PSFC/JA-17-54

A Field-Shaking System to Eliminate the Screening Current-Induced Field in the 800-MHz HTS Insert of the MIT 1.3-GHz LTS/HTS NMR Magnet:

A Small-Model Study

Jiho Lee, Dongkeun Park, Philip C. Michael, So Noguchi, Juan Bascuñán, and Yukikazu Iwasa

August 2017

Plasma Science and Fusion Center Massachusetts Institute of Technology

Cambridge MA 02139 USA This work was supported by the National Institute of Biomedical Imaging and Bioengineering and the National Institute of General Medical Sciences. Reproduction, translation, publication, use and disposal, in whole or in part, by or for the United States government is permitted.

Mon-Af-Po1.03-03

Template version 8.0, 27 July 2017. IEEE will put copyright information in this area

See http://www.ieee.org/publications_standards/publications/rights/indes.html for more information.

1

A Field-Shaking System to Eliminate the Screening Current-Induced Field in the 800-MHz HTS Insert of the MIT 1.3-GHz LTS/HTS NMR Magnet: A Small-

Model Study

Jiho Lee, Dongkeun Park, Philip C. Michael, So Noguchi, Juan Bascuñán, and Yukikazu Iwasa

Abstract— In this paper, we present results, experimental and

analytical, of a small-model study, from which we plan to develop

and apply a full-scale field-shaking system to minimize or even eliminate the screening current-induced field (SCF) in the 800-MHz HTS Insert (H800) of the MIT 1.3-GHz LTS/HTS NMR

magnet (1.3G) currently under construction—the H800 is com-posed of 3 nested coils, each a stack of no-insulation (NI) REBCO double-pancakes. In 1.3G, H800 is the chief source of a large error

field generated by its own SCF. To study the effectiveness of the field-shaking technique, we use two NI REBCO double-pancakes, one from 2 coils (HCoil2 and HCoil3) of the 3 H800 coils, and place

them in the bore of a 5-T/300-mm room-temperature bore external magnet. The external magnet is used not only to induce SCF in the double-pancakes but also eliminate it by the field-shaking. For

each run, we induce SCF in the double-pancake at an axial location where the external radial field Br > 0, then for the field-shaking, move to another location where the external axial field Bz >> BR.

To examine if other SCF eliminating techniques, e.g., the current-sweep-reversal (CSR) method, is applicable to H800 even when L500 and H800 are series-connected, we perform similar se-

quences of test for other combinations of the double-pancake axial locations. In this paper, we report 77-K experimental results, de-velop an analysis that satisfactorily explains the results, and apply

the analysis to design a field-shaking system for 1.3G at full oper-ation.

Index Terms—High field magnet, HTS insert, REBCO, SCF, Screening current

I. INTRODUCTION

OR the development of >1 GHz NMR, with LTS outsert

coil, HTS insert coil must be used owing to its large in-field

current-carrying capacities [1]. For a high-resolution NMR, its

magnetic field must be uniform with an error field of ~0.01 ppm

over a sample volume, making field shimming a must in the

NMR magnet. The screening current-induced field (SCF), a di-

amagnetic field generated by each turn of HTS coil, is major

field error to incorporate an HTS insert for a high field

LTS/HTS magnet. The magnitude of diamagnetic field, SCF is

This work was supported by the National Institute of Biomedical Imaging

and Bioengineering and the National Institute of General Medical Sciences.

J. Lee, D. Park, P. C. Michael, J. Bascunan, and Y. Iwasa are with the Magnet Technology Division, Francis Bitter Magnet Laboratory, Plasma

Science and Fusion Center, Massachusetts Institute of Technology, Cambridge,

proportional to the superconductor size and critical current den-

sity [2]. Various studies have reported the screening current-

induced field (SCF) generated by HTS coils [3-5].

Although much less serious with LTS magnets than with

HTS magnets, the so-called field-shaking technique to mini-

mize SCF error fields was proposed in 1986 for LTS magnets

[6-8]. This technique has since been demonstrated, theoretically

and experimentally to be applicable to HTS magnet [9-15].

Since SCF by an HTS insert can be >100 times greater than

those typical by an LTS outsert, it is critical to minimize or even

eliminate the SCF-generated error field.

Francis Bitter Magnet Laboratory (FBML) of Massachusetts

Institute of Technology (MIT) has developed a high-resolution

1.3-GHz/54-mm LTS/HTS NMR magnet (1.3G). 1.3G consists

of the 800-MHz HTS insert (H800) composed of 3 nested coils,

each a stack of no-insulation (NI) REBCO double pancakes,

and the 500-MHz LTS outsert (L500). In 1.3G, H800 is the

chief source of a large error field generated by its own SCF.

Although our earlier field-shaking works [9-11] has been useful

to us, an experiment in this paper help us design a field-shaking

system most suitable for our 1.3G. Unlike our previous field--

shaking experiments in which we used either a stack of 3

REBCO 94-mm OD NI double-pancakes [10] or a pile of

12mm×12mm square NI REBCO strips [11], we use 2 NI

REBCO DPs, one from each of two H800 coils (Coils 2–3) for

our new experiment.

By a 5-T/300-mm room-temperature bore external magnet,

we induce the screening current at DP coils. 5-T external mag-

net also eliminates the SCF generated by the screening current

by the trapezoidal field-shaking patterns, confirmed as effective

for elimination of SCF. Considering the geometry of the

LTS/HTS combination and the magnetic field within the HTS

tape, the largest SCF error field occurs in the top and bottom

end regions of a magnet, where the radial fields, applied- and

self-field, are greatest. Therefore, we induce SCF in the pres-

ence of the maximum external radial field and eliminate it by

MA 02139 USA (email: jiho@mit.edu, dk_park@mit.edu,

michael@psfc.mit.edu, bascunan@mit.edu, iwasa@jokaku.mit.edu).

S. Noguchi is with the Graduate School of Information Science and Technol-ogy, Hokkaido University, Sapporo 060-0814, Japan (email: noguchi@ssi.ist.ho-kudai.ac.jp).

F

2

the field-shaking in absence or presence of the external radial

field.

II. TEST COILS AND EXPERIMENTS PROCEDURE

In this study, for SCF induction and field-shaking tests, we

used 2 NI REBCO DP coils, one from each of 2 H800 coils

(Coils 2–3). Specifications of 2 REBCO DPs are shown in Ta-

ble I.

The procedures to induce the screening current and eliminate

the SCF caused by the screening current are as below:

1. Place the NI REBCO DP, at room temperature, at an ax-

ial location, 𝑧 = −175 𝑚𝑚, below the mid-plane of the

external magnet.

2. Apply the magnetic field of 5 T to the DP set. Each coil

experiences the external radial field, the perpendicular

field to the tape surface, at this location.

3. Turn the DP set into the superconducting state, by cool-

ing in a bath of liquid nitrogen.

4. Reduce the magnetic field to zero, the process of which

induces SCF in the double-pancake.

5. Relocate the double-pancake to the mid-plane of the ex-

ternal magnet.

6. Apply a shaking field with the external magnet using the

trapezoidal field injection, confirmed as effective for

elimination of SCF.

7. Map the radial field along the z-axis at each double-pan-

cake’s innermost turn

The magnetic field near the ends of the H800 contains signif-

icant radial field component. To better simulate conditions

throughout the H800, we examine the effect of field-shaking on

SCF in the absence or presence of radial field. This can be

achieved by testing each NI REBCO DP at selected position

above or below the mid-plane of 5-T external magnet. From the

geometry of the coil winding of 5-T external magnet, that se-

lected position which has the maximum radial field, 2 NI

REBCO DPs experience, is 𝑧 = ±175 𝑚𝑚.

III. INDUCTION OF SCF AND FIELD-SHAKING TEST

A. Induction of SCF and field-shaking at mid-plane

Following the above noted procedure, we induced the screen-

ing current of HCoil2 DP at 𝑧 = −175 𝑚𝑚 and moved the

double-pancake to 𝑧 = 0 𝑚𝑚 to eliminate its SCF by field-

shaking without the radial field. Fig. 2 shows the magnetic flux

density at magnet center of the external magnet to induce SCF

and perform the field-shaking. The measurement of magnetic

flux density distribution along the z-axis were repeated after

every field-shaking experiment. Fig. 3 shows the measured ra-

dial field, BR at the innermost turn of HCoil2 DP before and

after field-shaking. From the first attempt of field-shaking of

0.6 T based on the magnet flux density at magnet center, the

considerable SCF was decreased to 20% of its originally in-

duced one. However, after then, even with the multiple at-

tempts, same field-shakings of 0.6 T didn’t show the notable

effect on the elimination of SCF.

Fig. 4 shows the measured BR at the innermost turn of HCoil3

DP before and after field-shakings. From five field-shaking of

0.6 T lowered SCF to 55-50% of its originally induced one. And

then four more field-shaking of 0.8 T and 1.2 T lowered the

SCF to 44-48% of the original magnitude. To eliminate SCF to

~20% of the originally induced one same as the HCoil2 DP test

result, three field-shaking of 1.6 T was needed.

Fig. 1. Position of 5-T/300-mm external magnet (black), HCoil2 DP (blue),

and HCoil3 (red)

TABLE I

SPECIFICATIONS OF TESTED HTS DOUBLE PANCAKE COILS

Parameters HCoil2 DP HCoil3 DP

Conductor (REBCO) Width; thickness [mm; µm] 6.02; 76 6.04; 75 Cu stabilizer thickness [mm] 0.01 0.01 Ic @ 77 K [A] 188 190 NI Double pancake Coil ID [mm] 151.04 196.90 OD [mm] 169.18 211.50 Height [mm] 12.198 12.200 Turn per pancake 120 96 Self-inductance [mH] 18.06 16.60 Characteristic resistance, Rc [µΩ]

322.1 491.9

Time constant, τ [s] 56.07 33.75 Ic @ 77 K, self-field [A] 42.81 67.41

TABLE II

SPECIFICATIONS OF 5-T/300-MM ROOM-TEMPERATURE BORE EXTERNAL

MAGNET

Parameters Values

Overall i.d.; o.d.; height [mm] 327; 415; 338 Magnet constant [mT/A] 61.52

Center field at Iop of 81.268 A [T] 5.0

Self-inductance [H] 143

3

B. Induction of SCF at mid-plane and field-shaking at end-

plane

To simulate the conditions of the end-plane of H800 which

has the maximum external radial field, another field-shaking

test about HCoil2 DP was performed without changing its axial

location. When the magnetic flux density of the external magnet

is 0.98 T, the axial and radial magnetic flux densities are 0.6

and 0.14 T, respectively. To compare the effect of the pres-

ence/absence of the radial field on field-shaking, shaking field

of 0.6 T with radially shaking field of 0.14 T was injected to

HCoil2 DP. Fig. 5 shows the test result of the field-shaking with

the radial field. With the radial field of 0.14 T, 0.6 T field-shak-

ing didn’t show the effective reduction/elimination of SCF

comparable than the test result shown in Fig. 3 which is the re-

sults of field-shaking without the radial field.

Fig. 2. ”Time-versus-external magnetic flux density” profile of 5-T/300-mm

external magnet to induce the screening current and to eliminate its SCF

through field-shaking with trapezoidal waveform, for HCoil2DP test

Fig. 3. Measured radial magnetic flux density, BR, at innermost turn region of HCoil2 DP along the z-axis when coil position during field-shaking was 0

(𝑧 = 0 𝑚𝑚)

Fig. 4. Measured radial magnetic flux density, BR, at innermost turn region

of HCoil3 DP along the z-axis when coil position during field-shaking was 0

(𝑧 = 0 𝑚𝑚)

Fig. 5. Measured radial magnetic flux density, BR, at innermost turn region of HCoil2 DP along the z-axis when coil position during field-shaking was -

175 (𝑧 = 175 𝑚𝑚)

Fig. 6. Normalized value of the line integral ∫|𝐵𝑠𝑐𝑓| 𝑑𝑧 according to the

field-shaking iteration

4

C. Test summary

To quantify and compare the effect of field-shaking for the

elimination of SCF, we calculated the line integral of the meas-

ured magnetic flux density along z-axis, ∫|𝐵𝑠𝑐𝑓| 𝑑𝑧 which has

the unit of Wb/m and compare their normalization values. Fig.

6 shows the normalization value of that line integral according

to the field-shaking iterations. Three tests successfully elimi-

nated the SCF to approximately 20% of their initially induced

values. Once the degradation of SCF shows the tendency of sat-

uration, unless the shaking-field was increased, the repetition of

the field-shaking didn’t make the elimination of SCF more. The

field-shaking of 0.6 T with the radial field of 0.14 T which

shows 9% reduction (100→89%) makes the field-shaking of 0.6

T which shows 80% reduction (100→20%) significantly less

effective.

During the induction of SCF to each NI double-pancake and

the field-shaking, the electromotive force developed by the ex-

ternal magnet caused the azimuthally induced current due to its

NI winding technique which makes the closed loop inside the

coil. Electromotive force is proportional to the change rate of

the external magnetic field, which is proportional to the ramp

rate of the power supply current for the external magnet. During

the induction of SCF for HCoil3 DP, the trace of the coil voltage

shows superconducting-to-normal transition. HCoil3 DP which

has the bigger inner/outer radius than HCoil2 DP should expe-

rience the higher radial field. Therefore, its critical current at 3-

4 T could be high enough to make the superconducting-to-nor-

mal transition. Most of NI HTS double-pancakes are used for

high-field magnets themselves or their insert to elevate the

magnetic flux density at magnet center. Therefore, the field-

shaking could make the operation of HTS coils closer to the

critical condition.

IV. CONCLUSION

The measurement of SCF and the effect on the field-shaking

for its elimination was studied experimentally. Two NI REBCO

double-pancakes, one from each of two H800 coils (Coils 2–3)

were tested. Screening current and SCF were induced at axial

location which experiences the largest radial field. The field-

shaking tests were performed at various amplitudes of shaking-

field and the axial location. Based on the experimental results

and analyses to date, we may conclude that:

Field-shaking method was successfully applied to elimi-

nate the considerable amount of SCF in a couple of at-

tempts. The higher shaking field is more effective on the

elimination of SCF.

Presence of the radial field during field-shaking made

field-shaking significantly less effective than absence of

one. Though with the same z-axial shaking field of 0.6 T,

the presence of the radial field, 0.14 T makes the field-

shaking almost ineffective.

Double-pancakes applying NI winding have the azimuth-

ally induced current due to the electromotive force which

is proportional to the ramping-rate of field-shaking magnet.

In case of operating condition, this azimuthally induced

current could be helpful to eliminate SCF similar to the cur-

rent sweep reversal (CSR) method or make the HTS coils

closer to their operating current margin.

REFERENCES

[1] Yukikazu Iwasa, Juan Bascuñán. Seungyong Hahn, John Voccio, Youngjae Kim, Thibault Lécrevisse, Jungbin Song, and Kazuhiro

Kajikawa, “A High-Resolution 1.3-GHz/54-mm LTS/HTS NMR

Magnet,” IEEE Trans. Appl. Supercond., vol. 25, no. 3, pp. 1–5, 2015. [2] Y. Yanagisawa, H. Nakagome, Y. Koyama, R. Hu, T. Takao, M. Hamada,

T. Kiyoshi, M. Takahashi, and H. Maeda, “Effect of current sweep

reversal on the magnetic field stability for a Bi-2223 superconducting solenoid,” Phys. C Supercond. its Appl., vol. 469, no. 22, pp. 1996–1999,

2009. [3] Seungyong Hahn, Juan Bascuñán, Woo Seok Kim, Emanuel S. Bobrov,

Haigun Lee, and Yukikazu Iwasa, “Field mapping, NMR lineshape, and

screening currents induced field analyses for homogeneity improvement in LTS/HTS NMR magnets,” in IEEE Transactions on Applied

Superconductivity, 2008, vol. 18, no. 2, pp. 856–859.

[4] Chen Gu, Timing Qu, and Zhenghe Han, “Measurement and calculation of residual magnetic field in a Bi2223/Ag magnet,” IEEE Trans. Appl.

Supercond., vol. 17, no. 2, pp. 2394–2397, 2007.

[5] Y. Koyama, T. Takao, Y. Yanagisawa, H. Nakagome, M. Hamada, T. Kiyoshi, M. Takahashi, and H. Maeda, “Towards beyond 1 GHz NMR:

Mechanism of the long-term drift of screening current-induced magnetic

field in a Bi-2223 coil,” Phys. C Supercond. its Appl., vol. 469, no. 13, pp. 694–701, 2009.

[6] Kazuo Funaki and Kaoru Yamafuji, “Abnormal transverse-field effects in

nonideal type ii superconductors i. A linear array of monofilamentary wires,” Jpn. J. Appl. Phys., vol. 21, no. 2 R, pp. 299–304, 1982.

[7] Kazuo Funaki, Teruhide Nidome and Kaoru Yamafuji, “Abnormal

transverse-field effects in nonideal type II superconductor II. influence of dimension ratios in a superconducting ribbon,” Japanese J. Appl. Phys.,

vol. 21, no. 2 R, pp. 1121-1126, 1982.

[8] Kazuo Funaki, Minoru Noda and Kaoru Yamafuji, “Abnormal transverse-field effects in nonideal type II superconductor III. a theory for an AC-

induced decrease in the semi-quasistatic magnetization parallel to a DC

bias field,” Japanese J. Appl. Phys., vol. 21, no. 2 R, pp. 1580-1587, 1982. [9] Kazuhiro Kajikawa, Gwendolyn V. Gettliffe, Yong Chu, Daisuke Miyagi,

Thibault P. Lécrevisse, Seungyong Hahn, Juan Bascuñán, and Yukikazu

Iwasa, “Designs and Tests of Shaking Coils to Reduce Screening Currents Induced in HTS Insert Coils for NMR Magnet,” IEEE Trans. Appl. Su-

percond., vol. 25, no. 3, p. 4300305, 2015.

[10] Daisuke Miyagi, Seungyong Hahn, Thibault L´ecrevisse, Min Cheol Ahn Jung-Bin Song, Juan Bascu˜n´an, and Yukikazu Iwasa, “An experimental

study on ‘field-shaking’ technique to reduce screening-current fields in a

noninsulated REBCO magnet of double-pancake coils,” Presented at the 2016 Appl. Supercond. Conf., September 2016

Fig. 7. Terminal voltage of HCoil2 DP and the power supply current of

external magnet during the 1st field-shaking at z=0

5

[11] Fei Liang, Timing Qu, Zhenyu Zhang, Jie Sheng, Weijia Yuan, Yukikazu

Iwasa and Min Zhang, “Vortex shaking study of REBCO tape with con-sideration of anisotropic characteristics,” Supercond. Sci. Technol., 2017.

[12] E. H. Brandt and G. P. Mikitik, “Why an ac magnetic field shifts the

irreversibility line in type-II superconductors.,” Phys. Rev. Lett., vol. 89, no. 2, p. 27002, 2002.

[13] G. Mikitik and E. Brandt, “Theory of the longitudinal vortex-shaking

effect in superconducting strips,” Phys. Rev. B, vol. 67, no. 10, pp. 1–8, 2003.

[14] E. H. Brandt and G. P. Mikitik, “The evaluation of E–J characteristics of

Bi-2223 silver-sheathed tape in a wide range of electric field,” Supercond. Sci. Technol., vol. 17, 2004.

[15] K. Kajikawa and Y. Okabe, “Reduction of Screening-Current-Induced

Fields in an HTS Tape Winding Using Toroidal Arrangement of Shaking Coil,” IEEE Trans. Appl. Supercond., vol. 26, no. 4, pp. 1–4, 2016.