, and high triangularity, & 2 0.4. Compatibility between such high plasma triangularity and acceptable divertor conditions demands a double-null configuration so that most of the power exhaust goes to the outboard divertor legs.

The demonstration of high+ plasmas for times long compared to Alvfh times strongly suggests that the termination of the high-performance phase of the discharge is due to evolution of the current profile on the resistive time-scale.

4

Hence, control of the plasma current profile is likely to be the key to extending such enhanced performance to long pulse and steady-state operation. The current density profile in the plasma core can be modified through the use of neutral beams, fast waves in the ion cyclotron frequency range, or high frequency electron cyclotron.frequency range waves. Current drive in the outer ha’if of the plasma (r/a = 0.5-1) can be accomplished with lower hybrid or electron cyclotron waves.

There is substantial theoretical and experimental evidence that plasma rotation is important to both MHD stability and enhanced confinement. To date, the most important driver for toroidal rotation in tokamaks has been tangential injection of neutral beams. Methods of generating significant plasma rotation with RF waves may also become important. For example, ICRF minority resonances might be employed to drive minority ions out of the tokamak, thereby charging the plasma and causing rotation; while recent experiments on PBX-M suggest that ion Bernstein waves can be used to generate local shear in the rotation velocity.

I

3. A Program for Developing Advanced Tokamak Physics

So, What is our Game Plan-how can we hope to develop the physics database required for an advanced tokamak reactor? I see it as having three key elements:

3.1. Develop candidate advanced operating modes on present tokamaks for time scales Zpulse ,< Zskin. Many promising operating modes have already been identified, such as the VH-mode as seen on DIII-D, the PEP-mode seen on JET, the LHEP mode seen on Tore Supra, and the high+, H-mode seen on JT-60U. These, and other promising advanced operating modes must be explored further in order to gain a better understanding of the operating limits in such advanced operating modes, and an understanding of how they can be extended throughout the duration of the tokamak pulse.

3.2 Extend advanced physics operation to long pulse, Zpulse >>T&in. This can be begun on present large tokamaks, and is the central mission of the TPX

-5-

program in the US. Key issues in extending advanced tokamak operation to long pulse and steady-state will be control of the particle recycling (and impurities), and control of the current profile. Hence, this will require not only a pulse length long compared to the resistive skin time, but also an actively pumped divertor and substantial non-inductive current drive.

*

3.3 Demonstrate advanced tokamak operation in a long-pulse, D-T plasma. This is an essential element of any strategy to develop a power reactor that takes advantage of advanced tokamak operating modes. Within the context of the US program, this step must be accomplished on ITER.

4. Reversed-Shear Modes Are a Leading Candidate for Advanced Tokamak Operation in TPX and ITER

The operating mode that shows the most promise of meeting our goals for high-P and enhanced energy confinement (Pn* - 6, and H - 3) is the reversed shear mode.3 ':4 MHD stability calculations show that these configurations will allow high Pn at high bootstrap current fractions with good alignment between the desired total plasma current density and the bootstrap current density (see Fig. 2).

A key feature of the reversed shear mode is a profile of the safety-factor, q(v), that decreases as you move out radially from a maximum at the magnetic axis until you reach a minimum in q(v) at r/a > 0.5. Plasma Transport should be strongly reduced in reversed-shear region because tearing modes stabilized by shear reversal, hence, reducing the transport expected from the Rebut-Lallia- Watkins model.5 Electrostatic micro instabilities are also stabilized by reversed shear.* Hence, drift-wave driven transport should be suppressed. Experiments on JET6 (PEP-mode) and Tore Supra7 (LHEP-mode) have demonstrated reduced transport in the reversed-shear region.

-6-

Recent TPX Calculation with Inverted q Profile

r I

l-4

I a. a a

m

m

4

CI)

.V \ A

7 V

3 . 4

3.2

3 . 0

2.8

2 . 6

2 . 4

2 . 2

i = 2 M A , a PN =4.8,

P .

ria

B, = 4T, pN* = 6.3,

Stable to n = 1,2,3, 00

? ? c

q(a) = 3.55 p = 4.8%) p* =: 6.3%

with wall at r = 1.3a

v 9 ? E9 N. N N N c N

R

93 O/O bootstrap cu r re n t

#RJG-327 0912 1/93

I

5. The TPX Program

The Tokamak Physics Experiment (TPX) is presently being designed within the US program in order to study advanced tokamak operating modes and demonstrate that they can be extended to steady-state. TPX is equippe; with a strongly shaped plasma, full non-inductive current drive with active control of the current profile, active particle recycling control, and high aspect ratio. These provide the device with the flexibility to explore in steady state advanced operating regimes with high PN, ZE, and Ibs/Ip that are required for an advanced steady-state fusion power reactor. Steady-state operation of TPX permits these studies to be extended to time scales far exceeding the global current-relaxation time and the plasma-wall equilibration time. The reference design and operating parameters are shown in Table I.

Table I. TPX Reference Design Parameters

Major radius, & (m)

. Minor radius, a (m) Aspect Ratio, R,/a Elongation, K~ Triangularity, ax Toroidal field, BT (T) Plasma Current, Ip (MA) Pulse Length (s) Neutron Budget (#/year) Heating and Current Drive:

Neutral Beam ( M W ) ICRF (MW) Lower Hybrid ( M W )

2.25 0.50 4.5 1.6 -+ 2.0 0.8 4.0 2.0 1000 4 00

6,1021

8 -32 8 +18 1.5 -+ 6.0

The ranges given in Table I indicate device flexibility ( K ~ ) , and device upgrade capabilities (pulse length, heating and current drive systems).

The TPX heating and current drive system includes 8 M'vZl's (upgradable to 18 MW) of fast wave power for electron heating and central current drive; 8 MW's of neutral beam power (upgradable to 24 MW) for ion heating, bulk

-8-

current drive, and plasma rotation control; and a 1.5 MW's of lower hybrid system (upgradable to 6 MW) featuring real-time control of the launched NII spectrum for active control of the plasma current profile.

6. Will ITER Support An Advanced Tokamak Physics Program?

TPX has been specifically designed to extend advanced tokamak physics to long pulse and steady-state operation. However, ITER'S fundamental goal is to achieve controlled ignition and extended burn of a deuterium-tritium plasma, with an ultimate goal of demonstrating steady-state operation. It is not immediately obvious that a tokamak designed for long-pulse ignited operation will be suitable for demonstrating advanced tokamak physics in a D-T plasma. We have examined the prospects for reversed shear operation in ITER, and project that ITER can sustain steady-state operation while producing 1500 MW of fusion power (assuming a 12.5% helium-ash fraction) with ZOO M W of auxiliary power. Details of this operating point are given in the column labeled "Steady State Physics -Mode" in Table 11; while the equilibrium configuration, and profiles of pressure, safety factor, and plasma current are shown in Fig. 3.

Figure 3. Steady-State Operation in ITER

a.

100.

0 .

-im .

-4m;

Reversed-Shear Equilibrium

mxl

em

-am

E i o n 3 s s o i : 2

Pressure and q-Profiles Current Profiles

0.0 0.5 1 .o

0.60

0.40

0.20

0.00 0.00 0.50 1.00

sq r t ( V / V - O )

-9-

s q r t ( v / v-0)

This operating scenario has been examined for MHD stability with the PEST code, and found to be stable for Pn* 50.067 (corresponding to pn-0.05) in the presence of a conducting wall at 1.25 times the plasma radius. The profiles of safety factor (solid curve), plasma pressure (dashed curve shows 10xP/Po ), total plasma current density in MA/m2 (solid curve), and bootstrap -';current density (long-dash curve) and driven current (short-dash curve) are shown in Figure 3. This operating mode has a bootstrap current density that is reasonably well aligned with the total current density with a high bootstrap current fraction (I~s/I~=0.80). We note that bootstrap current fractions exceeding 90% are possible with a small reduction in the plasma current (to about 12 MA) and a slight modification in p(v). We have chosen to emphasize a case with lower bootstrap current fraction to illustrate the importance of pressure profile control in achieving high bootstrap current fraction, and the possible requirement for non-inductive current drive in the outer half of the plasma to control the location of the shear-reversal point.

Table 11. Advanced Steady-State Operating Modes in ITER

Parameter

f BS

f He

Pn/Pn'

TE / TITER-P

Steady-State I DEMO Phys. I Blanket Test Phvsics Mode Mode Mode 1500/ 100 2000/50 4000/ 100 15 40 40 8.56/2.56 8.46 /2.5 8.46 / 2.68 1.79 /0.42 1.W0.46 1.71 / 0.44 5.4 4.34 6 4.46 3.7 4.4

80% 0.91 0.94 12.5% 12% 10% 3.0%/4.0% 5.0%/6.7% 3.6% /4.9%

II ~~ ~

0.96/ 1.52 -1 1.25z.O I1.2/1.9

II ~~

12.3/21 .O 110.2/18.4 I14.5/26.3

2.84 12.64- 12.35 II 2.23 12.59 12.18 II

-1 0-

The energy confinement time required at the "Steady-State Physics" operating point is 2.8 s (for an energy enhancement of 2.23 relative to the ITER-89 confinement scaling-greater than required for ITER ignited operation, but substantially less than what has been observed in JET PEP modes). s

The free-boundary MHD code TEQ has been used to study the compatibility of this advanced operating mode with the ITER poloidal field system and divertor. Appropriate gaps between the plasma and first wall can be maintained while mapping the scrape-off-layer (those flux surfaces that pass within 5 cm of the separatrix at the outboard mid plane) into the divertor chamber for a plasma with major radius Ro=8.56 m, minor radius a=2.56 m, elongation Kg5%=1.79 and triangularity <@5%> = 0.42. In particular, the separation between the plasma and the first wall at the polodial location of the ICRF antenna is within the range anticipated for 24 MA operation, so that this configuration does not pose difficulties for ICRF (or LH) coupling beyond those anticipated for ignited operation. This rather strongly shaped configuration can be achieved a t a flux state consistent with an inductive current ramp-up starting from a pre-bias that is within the capability of the ITER transformer, while maintaining the currents in all poloidal field coils well below present design limits. The growth rate of the vertical instability in the presence of the ITER passive stiucture has been computed. The lower current and higher triangularity of this plasma configuration compensate for the greater elongation and increased gap between the plasma and the inner wall to yield resistive growth rates similar to those obtained for the reference 24 MA operation. We conclude that this magnetic configuration is compatible with the ITER divertor, poloidal field, and vertical position control systems.

In addition to the ultimate goal of demonstrating steady-state operation, ITER could be employed to develop the physics data-base for a steady-state demonstration reactor. The column labeled "DEMO Physics Mode" in Table I1 shows how a high Pn* operating mode can be achieved in ITER without greatly increasing the fusion power by operating at decreased toroidal field. It is anticipated that, if ITER is to be used for high fluence blanket testing, the ITER blanket and first wall would be upgraded to allow a substantial increase in the fusion power. Attractive operating points for such high fluence testing must

-1 1-

have high duty factor and high fusion power. Steady-state operation would be desirable for high fluence blanket testing because this would allow the duty factor to approach I, while simplifying the interpretation of the nuclear testing data. The column labeled "Blanket Test Mode" in Table I shows an advanced steady-state operating mode that produces 4 GW of fusion power whilestaying within plausible limits on confinement and beta.

7. Issues for Advanced Tokamak Operation on ITER

At the present level of design it appears that ITER will be suitable for demonstrating advanced tokamak physics in a long-pulse D-T plasma. However, several issues have been raised in our examination of the prospects for such operation in ITER. These issues must be addressed in either the baseline ITER design, or by upgrading ITER in preparation for the enhanced performance phase provided for in the ITER mission statement* if ITER is to be employed to demonstrate the advanced tokamak physics required for a steady-state fusion power reactor.. These issues include:

7.1 Current Profile Control. Will the ITER heating and current drive system provide sufficient control of the current-profile? This is a problem that must be addressed by the ITER JCT. At present, the ITER workplan calls for the choice of a single H&CD system by March of 1995. However, a reversed-shear operating mode will require on-axis current drive to provide the "seed current" and off-axis current drive to control location of the shear reversal. Hence, two heating and current drive systems will probably be required.

7.2 Stability to Kink Modes. The MHD calculations refered to in the previous section assumed a perfectly conducting wall. However, we require ITER to be stable to kink modes on the resistive time scale of the first wall as well. Indications from tokamak experiments and theoretical work presented by Bondeson at this meeting are that full stability can be accomplished by the combination a conducting wall and plasma rotation. The present ITER design has a first-wall that is a toroidally and poloidally continuous laminate of 0.5 c m of copper, 0.5 c m of beryllium, and stainless steel located at r/a=l.l. Hence, the first wall will provide the required passive stabilization to kink modes. However

-1 2-

it is not yet clear how the required plasma rotation is to be maintained in ITER. Further refinement of the estimates of the rotation required for full stabilization is a task for the tokamak physics community, while devising a means to drive the required rotation in ITER is a job for the ITER JCT.

7.3 Pressure and Current Profile Control. The maximum bootstrap current fraction limited by our ability to control the pressure and current profiles. The paper by Moreau at this meeting illustrated the need to control these profiles, together with the overall fusion power on the resistive time-scale. This issue needs to be addressed on future long-pulse devices, like TPX, so that appropriate control schemes can be devised for ITER.

8. Conclusion.

Tokamak concept improvement is necessary for the development of economic fusion power reactors. Steady-state tokamaks operation at higher normalized p Znd energy confinement (Pn*4% and Hz2.5) than is presently envisioned in, for example, the ITER design (which takes &*“2.7% and H=2) shows great promise as a means to achieve such concept improvement. The advanced steady-state tokamak operating modes required for such a steady-state fusion reactor are being developed on existing tokamaks. These operating modes can be extended to pulse lengths long compared to he resistive skin time on TPX; and demonstrated in long-pulse, D-T plasmas on ITER. Potentially relevant advanced operating modes have been identified both in large tokamak experiments, and theoretically. TPX and ITER promise an opportunity to explore the exciting new physics issues raised by such advanced steady-state operation. Advanced tokamak operation is the central TPX mission, while the ITER JCT has demonstrated an interest in these operating modes. We can expect the efforts of the world tokamak community over the next few years in developing advanced steady-state operating modes to influence the ITER design-although, full advanced tokamak capability in a long-pulse D-T plasma may await ITER‘S enhanced performance operating phase.

-1 3-

ACKNOWLEDGMENTS

We wish to thank the Centro di Cultura Villa Monaster0 and Professor - Elio Sindoni for their hospitality during the Tokamak Concept Improvement Workshop. This work was performed under the auspices of the U.S. Department of Energy by the Lawrence Livermore National Laboratory under contract number W-7405-ENG-48.

REFERENCES

1M.A. Mahdavi I et al, Divertor Heat and Particle Control Experimenfs on the DIII-D Tokamak I Proceedings of the 1 l t h International Conference on Plasma Surface Interactions in Controlled Fus9ion Devices (May 1994, Ibaraki-ken, Japan).

2A.W. Hyatt et al, Plasma Shape Experiments for an Optimizzed Tokamak, Proceedings ef the 21st European Plysical Society Conference on Controlled Fusion and Plasma Physics (June 1994, Montpellier, France).

3T. Ozeki et al, Nuclear Fusion 33,1025 (1993).

4C. Kessel et al, Phys. Rev. Lett 72,1212 (1994); J. Manickam et al, Phys. Plasmas 1,1601 (1994).

5P.H. Rebut et al., Plasma Physics and Controlled Nuclear Fusion Research, (IAEA, Vienna, 1991).

6P. Kupscus, et ai., Proc. of the 18th European Conference on Controlled Fusion and Plasma Physics Research (Berlin, 1991) I, 1.

7Equipe Tore Supra, Plasma Physics and ControIIed Nuclear Fusion Research (IAEA, Vienna, 1993).

*ITER Council, ITER Council Proceedings: 1992 , ITER EDA Documentation Series No. 3 (IAEA, Vienna, 1994). See pg. 53.

-1 4-