Recent progress on SuNAM’s coated conductor development ...

Recent progress on SuNAM’s coated conductor

development; performance, price & utilizing ways

Superconductor, Nano & Advanced Materials

Seung Hyun Moon, Jae-Hoon Lee & Hunju Lee

SuNAM Co., Ltd.

2016. 09. 13.

CCA 2016, Aspen, Colorado, US.

IEEE/CSC & ESAS EUROPEAN SUPERCONDUCTIVITY NEWS FORUM (global edition), October 2016. Presentation IO-16 given at CCA 2016; Aspen, Colorado,USA, September 11 – 14, 2016. 1

SuNAM’s coated conductor; architecture, characteristic. Quality control tools for uniformity and yield

Higher Je : Thicker S.C. layer 1.6 m, >1,000 A/12 mm.

MCI(Metal Clad Insulation) 2G wire for high field magnet.

Solution for charging time delay problem in NI(No-Insulation) coil.

Higher Je : metal substrate removal process.

Summary

Contents


SuNAM’s Coated Conductor



스티커 노트

HTS wire requirements in performance and price for various application devices were being studied at least for 10 years.


스티커 노트

Unfortunately, performance and price goals have not been met until recently.

Applications of Superconductivity

Can carry extremely large current without loss. Can generated extremely large magnetic field. High energy efficiency with compact volume & mass.

Current carry

HTS 2G Wire

NMR

MRI Crystal growth

Magnetic seperator

Motor

MagLev

FCL Motor & Generator Transformer

Cable

Fusion

Switcing

Wind

Generator

Biz Chance

in near(?) future


스티커 노트

Electric power devices such as superconducting cables and fault current limiters are beginning to be deployed on commercial basis, and coil-based applications such as rotating machines and MRI/NMR are thought to be the next candidate.

Throughput : growth rate & large deposition area

Yield : process margin & (in-line) Quality Control

Robustness : shelf life, stability (mechanical,

thermal cycling, thermal expansion…)

Customer friendly : joints, easy to use…

In-line production, automation…

For reasonable size market creation,

Target price ($/kA-m) : 50, 25, or less?

Availability : ~ 1,000 km/yr or /month or ??

RCE DR : ~ 100 nm/sec or faster (SuNAM) The highest throughput process

RCE-DR process : easy to scale-up to wide strip.

How can we realize practical HTS 2G wire?


스티커 노트

Manufacturing cost of 2G wire surely depends on production/sales volume, but ultimately, among the variety of manufacturing processes currently employed, high throughput process should dominate. Reactive Co-evaporation enables high throughput production due to high deposition rate and scalability to wider tapes.

Structure

- Typical Ic ~ > 700A/12mmW at 77K Self-field (Jc ~ >5 MA/cm2)


스티커 노트

Materials, deposition methods, thicknesses, and functions of each layer constituting 2G wire.

Scale up Issues: IBAD & in-Situ High Rate E-Beam

Robert H. Hammond (Stanford Univ.)

New Ideas, Directions?

High rate, large area, high IC and low cost of

materials processes will eventually be required –

not immediately but in 10 years.

High rate may require growth in liquid flux.

Cost Example


스티커 노트

Among the many factors influencing price performance ratio($/kAm), high deposition rate can be achieved with the help of liquid flux which exists at a certain area in the phase diagram.

SuNAM RCE-DR process

RCE-DR : Reactive Co-Evaporation

by Deposition & Reaction (SuNAM,

R2R)

High rate co-evaporation at low

temperature & pressure to the target

thickness(> 1 m) at once in

deposition zone (6 ~ 10nm/s)

Fast (<< 30 sec. ) conversion from

amorphous glassy phase to

superconducting phase at high

temperature and oxygen pressure in

reaction zone

Simple, higher deposition rate & area,

low system cost

Easy to scale up :single path SuNAM DR path

(single time)

Conventional CDR path (repeat more than >>1,000

times)


스티커 노트

Lower right graph depicts the path taken by amorphous precursor of Gd2O3, BaO, CuOx mixture in the T-PO2 phase diagram, and upper right schematic shows how we realize the path.

Superconductor, Nano & Advanced Materials

500 nm

Gd

Ba

Cu

O

Cu

Cu

Cu

Cu Gd

Ba

Cu

O

O

Cu Cu

Gd

Ba O

O

L

Gd2O3

GdBCO

Gd2O3

Cu-O

• Very low PO2 zone (~ 10-5 Torr): Amorphous Film

• Lower PO2 zone (~30 mTorr): Gd2O3 + Liquid (< 5 sec)

• Higher PO2 zone (~100 mTorr): GdBCO Film (< 20 sec)

GdBCO growth mechanism: a seeded melt-textured growth!!!

Growth mechanism of the GdBCO film by RCE-DR


스티커 노트

You can see how the Gd2O3 seed in the BaCuO liquid nucleates GdBaCuO (right) as the precursor sweeps the path in the phase diagram (left) (I’m sorry that PPT animation cannot be seen in PDF format.)

Daily Production 2G wire performances

0 100 200 300 400 500 600 7000

100

200

300

400

500

600

700

800

900

1000

Ic (

A / 1

2 m

m )

Length ( m )

0 100 200 300 400 500 600 7000

100

200

300

400

500

600

700

800

900

1000

Ic (

A / 1

2 m

m )

Length ( m )

0 100 200 300 400 500 600 7000

100

200

300

400

500

600

700

800

900

1000

Ic (

A / 1

2 m

m )

Length ( m )

0 100 200 300 400 500 6000

100

200

300

400

500

600

700

800

900

1000

Ic (

A / 1

2 m

m )

Length ( m )

0 100 200 300 400 500 600 7000

100

200

300

400

500

600

700

800

900

1000

Ic (

A / 1

2 m

m )

Length ( m )

0 100 200 300 400 500 600 7000

100

200

300

400

500

600

700

800

900

1000

Ic (

A / 1

2 m

m )

Length ( m )0 100 200 300 400 500 600 700

0

100

200

300

400

500

600

700

800

900

1000

Ic (

A / 1

2 m

m )

Length ( m )

0 100 200 300 400 500 600 7000

100

200

300

400

500

600

700

800

900

1000

Ic (

A / 1

2 m

m )

Length ( m )

0 100 200 300 400 500 600 7000

100

200

300

400

500

600

700

800

900

1000

Ic (

A / 1

2 m

m )

Length ( m )

Length

(m) IC IC,max IC,min 1 sigma

739 820 857 761 1.7

Length


700 800 865 615 3.4

Length


682 801 860 654 3.2

Length


734 810 856 713 2.2

Length


728 820 866 729 2.4

Length


710 804 859 716 2.5

Length


718 800 869 707 3.6

Length


743 803 840 651 1.8

Length


743 813 847 768 1.6

( ~ 6 hrs deposition time (120 m/hr))

IEEE/CSC & ESAS EUROPEAN SUPERCONDUCTIVITY NEWS FORUM (global edition), October 2016. Presentation IO-16 given at CCA 2016; Aspen,,Colorado,USA, September 11 – 14, 2016. 12

스티커 노트

We typically produce ~700 meters of wire in a single run with the speed of ~ 120 m/hr (12 mm wide tape). Ic variations along the lengths are shown for 9 runs.

RCE-DR Results on Stainless Steel Substrate

Min Ic (A/cm-width) x L (m) > 0.6 Million A-m

Production speed of 120 m/hr (12 mm width)

( 1 km for ~ 8 hrs)

Particle IBAD

defect

s

Compositio

n


스티커 노트

Some point defects, the origin of which we can trace, and rather slow composition variations are not totally under control upto now.

<QCM Feedback program>

<RHEED spot Feedback program>

Control computer

An appropriate feedback algorithm can keep the shape of the RHEED spot in the

specific range, while QCM monitoring to adjust the e-gun power.

Before optimization After optimization

Quality Control : RHEED Vision System


스티커 노트

In-line quality control is very important for producing high quality wire with minimal number of defects and/or non-uniformity. This page describes how we maintain the crystallinity of IBAD-MgO with RHEED pattern as a feedback input.

14

67 68 69 70 71 72 73 74 75 76 77

2.8

3.0

3.2

3.4

3.6

3.8

4.0

(

deg

ree)

Rheed Spot Angle (degree)

6.0

6.2

6.4

6.6

6.8

7.0

(d

eg

ree)

Feedback route based on RHEED spot analysis

(110) spot

Because of different evolution of & R,

optimization is very important for high quality 2G

wire.

Intensity & tilt angle of MgO (110) spot is one of the

most important parameter.


스티커 노트

In-plane and out-of-plane crystallinity (Δϕ and Δω) behave differently with respect to RHEED spot angle, forcing us to compromise.

Quality Control : RCE Vision Inspection System

Based on color dependence of composition DB, optimum composition level is automatically controlled by PC. (Slow feedback)


스티커 노트

With the knowledge that the surface color (RGB) of the SC layer has a strong correlation with composition (RE:Ba:Cu), we set up another feedback system for the SC layer.

[Start] [End]

Start color

End color

RCE Vision System will be introduced for increasing the uniformity of composition in

RCE-DR process. The control computer takes (RGB) values in three-dimensional

vector space which is transformed from the color of the tape surface.

Control

the power

Color detection

Is the (RGB) vector

in the range?

Yes

No

Quality Control : RCE Vision Inspection System

(Composition DB)


스티커 노트

The feedback system reads RGB of surface, looks up DB to figure out the composition, and then adjust the deposition rate of each metal accordingly, until the desired composition/color is restored.

Higher Je : Thicker S.C. layer


18 Normal RCE-DR process : before optimization

1.2 1.4 1.6 1.8 2.0 2.20

100

200

300

400

500

600

700

800

I C (

A/1

2 m

m)

Thickness (m)

IC

0

1

2

3

4

5

6

JC

JC (M

A/c

m2)

Thickness dependence of Ic and surface color for GdBCO

1 2 3 4 5

Thickness 1.3㎛ 1.6㎛ 1.8㎛ 2.0㎛ 2.2㎛

Surface color for GdBCO

Ic 750A/12mm 600A/12mm 300A/12mm 100A/12mm 0A/12mm

TEM analysis

→ 1.3 ㎛-thickness:

Gd2O3 are randomly distributed

→ 1.9 ㎛-thickness:

Gd2O3 are distributed the boundary of the layers

1

2 3

As increasing the thickness, Jc and Ic are decreased. All the samples were prepared by same process speed.


스티커 노트

When the deposition rate(=thickness of superconducting layer) is increased, both Ic and Jc are sharply reduced

Optimization of deposition region for making thick GdBCO films

For uniformity, 1. Decrease deposition region from 55 cm to

45 cm. 2. Increase distance between source and

substrate. 3. Increasing turns of deposition region (14

turns 19 turns)

14 19

As increasing the thickness, Jc is decreased

As increasing the thickness, Jc is not decreased

1.6 m-thick

All the samples were prepared by same process speed.


스티커 노트

We found the deposition region in which all the compositions (RE, Ba, Cu) can be uniformly deposited and increased the number of turns in order to increase the thickness of the superconducting layer.

0 1 2 3 4 5 6 7 8 9 100.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

Gd Cu

Co

mp

os

itio

n r

ati

o

Gd

Ba

Cu

Position

Ba

7th turn

0 1 2 3 4 5 6 7 8 9 100.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

Gd Cu

Co

mp

os

itio

n r

ati

o

Gd

Ba

Cu

Position

Ba

7th turn

Optimization of Deposition region

Distance between source and substrate : ~ Distance between source and substrate : ~


스티커 노트

In addition, the metal sources (RE, Ba, Cu) are put further away from the substrate to increase the deposition uniformity.

0 100 200 300 400 500 600 700 800 900 1000

0.0

2.0x10-5

4.0x10-5

6.0x10-5

8.0x10-5

1.0x10-4

Criteria = V/cm

IC = 905 A / 12 mm

N-value = 44.2

V (

Volt

age

)

I ( A / 12 mm )

(77 K, s.f.)

0 20 40 60 80 1000

200

400

600

800

1000

1200

I C (

A )

Position(m)

Optimization of RCE-DR process for thick superconducting layer


RCE-DR results (with optimization deposition region)

1,050A/12mm-w

(→ 875A/cm-w)

>800A/cm

0 200 400 600 800 1000 1200

0.0

1.0x10-5

2.0x10-5

3.0x10-5

4.0x10-5

5.0x10-5

6.0x10-5

Vo

ltag

e (

V)

Current (A)

@ 77 K, self-field

Speed

(m/min) Turns

Thickness

(㎛)

IC

(A/cm)

JC

(MA/cm2)

2 14 1 500 5

2 14 1.3 600 4.6

2 14 1.9 400 2.1

2 16 1.6 800 5

2 > 20 2 ~ 2.5 > 1,000 > 5

2016 Plan for making 400 A / 4 mm CC

0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.20

200

400

600

800

1000

1200

1400

IC

I C (

A/1

2 m

m)

Thickness (m)

0

1

2

3

4

5

6

7

JC

JC (M

A/c

m2)

plan

plan

results

1.6 m-thick 5.5 MA/cm2

0 100 200 300 400 500 600 700 800 900 10000

200

400

600

800

1000

I C (

A/c

m)

Length (m)

1,000 A/12 mm

The same process speed(120 m/hr ).


스티커 노트

We have achieved Ic = 1,050 A/12mm-width by applying the optimal deposition region. We are now working to attain Ic = 1,200 A/12mm-width in the second half of 2016.

MCI(Metal Clad Insulation) 2G wire

for high field magnet


Multi-width Double Pancake Coils

Stacked Double

Pancake Coils Fully assembled

No-insulation, multi-width, and compact !

4.1mmW 5.1mmW 6.1mmW 7.1mmW 8.1mmW

327 m

m

O.D. 172mm

I.D. 35mm Immersed in liquid Helium

26.4 T all 2G wire one-body(non-nested) magnet

(Designed by S. Hahn (MIT NHMFL/FSU)


https://nationalmaglab.org/


스티커 노트

Through a collaboration with Prof. S. Hahn, we succeeded in manufacturing HTS magnet, generating 26.4 T of center field at liquid Helium temp. This magnet is the first one wound only with 2G HTS wire in non-nested fashion, that generated such a high field.

NI-MW winding technic – No insulation

No-insulation winding technic :

Pros :

Compactness : without thick stabilizer

Strong mechanical strength : without soft insulation material

Self protection : automatic bypass

Rapid quench propagation

“Automatic bypass”

of the exceeding

current and better

protection to quench Without insulation

Quench

bypass current

(by S. Y. Hahn)

𝑄𝑅𝐶 W = 𝐼𝑅

2 ∙ 𝑅𝐶

where, 𝐼𝑅 = 𝐼𝑃𝑆 − 𝐼𝜃 = 𝐼𝑃𝑆 −𝐵0

𝑘0

Cons :

Charging time delay.

(excess heat generation/

Impractically slow for

charging)

(by S. Hahn)




스티커 노트

Pros and cons of no-insulation winding, and an equivalent circuit which is very useful in describing the behavior.

Metal Clad HTS 2G wire & coil

0

10

20

30 Power supply current

Cu

rren

t [A

]

30 A operating at 77 K

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

45.3 seconds

3.1 seconds

Copper plated HTS coil

Stainless steel cladded HTS coil

Mag

net

ic F

ield

at

SP

co

il c

ente

r [k

G]

Time [min]

95 % target field

15 times shorter !!!

J. Kim et al, “Effect of Resistive Metal-Cladding of HTS Tape on the Characteristic of No-Insulation Coil,” IEEE Trans. Appl. Supercond., 2016

Parameters Values

Inner diameter 58 mm

outer diameter 115 mm

Turns 275

Inductance 7.8 mH

< Copper plating CC tape >

Sputtering

Process


스티커 노트

Charging delay, one drawback of the no-insulation coil, can be reduced by metal-cladding to the tapes, adding extra resistance between turns.

Burn out test @ 77 K ( SPC with Copper stabilizer vs. MCI )

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30

0

50

100

150

200

250

300

350

400

450

5007.3 I

c Current

Voltage

Field

Time [min]

Cu

rren

t [A

]

EPL DPC Quench @ 473 [A]

Ic = 65 A

-0.25

-0.20

-0.15

-0.10

-0.05

0.00

0.05

0.10

0.15

0.20

0.25

Voltage

[V]

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

Field

[kG]

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30

0

50

100

150

200

250

300

350

400

450

500

SUS DPC Quench @ 420 [A]

Current

Voltage

Field

Time [min]

Curr

ent

[A]

0

1

2

3

4

5

6

7

8

9

10

Voltage

[V]

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

Field

[kG]

Ic = 65 A

6.5 Ic


스티커 노트

Added resistance reduced charging delay, but did not compromise quench stability.

Magnet Operation Results

Time constant, t, is calculated to 11 seconds.

Contact resistance between turns, RS,STS = 165 [W·cm2]

(By J. Kim et. al., CP-14, yesterday)


https://www.kimm.re.kr/


스티커 노트

We adopted metal-cladded tapes to 3 T demo magnet which is a scale-down version of 9.4 T/400 MHz NMR magnet.

STS cladded tape Charging time is 41 times shorter

Reduction of Charging Delay

If magnet was wound with copper plated tape,

(By J. Kim et. al., CP-14, yesterday)




스티커 노트

Reduction of charging delay is impressive.

Higher Je : metal substrate removal

process


Before combining After combining

0 100 200 300 400 500 600 700 800

0.0

5.0x10-6

1.0x10-5

1.5x10-5

2.0x10-5

2.5x10-5

3.0x10-5

12 mm-width : 740 A

4 mm-width Center : 234 A

4 mm-width Left : 235 A

4 mm-width Right : 240 A

Volta

ge (

V)

Current (A)

Cutting side

After bending

System for Al2O3 barrier & Y2O3 seed System for IBAD, homo-epi & buffer

Combining Barrier, Seed, IBAD, Buffer Systems in One

For standard process,

Stainless steel ~ 100 m thick

Hastelloy ~ 60 m thick


스티커 노트

We connected IBAD and sputtering (deposition process of Al2O3, Y2O3) to build an in-line system for prevention of contamination by external exposure. -> Improvement of adhesive property of coated conductor

Utilize this properties !!


스티커 노트

We utilize small the value of the peel/cleavage stress, which sometimes degrade the performance of impregnated coils, to our advantage. We cleave the 2G tape such that the SC layer sticks to the thin protecting layer rather than to the thick substrate.

High Je wire by removal of thick metal substrate

For Je, substrate thickness must be thin

Superconductor layer

Substrate

Stabilizer

For thin substrate, easy to damage during the reel

to reel process

Superconductor layer

Substrate

Stabilizer

Soldering thin substrate

on top of CC

Remove bottom substrate

Improvement of Je

4 roll 도면 or 사진

Intentionally making a weak interface

by some treatment


스티커 노트

Schematic of peel/cleavage process and how and where we invoke cleavage.

Demonstration of High Je wire by removal of thick metal substrate

0 2 4 6 8 10 120

50

100

150

200

250

300

I C (

A/4

mm

)

Position (m)

0 2 4 6 8 10 120

50

100

150

200

250

300

I C (

A/4

mm

)

Position (m)

45 m-thick (brass)

100 m-thick

Removing

Bottom

Substrate

Laminated

CC

Substrate

Removed

CC

Bottom

Substrate

Laminated

CC

Substrate

Removed

CC

Substrate remove machine

- Easily reduce the thickness ~ < 20 m

- Choice of any materials(SUS, Copper…)

May possible…

2 m

5 m

20 m

Superconductor

layer

Stabilizer

Superconductor

layer

Stabilizer

Laminated

Substrate

Laminated

Substrate


스티커 노트

Resulting tapes, tape performances, and apparatus for the processing.

SuNAM has been producing high IC coated conductors consistently.

Introduction of in-line Q.C. measures enhanced wire uniformity &

production yield.

With thicker(1.3 m 1.6 m) S.C. layer, we achieved >1,000 A/12 mm

in production.

We demonstrated 3 T magnet using MCI coated conductor.

Initial test of substrate removal & suggesting a new way of high Je wire

structure.

Summary


Price Reduction in RCE DR process

100

25 10

(Unit: USD / kAm)

12 mm 120 mm 360 mm Width :

Capacity : 1,000 km/y 15,000 km/y 75,000 km/y

Achievable with Existing Line of

SuNAM

“Increasing Demand for HTS

2G wire has surpassed the

supply”

“For market entrance $ 50 /

kAm is the threshold ”

“Price Reduction will ignite

an exponential growth of

demand for HTS 2G wire”

“High throughput, low

material cost, High yield is 3

Critical Success Factor”

Direction of Technology Development in the Future


Thanks for

Attention !

www.i-sunam.com


SuNAM : J. H. Lee, H. K. Kim, B. J. Mean, Y. S. Kim, S. W. Yoon, K. K. Cheon,

and H. J. Lee.

Seoul Nat’l Univ. : J. W. Lee, S. M. Choi, S. I. Yoo.

KERI : H. S. Ha, S. S. Oh.

Korea Polytech. Univ. : G. W. Hong, H. G. Lee.

Andong Nat’l Univ. : H. S. Shin.

Stanford Univ. : R. H. Hammond.

iBeam Materials : V. Matias

Univ. of Cambridge : J. M. Driscoll

FSU/NHMFL : S.Y. Han, D. Labalastier

MIT : Y. Iwasa

Acknowledgement

Thanks for

Attention !


Recent progress on SuNAM’s coated conductor development ...

Documents

Transcript of Recent progress on SuNAM’s coated conductor development ...