N A S A T E C H N I C A L M E M O R A N D U M

NASA TM X-52269

* ICODEI / d M - IPAGESJ

INASA CR OR TMX OR AD NUMBER1 ; /A/lx-,rJa

SUPERCONDUCTIVE MAGNETS AT LEWIS RESEARCH CENTER OF NASA GPO PRICE $

CFSTl PRICE(S) $

by J a m e s C. Laurence Lewis Resea rch Center Cleveland, Ohio

Hard copy (HC) 7?. r 9 v

Microfiche (MF) .

We53 July65

TECHNICAL PAPER proposed fo r presentation at International C r y ogeni c Engineering Conference Kyoto, Japan, Apri l 9-13, 1967

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION WASHINGTON, D.C. 1967

~

I I .

SUPERCONDUCTIVE MAGNETS A T LEWIS RESEARCH CENTER OF NASA

by J a m e s C. Laurence

Lewis Research Center Cleveland, Ohio

TECHNICAL PAPER proposed for presentation at

International Cryogenic Engineering Conference Kyoto, Japan, Apr i l 9-13, 1967

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION

co r- r- M I w

SUPERCONDUCTIVE MAGNETS AT LEWIS RESEARCH CFNTER OF NASA

by James C. Laurence

Lewis Research Center National Aeronautics and Space Administration

Cleveland, Ohio

ABSTRACT

Construction details and performance data of special-purpose,

superconductive magnets planned and in use at the Lewis Research Center

of NASA are presented in this paper.

(1) a pair of 2.5-tesla force-reduced toroids, 30 centimeters in diam-

eter with a winding cross section 2.5 centimeters in diameter; ( 2 ) a

The magnets described include

15-centimeter-bore magnetic bottle with 5-tesla mirrors, a 2.5-tesla

center coil, and a quadrupole Ioffe coil producing a cusp field; and

(4) a 6.25-centimeter-bore, 10-tesla solenoid.

INTRODUCTION

The Lewis Research Center has a need for large-volume, high field-

strength magnets for use in its program on advanced space propulsion and

power concepts, and also for its research in solid-state, plasma, and

low-temperature physics. The development of such magnetic fields has

been in progress for about 9 years. The work in conventional water-

cooled and cryogenically cooled, but not superconducting, magnets has

- - u c e u - dc;,,, ̂- -.".;hoA I u b - grevi m1sly. to 5, These references describe the mag-

nets, built and used at Lewis Research Center, which produce magnetic

fields up to 20 teslas in an 11-centimeter bore and 16 teslas in a

30-centimeter bore. These magnets are now in use in several research

programs.

TM X-52269

2

Enormous amounts of electrical power are required for room-

temperature electromagnets, and the operation time is limited in

cryogenically-cooled, but not superconducting, magnets. The operating

time limitation results because a large supply of cryogen must be avail-

able for cooling the coils if boiling heat transfer is used.

conductive coils, on the other hand, require a liquid-helium environ-

ment. After the coils are charged, the helium boiloff rate is limited

to the heat leaking into the dewar.

small by careful design and construction of the dewar; hence, operation

of the coils is very economical.

Super-

This amount can be made very

The NASA Lewis Research Center has a large and efficient helium

recovery and liquification system(6) with a capacity of 300 liters of

liquid helium per hour.

easily provided for superconductive coils.

Thus, the helium environment can be quite

Studies(7 to '1 have shown that, for such possible space applica-

tions as magnetohydrodynamic power generation and thermonuclear power

generation and propulsion, the only practical solution to the required

magnetic fields is the superconductive magnet.

propulsion and power research laboratory, the Lewis Research Center

i s especially interested in the development of superconductive magnets

The Status of the Lewis Program will be described in this paper.

As NASA's primary space

Force-Free Toroids

The concept of the force-free magnet has been known and used since

3

the early days in the production of high-field magnets. Pioneers such

as Kapitza(') and Cockcroft(l0) used the principle of uniform hydrostatic

pressure throughout the magnet windings. In more recent times, the

concept has been used in various laboratories concerned with the con-

struction of high-intensity fields of large volume for thermonuclear

re search. (11 and 12)

It can be easily demonstrated, of course, that no finite magnet

is truly force free. Nevertheless, a toroid can be wound so that every-

where the conductor is nearly parallel to the field. In this case,

jxB is small, and there is a reduced force on the conductor. + +

For superconductors, however, there is another advantage of winding

the conductor parallel to the local magnetic field. Experiments have

shown that superconductors carry more current when placed in longitu-

dinal fields than when placed in transverse fields.

Cullen (I4) made measurements which show that the critical current is

Sekula(13) and

strongly influenced by the angle that the conductor makes with the

field. These and other investigations showed that the current enhance-

ment can be as much as a factor of 10 or more. In figure 1, The vari-

ation of the critical current with the angle made by the conductor with

the magnetic field is shown f o r two 6il"ier'eiit YJyc I1 szperrnnctuctors.

niobium zirconium and niobium s3annide.

The two NASA force-reduced toroids are wound with niobium-0.25

zirconium, which has the short-sample critical current-field curves

shown in figures Z(a) and (b) . In figure 2 ( b ) , the current is at a

4’

4

maximum when the conductor is placed in a 2.5-tesla field parallel to

the direction of the current.

ate in a self-field of this magnitude to take advantage of this in-

creased performance, an enhancement factor of about 8. Not a11 this

enhancement was realized, however, in the wound coils. The critical

currents of the two coils were 110 and 108 amperes, respectively.

The NASA toroids were designed to oper-

High -Fi eld , Large -Bo re Solenoid As mentioned before, very intense magnetic fields have many

potential applications.

noids.

in a 15-centimeter bore, a solenoid design was specified. To produce a

field of this magnitude in such a volume by conventional, water-cooled

magnets would require many megawatts of power; thus, the advent of

high-current high-field superconductors made such a magnet feasible.

Such fields are most easily produced by sole-

For this reason, when it was decided to produce a 15-tesla field

The solenoid was built for the Lewis Research Center by the Radio

Corporation of America. It was wound with a ribbon conductor consisting

of niobium stannide vapor deposited on a substrate of stainless steel.

At the present, niobium stannide (Nb3Sn) has the highest critical field

(X2.0 T) and the highest critical temperature (18.4%) of any super-

Cnnc?_ir-rt.or, Indeed, extensive research during the past several years

has failed to find a superconductor that approaches niobium stannide

in the magnitude of these two parameters.

its substrate, is exceedingly strong and can be manufactured in a

This material, together with

5

range of sizes to carry a large range of currents.

of a typical ribbon used are presented in’ table I.

Physical properties

The ribbon material is silver plated for stabilization of the

superconductor and for protection of t .e superconductor during handling.

The ribbon can be wound, unwound, and rewound on a small radius

(-1 cm) without damage to the performance of the superconductor. The

short-sample performance is shown in figure 3 where the critical cur-

rent 1, is plotted as a function of the external field. In figure 3(15)

are shown the critical-current magnetic-field data for two types of

Nb3Sn ribbon that were manufactured for specific field ranges.

presents data for exceptionally high-field material that can be used to

wind the portion of a coil which is in the highest field.

Figure 4

Since large hoop stresses are encountered in magnet coils of this

size (15-cm bore), the effect of stress on the superconductor is an

important parameter. Figure 5 shows the effect of stress on copper-

and silver-plated niobium stannide ribbon samples that were oriented

transversely to the magnetic field with the field parallel to the plane

of the ribbon. This relation of current to field is of major concern in

large solenoids, since the innermost turns of the conductor are in

exactly this orfentation. The critical current density as a function

of stress was relatively constant for both types of platings (fig. 5).

The critical current density of the copper-plated ribbon increased

slightly with stress (fig. 5(b)). Generally, the increase was less than

c

4

6

With a l l the samples, the current densi ty decreased sharply, 5 percent.

and the sample became r e s i s t i v e just before t h e breaking point.

The 15-tesla , 15-centimeter-bore solenoid consis ts of 22 modules,

o r subcoils, and energy sinks as shown i n f igure 6. These modules (one

o f which is shown being wound i n f ig . 7 ) were designed t o withstand not

only t h e forces which a re encountered i n t h e i r se l f - f ie ld but a l s o the

forces developed when t h e solenoid i s placed i n ex terna l f ie lds i n t h e

form o f magnetic b o t t l e s used f o r plasma physics experiments.

energy sinks absorb some o f t he energy released when t h e magnet re turns

t o t h e normal state.

where the f i e l d is plot ted as a function o f posit ion.

is designed t o be 5 t o 7 hours, and the current i s supplied by f o u r

100-ampere power supplies.

charge the c o i l s i s qui te long, it is ant ic ipated t h a t the co i l s w i l l be

kept cold and charged somewhere near capacity a t a l l times.

The control c i r c u i t r y provides f o r automatic as w e l l as manual

The

The performance of the magnet i s shown i n f igure 8,

The charging t i m e

Since the time required t o cool down and t o

cont ro l of the charging and discharging rate. Instrumentation i s in-

cluded within the windings t o monitor both the magnetic f i e l d and t h e

temperature and t h e stress on the conductors.

In a.ddition t o the 15-tesla , 15-centimeter-bore solenoid, a 10-tesla ,

6.25-centimeter-bore magnet with be t t e r uniformity w a s b u i l t of niobium

stannide superconductive ribbon; similar t o t h a t described i n t ab le .1 .

This solenoid, however, was designed t o have a homogeneity of 0 .2 per-

7

cent within a 2.54-centimeter-diameter spherical volume and of 0.7 per-

cent within a 5.08-centimeter-diameter spherical volume.

This magnet, which will be used in the research program of Lewis

Research Center, is capable of being charged from zero field to full

field in less than 30 minutes. In the semipersistent mode, the maximum

critical field decay is less than ,500 gauss per hour.

Initially, the magnet will be applied to solid-state physics

research, which will include magnetostriction measurements on copper,

and magnetoresistance of very pure and also very dilute alloys of

alluminum, copper, and bismuth.

Magnetic Bottle

A system of coils whose combined field is shown in figure 9 was

built for plasma containment experiments. The system consists of two

mirror coils that produce 5 teslas each on the axis of the system and

a mirror ratio of 2:l with the central field. The more uniform central

field is provided by inserting a third coil between the two mirror coils,

as shown in figure 9. In addition, a cusp field is produced by a

system of linear conductors (Ioffe bars(16)) equally spaced around the

inner diaiileter sf t h e m i l forms. This scheme produces a minimum B

field which is more effective than the simple mirror field for plasma

containment.

8

A photograph o f the ac tua l c o i l is shown i n figure 10(a) and the

winding scheme o f t he I o f f e c o i l s i s shown i n f igure 10(b) .

co i l s are wound with Nb3Sn (copper clad f o r s t a b i l i t y ) .

qu i te s a t i s f a c t o r i l y and are not ea s i ly driven t o the normal state.

These Ioffe

They operate

The three main c o i l s are wound with a seven-strand Nb-0.25 Zr

cable which is copper plated and dipped i n indium.

s t ab i l i zed mater ia l is easy t o handle and wind in to co i l s . Between

turns , insulat ion i s provided by a Mylar f i l m , and aluminum f o i l pro-

vides an energy sink t o absorb some of the energy when the co i l s a re

driven normal. Monofilament nylon threads (0.05 cm i n diameter) are

wound back and f o r t h across the windings t o provide passages f o r the

l i qu id helium t o penetrate t h e windings. The complete magnet i s shown

i n t h e photograph o f f igure 11.

This quasi-

The performance of t he system o f co i l s is best described by showing

the var ia t ion o f c o i l current as it i s increased with time. Figure 1 2

shows t h a t a mirror c o i l can be charged in less than 2 minutes and t h a t

t h e design f i e l d o f 5 t e s l a s w a s achieved.

To provide f o r improved performance o f t he co i l s , it is planned

t o subs t i t u t e f o r t h e seven-strand NbZr conductor a f u l l y s t ab i l i zed

11~x1 coiidi;&,sr tz prnvirie f o r greater s t a b i l i t y and b e t t e r performance

of t h e coi ls .

-- -.

I n addi t ion t o the magnets previously described, there are i n use

a t t h e L e w i s Research Center several other superconductive magnets.

. 9

These are single solenoids and split pairs, ranging in size from 2 to

20 centimeters in inner diameter and having field strengths from 2.5

to 10.0 teslas. These magnets and magnet pairs are being used in

solid-state, nuclear, and plasma physics research. The research as

well as the facilities have been described in available NASA publi-

cation~'~~ to 18).

cost magnetic-fieid equipaent which has been made possible by the

advent of high-field superconductive materials and the consequent

elimination of the high-cost power supply and cooling water supply

required by conventional copper magnets.

These magnets are all examples of the type of l o w -

REFERENCES

(1) J. C. Laurence, G. V. Brown, J. Geist, and K. Zeitz, in: High

Magnetic Fields, M.I.T. Press, Cambridge and john Wiley and Sons,

Inc., New York (1962), p. 170.

(2) J. C. Laurence and W. D. Coles, in: Proceedings of the International

Symposium on Magnet Technology, Stanford Linear Accelerator Center,

Stanford University, Stanford (1965), p. 574.

(3) J. C. Laurence and W. D. Coles, in: Advances in Cryogenic

Engineering, Vol. 11, Plenum Press, New York (1966), p. 643.

{4) TG. 2. C o k s , 2 ; C. Fakan, and J. C. Laurence, "Superconducting

Magnetic Bottle for Plasma Physics Experiments," NASA Tech.

Note TN D-3595 (August 1966).

. 10

(5) J. C. Fakan, in: High Magnetic Fields, M.I.T. Press, Cambridge

and John Wiley and Sons, Inc., New York (1962), p. 211.

(6) H. Mark, R. D. Sommers, and M. J. Mirtich, AIAA J. 4(10): 1811

(October 1966).

(7) G. W. Englert, in: 3rd Symposium on Engineering Aspects of

Magnetohydrodynamics, Gordon and Breach Science Publishers, Inc.,

New York (1964), p. 645.

R. R. Akers, W. A. Stewart, T. C. TUSU, R. E. Kothman, and 0. T.

Beecher, "Superconducting Magnet Research and Application MHD

Power Generators," Westinghouse Electric Corp., Air Force Report

RTD-TDR-63-4170, AD-426776 (December 31, 1963).

(9) P. Kapitza, Roy. SOC. Proc. 115, 658 (August 1927).

(8)

(10) J. D. Cockcroft, Roy. SOC. Phil. Trans. 227, 317 (1928).

(11) D. R. Wells and R. G. Mills, in: High Magnetic Fields, M.I.T.

Press, Cambridge and John Wiley and Sons, Inc., New York

(1962), p. 44.

(11) H. P. Furth, M. A. Levine, and R. W. Waniek, Rev. Sci. Instr.

28( 11) : 949 (Novenber 1957).

G. V. Brown, in: High Magnetic Fields, M.I.T. Press, Cambridge

and John Wiley and Sons, Inc., New York (1962), p. 370.

(13)

(14) S. T. Sekual, R. W. Boom, and C. J. Bergeron, Appl. Yhys. LeLi5ei-a

2(5) : 102 (March 1, 1963) -

11

(15) G. W. Cullen, G. D. Cody, and J. P. McEvoy, Jr., F'hys. Rev. 132(2):

577 (October 15, 1963).

(16) E. R. Schrader and P. A. Thompson, IEEE Trans. on Magnetics

MAG-2( 3) : 311 (September 1966).

(17) W. R. Hudson, "Effect of Tensile Stress on Current-Carrying

Capacity of Commercial Superconductors," NASA Tech. Note TN D-3745

(November 1966).

(18) Yu. V. Gott, M. S. Ioffe, and V. G. Telkovsky, Nucl. Fusion,

Suppl. Z(3): 1045 (1962).

(19) J. R. Roth, "Nonadiabatic Particle Losses in Axisymmetric and

Multipolar Magnetic Fileds," NASA Tech. Note TN D-3164 (December

1965), p. 40.

1 2

.

TABLE I. - PHYSICAL FROPERTIES OF NIOBIUM

STANNIDE STAINLESS STEEL RIBBON

Property I Width, cm Thickness, cm Length, km Current density, A/cm2 Tensile strength, N/m2

Stainless s t e e l substrate

0.23 0.008

88

16x108 - - -

Niobium stannide superconductor

0.23 0.001

88 106 16x108 (without damage t o super- conductor )

300

200

100

0

U - 0 -u 1.1

0

i 1.0

Y -... I U

Y

c m L L

3 U

U c L

. 9 ._ ._ u

. a

. 7

.6

. 5

. 4

.3

.2

.1

0

( a ) Niobium zirconium superconductor.

Maqnetic field, H, T

ref. 13

0 0 0

I K -=- I(0") H s in 8 + K

I - \

0 0 .

0 0 Y - 0

I 10 20 30 40 50 60 70 80 90

Angle, 8, deg

(b ) Niobium stannide superconductor.

Figure 1. - Crit ical current as function of angle made by conductor with magnetic field. Sample, 0.025-centimeter-diameter niobium - 0.25 zirconium wire, magnetic field, 3teslas; temperature, 4.2" K (ref. 19).

, a IC IC M I W

In N

0

.

- \ 800 -

- L L

2 100- 80-

60 -

- -

- 40- -

20 I 0 1 2 4 6 8 10 12 14 16

Total field, T

Figure 3. - Cr i t ica l -current magnetic-field characterist ics of n iob ium stannide r ibbon (ref. 17).

Figure 4. - Ranges of application of minimum-stabi l ized superconductive r ibbon (ref. 17).

(D I IC

IC N-I

I w 4

3 N E

s .= a.

a

P 2 -

I- Current

c c E " I L 3

Current

Force Force -?- (a1 Silver plated.

Sample 0 4 0 5 0 6

I Magnetic field

0 2 4 6 8 10 12 14 16 18 Stress, Nlm2

(bl Copper plated.

Figure 5. - Effect of stress on current density of niobium stannide ribbon oriented with magnetic field perpendicular to current i n plane of ribbon.

C-67-391 F igure 6. - Components of 15-tesla magnet.

T --

F i g u r e 7. - hlodule of a 15- tes la magnet being wound.

.

0 Frn 6.52 crn 13.0 cm

21.8 cm 3 . M 5.41

1.24 18.4 cn l

4.55 4.67 5.01 5.94 8.07

7.43 7.69

10.30 10.21 9.94 9.61 9.38

c) f 14.9 cm - -

c .I 11.70 10.47

1 3 : a 13.24 12.56 11.46 10.09 11.4 c m - - - -

6.43

6.86

7.23

f /

.A .r I I 12.35 I 7.52 1 - I . - . J/ 4 8.Ocm

15.91 15.64 14.86 13.42 10.84 7.73

15.51 15.23 14.37 12.81 10.43 7.83

15.23 14.93 14.02 12.43 io. 23 7.9C

15.06 14.75 13.81 12.23 10.13 7.92

15.00 14.69 13.75 12.17 10.97 7.92

- w t - r - H /

e , t , W ~ * ~

U H C , H C , H

4 -------------

Figure 8. - NASA 15-centirneter-bore, 15-tesla (1% kG) magnet.

0

Relative distance f rom center of coi l B

Figure 9. - Magnetic mir ror configuration for plasma physics experiments.

.

C-65-1617 ( a ) loffe coils,

(b) Schematic of current path in coil.

Figure 10. - loffe bars (quadrupole configuration).

Figure 11. - Magnet ic bottle,

49.5 kG

I I I I I I I I ,

0 50 103 150 200 250 3M1 Time, sec

Figure 12. - Charging history, coi l A alone.

NASA-CLEVELAND, OHIO E-3776