electromechanical properties of technical superconductors · hg] year mgb 2 nbti nb 3 sn bi2223...

Post on 11-Mar-2020

3 Views

Category:

Documents

0 Downloads

Preview:

Click to see full reader

TRANSCRIPT

Group of Applied SuperconductivityDepartment of Quantum Matter Physics

University of Geneva, Switzerland

Electromechanical Properties

of Technical Superconductors

Carmine SENATORE

http://supra.unige.ch

Outline

Forces, stress and strain in a magnet

• Reversible vs. irreversible effects

• Mechanisms behind the irreversible degradation of the critical current

Focus on two materials: Nb3Sn and YBCO

Technical superconductors under mechanical loads

• Intrinsic vs. extrinsic effects

From superconducting materials…

…to technical superconductors

1. Superconducting ? 10’000

2. Tc>4.2K & Bc2>10T ? 100

3. Jc>1000 A/mm2 ? ~10

1900 1940 1980 1985 1990 1995 2000 2005 2010 20150

10

20

30

40

50

100

150

200

PbMo6S

8

H2S @ 140 GPa

AEFeAs

LaFeAsO

LaFePO

FeSe-1 layer

HgBaCaCuO @ 30 GPa

Night on

the Moon

Surface of Pluto

Liquid nitrogen

Liquid neon

Liquid hydrogen

Liquid helium

FeSeTe

SmFeAsO

HgTlBaCaCuO

HgBaCaCuO

TlBaCaCuO

BiSrCaCuO

LaBaCuO

YBaCuO

YC6

CaC6

CNT

diamondCNT

K3C

60

RbCsC60

CeCoIn5UPd

2Al

3CeCu2Si

2UPt

3

MgB2

BKBONb3Ge

V3Si

Nb3Sn

NbN

NbTiNbPb

Hg

Tem

per

atu

re [

K]

Year

MgB2

NbTi

Nb3Sn

Bi2223

Bi2212

YBCO

Operate at high current density is a necessary condition for applications, but it is not sufficient

Other crucial requirements:

• Have high tolerance to stress

• Be safe in case of magnet quench

• Have low magnetization

• Have a persistent joint technology

Quench detection, NZPV

Applications to NMR, MRI, HEP magnets

Applications to NMR, MRI

Magnetic forces

Operate at high current density is a necessary condition for applications, but it is not sufficient

Other crucial requirements:

• Have high tolerance to stress

• Be safe in case of magnet quench

• Have low magnetization

• Have a persistent joint technology

Quench detection, NZPV

Applications to NMR, MRI, HEP magnets

Applications to NMR, MRI

Magnetic forces

Introduction to Forces and Stresses in a Magnet

Infinite solenoidthin wall

B=B0 B=0

0 0B JdThe magnetic field is

The expression of the magnetic force density (per unit of volume) is

f J B

The average field in the winding is 0

2

B

F

dr

0 2

BJ r z d

20

0

2

Br z

20

02m

Bs p s

The (radial) force on a winding volume element is

0

2

BF J v

0( 1 ) 4 mp B T bar 0( 10 ) 400 mp B T bar

Hoop stress in a ring

r

loop

F dvf dvJ B

A ring carrying a current I in a field B

2 R I BB

I = J·Swire

The total radial magnetic force on the ring is

R

x x x x x x x x

x x x x x x x x

x x x x x x x x

x x x x x x x x

x x x x x x x x

x x x x x x x x

x x x x x x x x x x x x x

x x x x x x x x x x x x x

x x x x x x x x x x x x x

x x x x x x x x x x x x x

x x x x x x x x x x x x x

x x x x x x x x x x x x x

x x x x x x x x x x x x x

x x x x x x x x x x x x x

x x x x x x x x x x x x x

x x x x x x x x x x x x x

Hoop stress in a ring

r

loop

F dvf dvJ B

A ring carrying a current I in a field B

2 R I B

radial force2

rF F

A tension is developed within the ringTT /2/2

2 sin 22 2

T T T

R I B2

rFT

The so-called “hoop stress” on the wire is wire

TR J B

S

The total radial magnetic force on the ring is

Hoop stress levels above 100 MPa are common, the NHMFL 32 T magnet operates at 400 MPa

Electromagnetic stresses in a finite solenoidIn a winding adjacent turns will press on each other and develop a radial stress r which modifies the hoop stress

The condition for equilibrium between radial stress r , hoop stress and body force BJr is given by the equation

2

2

1 1d du ur BJ

r dr dr Er

Considering the solenoid as a continuous uniform medium, from the Hooke’s law

21r

E du u

dr r

21

E u du

r dr

and

where u is the local displacement in the radial direction, E is the Young’s modulus and is the Poisson’s ratio

F

Electromagnetic stresses in a finite solenoid

magnetic lines of force vectors of electromagnetic force per unit volume

20

02

Bp

r

bore radius

' BJr

Example: for B0 of the order of 10 T, on the winding of > 200 MPa (> 2000 bar)

Colour mapping of and r in a finite solenoid

Hoop stress Radial stress

Electromagnetic stresses in an accelerator dipole

magnetic lines of force vectors of electromagnetic force per unit volume

In straight-sided coils such as dipolesand quadrupoles the conductor is unableto support the magnetic forces in tension

20

02

Bp

For B0=6T and (a+b)/2=100 mm

Fx=200 tons/m

Collaring and Pre-stress in accelerator magnets

175 tons/m

85 tons/mF

The windings of accelerator magnets are clamped in a solid collar

Collar provide pre-stressing to prevent the movement of the coil in the presence of electromagnetic forces

Thermal precompression of the superconductor

Mismatch in thermal contraction within the composite induces a precompression in the SC

650°C 4.2 K

m

When the SC is cooled at the operating temperature,its crystal structure deforms and this induces changeson Tc , Bc2 (and thus on Jc)

MgB2

NbTi

Nb3Sn

All technical superconductors are composite

Bi2223

Bi2212

YBCO

In the following, the focus will be on

Today high field superconductor, up to 23.5 T in

solenoidal coils, the material for and

and only candidate material for the 16 T dipoles of

Nb3Sn

YBCO Tomorrow high field superconductor, more than 40 T

demonstrated in solenoidal insert coils, the way to get

20 T dipoles and beyond

In the following, the focus will be on

Today high field superconductor, up to 23.5 T in

solenoidal coils, the material for and

and only candidate material for the 16 T dipoles of

Nb3Sn

YBCO Tomorrow high field superconductor, more than 40 T

demonstrated in solenoidal insert coils, the way to get

20 T dipoles and beyond

Industrial fabrication of Nb3Sn wires

Three technologies have been developed at industrial scale

• Bronze route

• Internal Sn diffusion

• Powder-In-Tube (PIT)

The Sn source is the main difference

Presently produced by

Nb

Sn

Critical current density vs. magnetic fieldBest performance achieved so far in industrial wires

12 14 16 18 20 220

500

1000

1500

2000

2500

3000

FCC Target

n

on

-Cu

Jc [

A/m

m2]

B [T]

PIT - BKr

= 26.5 T

Internal Sn - BKr

= 24.9 T

Bronze - BKr

= 25.6 T

@ 4.2 K

J. Parrell et al., AIP Conf. Proc. 711 (2004) 369

T. Boutboul et al., IEEE TASC 19 (2009) 2564

V. Abächerli et al., IEEE TASC 17 (2007) 2564

Strain-induced changes in the critical current

m irr applied strain

crit

ica

l cu

rren

t

The applied force reduces the effects of the thermal precompression and the critical current increases up to a maximum

The applied force induces breakages within the filaments and the critical current is irreversibly degraded

0

Effects of the longitudinal strain

Strain-induced changes in the critical current

m irr applied strain

crit

ica

l cu

rren

t

0

Effects of the longitudinal strain

REVERSIBLE REGIONIc recovers after force unload

IRREVERSIBLE REGIONIc does not recover after force unload

Technology = m-irr [%]

Bronze Route 0.4 – 0.6

Internal Sn 0.05 – 0.2

Powder-In-Tube

0.15 – 0.3

Wire cabling for the ITER coils

CICC = Cable-In-Conduit Conductor

Wire cabling for the ITER coilsReversible variation of Ic under strain matters

CICC = Cable-In-Conduit Conductor

Because of the thermal precompression, wires in a CICC experience an effective axial compressive strain ranging between -0.6% and -0.8%

-0.8 -0.6 -0.4 -0.2 0.00.0

0.2

0.4

0.6

0.8

1.0 01UL0299A01

no

rmal

ize

d I c

intrinsic strain [%]

@ 4.2 K, 12 T

The performance of the superconductor is limited to ~40% of the maximum achievable current

How to measure Ic vs. axial strain The WASP (Walters Spring) probe

Ti-6Al-4V

Cu

B. Seeber et al., Rev. Sci. Instr. 76 (2005) 093901

Motor @ room temperature

Max current 1000 A

Sample length 1.1 meter

Voltage taps distance 126 mm

Electrical field criterion 0.01 μV/cm

Strain (ε) up to ± 1.2 %

Why do superconducting properties depend on strain ?

1. Reversible effects of strain

Why superconducting properties depend on strainReversible effects of strain

*

1.04 1exp

1.2 1 0.62c

B

Tk

2

FN E

Nb3Sn exhibits strong-coupling superconductivity

is an average phonon frequency

* gives the Coulomb repulsion at the Fermi surface

2

0

F2 d

McMillan parameter

electron-phonon coupling phonon DOS

So far the question whether the strain sensitivity is due mainly to electrons orphonons has not been settled

Strain induces change both in the phonon spectrum and the electronic bands

Why superconducting properties depend on strainReversible effects of strain

Some recent theoretical studies

G. De Marzi et al., J. Phys.: Condens. Matter 25 (2013) 135702Investigation of the strain sensitivity of Nb3Sn from first principle calculations based on the density functional theory

M. Mentink, PhD thesis (2014), University of Twente

Ab-initio calculations used to evaluate electronic, and vibrational properties, as well as Tc and Bc2 as a function of disorder, crystal orientation, and strain

D. Valentinis et al., Supercond. Sci. Technol. 27 (2014) 025008A theory of the strain-dependent critical field in Nb3Sn based on anharmonic phonon generation

D. Taylor and D. Hampshire., Supercond. Sci. Technol. 18 (2005) S241Microscopic theory analysis suggests that the uniaxial strain effects are predominantly due to changes in the average phonon frequencies

D. Markiewicz, Cryogenics 44 (2004) 767This work enphasizes the role of anharmonicity as a source of the strain dependence in Nb3Sn

Critical temperature vs. hydrostatic compression Reduced critical temperature vs. unaxial strain

D. Markiewicz, Cryogenics 44 (2004) 767S. Tanaka et al., J. Phys. Soc. Jpn. 81 (2012) SB026

Reversible effects of strainHydrostatic vs. uniaxial

Cubic-shaped stress-free cell for Nb3Sn

Precompression induced by cooling to low temperature has two components

HYDROSTATICChange of the cell volume

DEVIATORICChange of the cell shape

Reversible strain effects in Nb3Sn wires

Reversible strain effects in Nb3Sn wires: lattice parameters

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.85.260

5.265

5.270

5.275

5.280

5.285

5.290

5.295

5.300

latt

ice

par

ame

ter

[Å]

applied strain [%]

C. Scheuerlein et al., IEEE TAS 19 (2009) 2653

XRD experiments @ ESRF Grenoble

L. Muzzi et al., SUST 25 (2012) 054006

Bronze route wire: Nb3Sn lattice parameters vs uniaxial strain @ 4.2 K

ZERO APPLIED STRAINNb3Sn is precompressedHydrostatic+Deviatoric

CROSSING POINTCubic cell recoveredHydrostatic component still present

HIGH APPLIED STRAIN Again Hydrostatic+Deviatoric

Lattice parameters and Ic under axial strain

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.85.260

5.265

5.270

5.275

5.280

5.285

5.290

5.295

5.300

#26

latt

ice

par

ame

ter

[Å]

applied strain [%]

@ 19T, 4.2K

50

100

150

200

250

300

crit

ical

cu

rren

t [

A]

Bronze Route wire

MAXIMUM OF Ic

The maximum of the critical current occurs when the cubic cell is restored

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.80.0

0.2

0.4

0.6

0.8

1.0 PIT Internal Sn Bronze

0.4

2%

< ir

r<0.4

5%

irr=0

.62

%

I c / I cm

ax

applied strain [%]

0.4

1%

< ir

r<0.4

4%

@ 19T, 4.2K

Bronze Route, Internal Sn and PIT

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.80

50

100

150

200

250

300

350

400

450

E0=125 GPa

Rp0.2

=218 MPa

@ 4.2K

stre

ss [

MP

a] strain [%]

Bronze

Ic vs. axial strain

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.80

50

100

150

200

250

300

350

400

450

E0=125 GPa

Rp0.2

=218 MPa

@ 4.2K

stre

ss [

MP

a] strain [%]

Internal Sn

Bronze E0=101 GPa

Rp0.2

=194 MPa

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.80

50

100

150

200

250

300

350

400

450

E0=111 GPa

Rp0.2

=115 MPa

E0=101 GPa

Rp0.2

=194 MPa

E0=125 GPa

Rp0.2

=218 MPa

@ 4.2K

stre

ss [

MP

a] strain [%]

PIT

Internal Sn

Bronze

Technology irr

Bronze Route 330 MPa

Internal Sn 210 MPa

Powder-In-Tube 150 MPa

0 50 100 150 200 250 300 350 4000.0

0.2

0.4

0.6

0.8

1.0

PIT Internal Sn Bronze

20

7 M

Pa

<

irr<

22

0 M

Pa

ir

r=32

8 M

Pa

I c / I cm

axaxial stress [MPa]

15

3 M

Pa

<

irr<1

64

MP

a

@ 19T, 4.2K

Ic vs. axial stress

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.80

50

100

150

200

250

300

350

400

450

E0=111 GPa

Rp0.2

=115 MPa

E0=101 GPa

Rp0.2

=194 MPa

E0=125 GPa

Rp0.2

=218 MPa

@ 4.2K

stre

ss [

MP

a] strain [%]

PIT

Internal Sn

Bronze

Reversible behaviour and irreversible limit

Why do superconducting properties depend on strain ?

2. Irreversible effects of strain

Irreversible degradation phenomenaTwo irreversible phenomena play together

• Crack formation in Nb3Sn

• Plastic deformation of the Cu matrix

PIT Nb3Sn wire under transverse loadStress map of the Cu matrix

Yield strength of Cu y ~ 90 MPa

Plastically deformed Cu imposes a stress on Nb3Sn after force unload (= lattice deformation)

What is the residual stress on Cu after unload ?

resCu = max(Mises

Cu – y , 0 )

Nb3Sn is a brittle material and is characterized by a strong propensity to fracture

Voids formed during the reaction cause localized stress concentrations where cracks nucleate

Voids in Nb3Sn wires

Bronze Route121 x 121 filaments

Internal Sn132/169 subelements

PIT192 filaments

XRD microtomography reconstruction

Can we quantify the impact of voids on the electromechanical limits?

XRD microtomography at ESRF Grenoble

sample

- XRD photon energy = 89 keV

- 360° rotation of the sample

- 30’000 projections

- 2560 x 2160 pixels

- 0.57 m/pixel resolution

ab

sorp

tio

n

parallel XRD beam

rotate

Non-destructive 3D volume reconstruction with separation of internal features

Bronze Route wire

Virtual longitudinal cut

XRD microtomography at ESRF Grenoble

A case study on Bronze Route Nb3Sn wires

Why Bronze Route wires ?

Because by Hot Isostatic Pressing (HIP) treatment we can reduce the void fraction without degrading the critical current

What did we do ?

Experimental determination of the electromechanical limits

Statistical analysis of the voids size, shape and distribution

Finite Element Modelling (FEM) to quantify the impact of voids

Possibility to investigate the same wire type with and without voids

A case study on Bronze Route Nb3Sn wires

NbCu

Bronze

Nb3Sn

manufacturer

wire diameter 1.25 mm

# of filaments 121 x 121

filament size 4.5 m

irr =

0.6

8 %

max

= 0

.20

%

0.0 0.2 0.4 0.6 0.8 1.00

50

100

150

200

regular HT

crit

ical

cu

rre

nt

I c [A

]

applied strain [%]

@ 4.2 K, 19 T

Regular HT: 600°C/100h + 670°/150hc = irr– max = 0.48 %

HIP 550°C/1h/200MPa + Regular HT

c = irr– max = 0.65 %

irr =

0.6

8 %

irr =

0.8

9 %

max

= 0

.24

%

max

= 0

.20

%

0.0 0.2 0.4 0.6 0.8 1.00

50

100

150

200

regular HT

HIP + regular HT

crit

ical

cu

rre

nt

I c [A

]

applied strain [%]

@ 4.2 K, 19 T

With HIP treatment c increases by +0.17 %

Bronze route wire: Void detection

Without HIP treatmentVoid fraction = 2.1 %

With HIP treatmentVoid fraction = 0.05 %

Void reduction by a factor 50 !!

Void detection and analysis

- A brightness distribution analysis is

used to detect voids

- Voids are approximated with ellipsoids

- Size and orientation are determined

• Major axis length

• Major axis angles and

• Eccentricity (sphericity)

- Position is recorded, including distance

to the nearest neighbors and location

(matrix, filament bundle, interface)

These data become input for FEM

Bronze route wire: Void analysis

Void size Void shape

0.0 0.2 0.4 0.6 0.8 1.01

10

100

1k

10k

100k regular HT

cou

nt

sphericity, 0.0 0.2 0.4 0.6 0.8 1.01

10

100

1k

10k

100k regular HT

HIP + regular HT

cou

nt

sphericity,

HIP voids are morespherical

0 5 10 15 20 251

10

100

1k

10k

100k

cou

nt

principal axis length [m]

regular HT

HIP void size is reduced

0 5 10 15 20 251

10

100

1k

10k

100k

cou

nt

principal axis length [m]

regular HT

HIP + regular HT

Statistical FEM analysis

- 3D model, real dimensions, 0.1 mm long wire

- Bottom side is fixed, top side is displaced to apply a mechanical strain

The output of the model is the

von Mises stress distribution in the

Nb3Sn volume at a given strain

Statistical FEM analysis: 1) no voids - The experimental c corresponds to an irreversible reduction of Ic by 5%

- Working hypothesis: 5% of Ic degradation damage in 5% of the filaments

Wire with HIP treatment

SIMULATION done at c = 0.65%

The 5% of the Nb3Sn volume at the highest

stress is highlighted in red

Here the red regions are at 275 MPa

The critical stress is thus c = 275 MPa

Statistical FEM analysis: 2) with voids - Voids are generated from the statistical analysis until the void fraction

of the wire is reached

- The goal is to predict the reduction of c induced by the voids

- Simulation runs at increasing until c in 5% of the filaments

c = 275 MPa

c = 0.65%

0.3 0.4 0.5 0.6 0.7 0.8150

200

250

300

350

400

HIP + regular HT

regular HT

up

pe

r 5

% s

tre

ss,

up

pe

r 5

% [

MP

a]

intrinsic strain, [%]

c = 275 MPa

c = 0.65%

0.3 0.4 0.5 0.6 0.7 0.8150

200

250

300

350

400

HIP + regular HT

regular HT

up

pe

r 5

% s

tre

ss,

up

pe

r 5

% [

MP

a]

intrinsic strain, [%]

c = 275 MPa

c = 0.65%

0.3 0.4 0.5 0.6 0.7 0.8150

200

250

300

350

400

HIP + regular HT

regular HT

up

pe

r 5

% s

tre

ss,

up

pe

r 5

% [

MP

a]

intrinsic strain, [%]

c = 275 MPa

c = 0.65%

0.3 0.4 0.5 0.6 0.7 0.8150

200

250

300

350

400

HIP + regular HT

regular HT

up

pe

r 5

% s

tre

ss,

up

pe

r 5

% [

MP

a]

intrinsic strain, [%]

c = 0.50%

c = 275 MPa

c = 0.65%

0.3 0.4 0.5 0.6 0.7 0.8150

200

250

300

350

400

HIP + regular HT

regular HT

up

pe

r 5

% s

tre

ss,

up

pe

r 5

% [

MP

a]

intrinsic strain, [%]

In the presence of voids the model predicts c = 0.50%

Irreversible limit in the presence of voidsExperiment vs Prediction

0.0 0.2 0.4 0.6 0.8 1.00

50

100

150

200

regular HT

HIP + regular HT

crit

ical

cu

rre

nt

I c [A

]

applied strain [%]

@ 4.2 K, 19 T

c = 0.48%

c = 0.65%

Ic vs strain measurement

c = 0.65%

FEM simulations

c = 275 MPa

0.3 0.4 0.5 0.6 0.7 0.8150

200

250

300

350

400

regular HT

HIP + regular HT

up

pe

r 5

% s

tre

ss,

up

pe

r 5

% [

MP

a]

intrinsic strain, [%]

c = 0.50%

The simulations predict the correct value of c when voids are introduced

What did we learn ?

Changes in the voids correlate quantitatively with the changes in the electromechanical limits

More details in

Case study on Bronze Route, what about Internal Sn and PIT ?

The same approach may lead to the prediction of how much c can be increased by reducing the void fraction

Reversible effects of strain

Influence on the stress tolerance of the way the mechanical stresses are exerted

Intrinsic mechanisms behind the irreversible degradation of the critical current

Strain-induced changes in Nb3Sn wires

Electromagnetic stresses in an accelerator dipole

magnetic lines of force vectors of electromagnetic force per unit volume

Accelerator magnets operate at ~10 kA to keepinductance low and ease magnet protection

Rutherford cables are used to get these largeoperating currents

LHC27 km, 8.33 T

14 TeV (c.o.m.)1300 tons NbTi

FCC-hh100 km, 16 T

100 TeV (c.o.m.)~9000 tons Nb3Sn

HE-LHC27 km, 16 T

27 TeV (c.o.m.)~3000 tons Nb3Sn

Geneva

PS

SPS

LHC

Courtesy of L. Bottura (CERN)

The Future Circular Collider Study

Degradation upon transverse loads

The 16 T FCC dipoles are being designed with a peak stress of 200 MPaat operation

Nb3Sn Rutherford cable for HL-LHC, 40 strands

Are the Nb3Sn wires in the cable able to withstand such a high stress level? Which degradation is tolerable?

• Nb3Sn wires are deformed during cabling

• Cables are braided with glass fiber

• The winding is impregnated with resin

The irreversible limit under transverse stress is influenced by parameters extrinsic to the wire

• the type of impregnation (the elastic modulus of the resin)

• the redistribution of the applied stress on the wire i.e. the deformation of the wire during cabling

Tested Epoxy L, Glass fiber + Epoxy L, Stycast

The WASP concept for Ic vs. transverse stress

4-WALL + impregnation

Pulling force

Resin

Sample

collaboration agreement K1629/TE (2009-2012)

Ic vs. transverse stressPIT 192 + Epoxy L

0 5 10 15 20 25 30 350.2

0.4

0.6

0.8

1.0

@ 4.2K, 19 T

I c / I c0

Transverse force [kN]

#31712-1 unload

95% Ic0

0 30 60 90 120 150 180 210 240Transverse stress [MPa]

ForceStress

groove length groove w th id

irr = 110 MPa

The corresponding irreversible stress limit is

where

The irreversible limit is defined at the force level leading to a 95% recovery of the initial Ic after unload

Here Firr = 16 kN

60 80 100 120 140 160 180

0.0

0.5

1.0

1.5

2.0

2.5

F = 0.5 kN

F = 15 kN

unload to 0.5 kN

@ 4.2 K, 19 T

Elec

tric

fie

ld [

V/c

m]

Current [A]

Epoxy L

Sample

Glass fiber

Ic vs. transverse stress: wire in a glass fiber sleeve

0 30 60 90 120 150 180 210 2400.2

0.4

0.6

0.8

1.0

PIT #31712 Ø = 1.0 mm

@ 4.2K, 19 T

95% Ic0

Transverse stress [MPa]

I c / I c0

sample #1 round

with glass fiber sleeve

PIT #14310 Ø = 1.0 mm

0 30 60 90 120 150 180 210 2400.2

0.4

0.6

0.8

1.0

PIT #31712 Ø = 1.0 mm

@ 4.2K, 19 T

95% Ic0

Transverse stress [MPa]

I c / I c0

sample #1 round

after force unload

with glass fiber sleeve

after force unload

PIT #14310 Ø = 1.0 mm

Glass fiber adds rigidity to the impregnation Shift of irr by > 50 MPa

Ic vs. transverse stress: Epoxy L vs. Stycast

0 5 10 15 20 25 30 350.2

0.4

0.6

0.8

1.0

PIT #31712 Ø = 1.0 mm

@ 4.2K, 19 T

I c / I c0

Transverse force [kN]

epoxy impregnation

stycast impregnation

95% Ic0

0 30 60 90 120 150 180 210 240

Transverse stress [MPa]

0 5 10 15 20 25 30 350.2

0.4

0.6

0.8

1.0

PIT #31712 Ø = 1.0 mm

@ 4.2K, 19 T

I c / I c0

Transverse force [kN]

epoxy impregnation

epoxy impregnation after force unload

stycast impregnation

stycast impregnation after force unload

95% Ic0

0 30 60 90 120 150 180 210 240

Transverse stress [MPa]

The change of resin, from epoxy to stycast, leads to an increase of irr by > 50 MPaThe result is comparable to the value found with epoxy + glass fiber sleeve

The Young modulus ofStycast is 3 to 4 timeshigher compared to thevalue of Epoxy L

Ic vs. transverse stress on flattened wires

0 50 100 150 200 250

60

80

100

120

140

160

180

sample #1 round

sample #1 round after force unload

sample #2 15% rolled

sample #2 15% rolled after force unload

Transverse stress [MPa]

PIT #31712 Ø = 1.0 mm

@ 4.2K, 19 T

I c [A

]

PIT 192

~7.5% Ic reduction by rollingShift of irr by ~ 40 MPa

NO Ic reduction by rollingShift of irr by ~ 15 MPa

50 100 150 200 250

60

80

100

120

140

160

180

@ 4.2K, 19 T

95% Ic0

Transverse stress [MPa]

I c [A

]

sample #1 round

sample #1 round after force unload

sample #2 15% rolled

sample #2 15% rolled after force unload

RRP 132/169 #114163 Ø = 1.0 mm

Internal Sn 132/169

Effects of the stress redistribution

This wire can withstand the peak stress of the FCC dipoles !!

12 14 16 18 20 220

500

1000

1500

2000

2500

3000

FCC Target

no

n-C

u J

c [A

/mm

2]

B [T]

PIT - Bc2

Kr= 26.5 T

Internal Sn - Bc2

Kr= 24.9 T

@ 4.2 K

T. Boutboul et al., IEEE TASC 19 (2009) 2564

@ 200 MPaCritical current density vs. magnetic field

Don’t forget the reversible effects of stress !!

12 14 16 18 20 220

500

1000

1500

2000

2500

3000

FCC Target

no

n-C

u J

c [A

/mm

2]

B [T]

PIT - Bc2

Kr= 23.2 T @ 200 MPa

Internal Sn - Bc2

Kr= 23.8 T @ 200 MPa

@ 4.2 K, 200 MPa

J. Parrell et al., AIP Conf. Proc. 711 (2004) 369

Best performance achieved so far in industrial wires

High Temperature Superconductors are

different animals …

Bi2212Bi2Sr2CaCu2O8+x

Bi2223Bi2Sr2Ca2Cu3O10+x

Y123YBa2Cu3O7-x

3 CuO2 planes

2 CuO2 planes

CuO chains

2 CuO2 planes

O

Ca

Cu

Sr

Bi

O

Cu

Y

Ba

HTS materials for applications

Y123 Coated Conductor

Bi2212Powder-In-Tube wire

Bi2223 Powder-In-Tube tape

YBa2Cu3O7-x (YBCO) coated conductors

Ag cap layer

Cu stabilisation

YBCO layer metallic substratebuffer layers

~1 µm of YBCO in a ~100 µm thick tape

[001] tilt grain boundaryH. Hilgenkamp and J. Mannhart, RMP 74 (2002) 485

Ag cap layer

Cu stabilisation

YBCO layer metallic substratebuffer layers

Biaxial texturing – within < 3° – is obtained

a

b

a

b

a

b

a

b

a a

b

a

b

Top view

The template is a metallic substrate coated with a multifunctional oxide barrier

Presently produced by

• complex and expensive manufacturing process

• pronounced anisotropic behaviour

but with some also drawbacks:

YBa2Cu3O7-x (YBCO) coated conductors

~1 µm of YBCO in a ~100 µm thick tape

Goyal et al., APL 69 (1996) 1795

Iijima et al., APL 60 (1992) 769

REBCO, RE = Rare Earth

Tc

Tc

Tc , Ic

0.00 0.05 0.10 0.15 0.20 0.25 0.3090.6

90.8

91.0

91.2

91.4

91.6

91.8

p // a

p // c

T c [K

]

p [GPa]

p // b

Critical temperature vs. uniaxial strain in single crystals

U. Welp et al., Phys. Rev. Lett. 69 (1992) 2130

a

b

a

b

a

Critical current vs. uniaxial strain in coated conductors

D. van der Laan et al.,Supercond. Sci. Technol. 24 (2011) 115010

@ 76 K, s.f.

Reversible effects of strain on YBCO

In the following, the focus will be on

Today high field superconductor, up to 23.5 T in

solenoidal coils, the material for and

and only candidate material for the 16 T dipoles of

Nb3Sn

YBCO Tomorrow high field superconductor, more than 40 T

demonstrated in solenoidal insert coils*, the way to get

20 T dipoles and beyond

*tested in a resistive outsert

All superconducting magnets beyond LTS The World cup of high magnetic fields

Maximum field [T]

HTS insertHTS coil field

[T]Winding

technology

32 REBCO 17 DP

27.6Bi2223 / REBCO

4.5 (Bi2223) 6 (REBCO)

LW

IEE CAS 27.2 REBCO 12.2 DP

26.4 REBCO 26.4 DP

IEE CAS 25.7 REBCO 10 DP

25 REBCO 4 LW

HFLSM 24.6 Bi2223 10.6 DP

24.2 Bi2223 3.66 LW

Stress management in Ultra High Field magnets

Hoop stress levels above 100 MPa are common, the NHMFL 32 T magnet operates at 400 MPa

In the case of layer winding the directionchanges going from one layer to the next impose in-plane hard bending of the tape

Other constrains to the winding come from the tape geometry of the conductor

F I B hoop

wire

IBR

S

B I

Ic vs. axial strain

C. Barth, G. Mondonico and CS, Supercond. Sci. Technol. 28 (2015) 045011

REBCO CCs: Dependence of Ic on axial loads

0.0 0.2 0.4 0.6 0.8 1.0

0.6

0.7

0.8

0.9

1.0

SuperPower

0.66 < irr

< 0.68 %

Bruker HTS

0.70 < irr

< 0.72 %

SuNAM

0.67 < irr

< 0.69 %

Fujikura

0.55 < irr

< 0.57 %

SuperOx

0.45 < irr

< 0.47 %

no

rmal

ized

cri

tica

l cu

rren

t, I c /

I c,m

ax

Applied strain [%]

@ 4.2 K, 19 T

• Very low stress effect curves are flat in rev. region

• Irreversible stress limits above 500 MPa

• The only weakness is delamination...

• REBCO CCs are inherently strong, ~50% is a high strength alloy

0.0 0.2 0.4 0.6 0.8 1.00

200

400

600

800

1000

4.2 K

Stre

ss [

MP

a]

Strain [%]

SuperPower:

E = 155 GPa, Rp0.2

= 893 MPa

Bruker HTS:

E = 187 GPa, Rp0.2

= 736 MPa

SuNAM:

E = 168 GPa, Rp0.2

= 846 MPa

Fujikura:

E = 176 GPa, Rp0.2

= 787 MPa

SuperOx:

E = 172 GPa, Rp0.2

= 1083 MPa

Ic vs. axial stress

0 200 400 600 800 1000

0.6

0.7

0.8

0.9

1.0

SuperPower

800 < irr

< 830 MPa

Bruker HTS

740 < irr

< 750 MPa

SuNAM

830 < irr

< 840 MPa

Fujikura

750 < irr

< 760 MPa

SuperOx

770 < irr

< 800 MPa

no

rmal

ized

cri

tica

l cu

rren

t, I c /

I c,m

ax

Applied stress [MPa]

Reversible behaviour and irreversible limit

HTS for accelerator magnets

The goal of 20 T in an accelerator quality dipole calls for HTS

1970 1980 1990 2000 2010 2020 2030 20400

2

4

6

8

10

12

14

16

18

20

FCC-hh

HTS Dipole

HL-LHC

LHCSSC

HERA

Tevatron

Mag

ne

tic

fie

ld [

T]

Year

SPS

Accelerator magnets operate at ~10 kA tokeep inductance low and ease magnetprotection

The most promising HTS CONDUCTOR for

accelerator magnets is the Roebel Cable (10

kA-class cable)

A coil of Roebel cable: a Short Dipole Demonstrator

A 5 T, 40 mm bore HTS based dipole demonstrator

J. Van Nugteren, PhD thesis, 2016

~1000 m of REBCO tape

~70 m of Roebel cable

5 T in standalone configuration

40 mm aperture

A coil of Roebel cable: a Short Dipole Demonstrator

A 5 T, 40 mm bore HTS based dipole demonstrator

J. Van Nugteren, PhD thesis, 2016

Are Roebel cables able to withstand

large transverse loads?

The answer is YES: Irreversible stress limit 400 MPa

50 100 150 200 250 300 350 400 450 5006.5

7.0

7.5

8.0

8.5

@ 4.2 K, 10.5 T

Cri

tica

l cu

rren

t, I c [

kA]

Transverse stress [MPa]

Bruker tape - CTD-101K impregnation

50 100 150 200 250 300 350 400 450 5002.5

3.0

3.5

@ 4.2 K, 10.5 T

Cri

tica

l cu

rren

t, I c [

kA]

Transverse stress [MPa]

SuperPower tape - Araldite + filler

SuperPower tape - CTD-101K impregnation

Short Dipole Demonstrator results

3.35 T

1st coil wound with moderate Ic SuNAM cable (~40% of the Bruker cable)

J. Van Nugteren @ EUCAS2017

New coil with ultra-high Ic cable from Bruker already wound Expected field in stand-alone test ~7T

What we have learned…

• Why we care about mechanical loads

• What hides behind the reversible degradation of Ic

• How the stress tolerance depends also on factors extrinsic to the wire

Electromagnetic forces in solenoids and accelerator magnets

Deviatoric strain, intrinsic anisotropy,…

Load geometry, impregnation…

• Which are the intrinsic mechanisms that lead to the irreversible degradation of Ic

Detailed study of the stress concentration at the voids in Nb3Sn

• An overview of the electromechanical properties of Nb3Sn and YBCO

Thank you for the attention !

Carmine SENATORE

carmine.senatore@unige.ch

http://supra.unige.ch

…time for questions…

If you want to know more about applied superconductivity in Geneva, visit http://supra.unige.ch

top related