1 May 23, 2014A. Godeke – Critical Surfaces – WAMHTS-1 workshop – DESY Hamburg Hansestadt Hamburg around 1400… …pirates, the “Vitalienbrüder” or “Likedeelers”

1May 23, 2014 A. Godeke – Critical Surfaces – WAMHTS-1 workshop – DESY Hamburg

Hansestadt Hamburg around 1400… …pirates, the “Vitalienbrüder” or “Likedeelers” based at Helgoland

Klaus Störtebecker and Godeke Michels

Critical SurfacesA comparison between LTS and HTS

Arno GodekeLawrence Berkeley National Laboratory, Berkeley, CA, USA

WAMHTS-1 workshop – DESY Hamburg – May 23, 2014

This work was partly supported by the Director, Office of Science, High Energy Physics,U.S. Department of Energy under contract No. DE-AC02-05CH11231

3May 23, 2014 A. Godeke – Critical Surfaces – WAMHTS-1 workshop – DESY Hamburg

Motivation: Record dipole fieldsHE-LHC and MAP: 20+ T dipoles HTS: High Tesla Superconductors

(Bi-2212, YBCO, Bi-2223)

Nb-Ti

18 T (1.9 K)

Nb3Sn

10.5 T

Dietderich and Godeke, Cryogenics 48, 331 (2008)

Nb3Sn quality and quantity in wiresMagnet technology

1 decade ago

2 decades ago

20+ T dipoles require High Tesla Superconductors (at least presently)

Nb3Sn

4May 23, 2014 A. Godeke – Critical Surfaces – WAMHTS-1 workshop – DESY Hamburg

How much current density is needed?JW: In windings JE: In wire

• 2 layer, 40 mm bore, 25 mm Rutherford– R1 = 20 mm, R2 = 70 mm, m0H = 20 T

– JW = 637 A/mm2

Maximum stress

Meuser, LBNL Eng. note M5254 (1978)Caspi and Ferracin, Proc. PAC-2005

Godeke, et al., Unpublished

JE = 600 A/mm2 & JW = 450 A/mm2

(if stress = OK)

Meuser, LBNL Eng. note M5256 (1978)Caspi and Ferracin, Proc. PAC-2005

Increase R2

Drop JW

Reduce stress

Bottura and Godeke, Rev. Accel. Sci. Techn. 5 25 (2012)

Rmax

5May 23, 2014 A. Godeke – Critical Surfaces – WAMHTS-1 workshop – DESY Hamburg

LTS intrinsic limits and dipole performanceField – temperature limitations and achieved dipole fields

• NbTi– 10.5 T– 80% of Hc2(1.8 K)

• Nb3Sn– 16 T– 65% of Hc2(4.5 K)

– 80% of Hc2(4.2 K) = 20 T?

– 80% of Hc2(1.8 K) = 22 T?

?

Can Nb3Sn be pushed higher?Godeke, et al., IEEE Trans. Appl. Supercond. 17 1149 (2007)Godeke, et al., J. Appl. Phys. 97 093909 (2005)

6May 23, 2014 A. Godeke – Critical Surfaces – WAMHTS-1 workshop – DESY Hamburg

Prospects for A15 quality and quantity (LTSW 2003)

25%

40%

10%

25%

Non-Cu: Ternary 64h/675°C

C.M. FischerMS thesisUW-Madison

PIT measured: ~140nm, 40% A15 in non-Cu

OI-ST: ~170nm, ~60% A15 (?), 24%Sn

PIT, all = best bit, ~140nm, 40% A15

PIT, + = 65% A15, ~140nm, ~24%Sn

PIT, + = 65% A15, ~140nm, all best bit

Sn C

ON

TEN

T

Nb 3S

n FR

ACTI

ON

AN

D S

n CO

NTE

NT

Nb 3S

n FR

ACTI

ON

Simulations on SMI-PIT (Nb,Ta)3Snfor Jc(12T,4.2K)

Godeke, LTSW (2003)Godeke, et al., J. Appl. Phys. 97 093909 (2005)

Achieved:Bruker PIT: 2600 A/mm2 (2008)OST IT: >3000 A/mm2 (since 2003)Hypertech IT: 3400 A/mm2 (2008)

Academic limit = 5000 A/mm2

Practical limit = 4000 A/mm2 (2003)3800 A/mm2 measured in OST-RRP (2011)

Unless pinning can be improved!

7May 23, 2014 A. Godeke – Critical Surfaces – WAMHTS-1 workshop – DESY Hamburg

Pinning optimizations?Comparison between NbTi and Nb3Sn JE(H)

Godeke, et al., IEEE Trans. Appl. Supercond. 17 1149 (2007)

Hc2(1.9K) Hc2(1.9K)

NbTi is more efficient approaching Hc2 → WHY?

8May 23, 2014 A. Godeke – Critical Surfaces – WAMHTS-1 workshop – DESY Hamburg

What determines Jc?Jc is determined by the achievable pinning force FP

• And thus by the average grain size in Nb3Sn

Grain boundaries are the mainpinning centers in Nb3Sn

Godeke, Supercond. Sci. Techn. 19, R68 (2006)After Fischer M.Sc. Thesis U. Wisconsin – Madison (2002)

No pinning

FL < FP

F L > F P

Ec

Jc

9May 23, 2014 A. Godeke – Critical Surfaces – WAMHTS-1 workshop – DESY Hamburg

What is an optimal grain size?Ideal is 1 pinning center per flux-line• Schematic: Cubic grains and flux-lines

• Ideal: dav = a0

Flux-line spacing a0 is field dependent• E.g. at 12 T a0 = (3/4)¼(0/0H)½ = 12 nm• Grain size in Nb3Sn wires → 100 – 200 nm

NbTi: Nanometer scale α-Ti precipitates

Nb3Sn: Grain size determines FP,MAX

Grain size determines FP(H)Grain size Nb3Sn factor 10 too large

Pinning in Nb-Ti is fully optimized

Lee, et al., Wire J. Int. Feb. 2003

Godeke, et al., IEEE Trans. Appl. Supercond. 17 1149 (2007)

Godeke and Dietderich, Unpublished

10May 23, 2014 A. Godeke – Critical Surfaces – WAMHTS-1 workshop – DESY Hamburg

What happens when Nb3Sn grains are refined?

Pinning force predicted gains• 12 T, 4.2 K Jc increases by factor 3.6 (!)

– A factor 3.4 is measured in thin films that were made and tested at LBNL

Critical current• 20 – 25 T field regime is opened up

– Much more efficient approaching Hc2

• Finer grains are emerging for wires…

Dietderich and Godeke, Cryogenics 48 331 (2008)Godeke, et al., Unpublished (2014)

Gains confirmed on thin films.How to do in wires is emerging…

3225 A/mm2

11,567 A/mm2E.g. Xu, et al., Appl. Phys. Lett. 104 (2014)

FP normalized to non-Cu Jc

11May 23, 2014 A. Godeke – Critical Surfaces – WAMHTS-1 workshop – DESY Hamburg

Wrap-up: What are the present dipole limits using LTS?

Nb-Ti reaches 80% of Hc2(T)• a-Ti precipitates form pinning centers

Nb3Sn reaches 65% of Hc2(T)• Inefficient high field pinning

Geometricand stresslimitations

Limits:10.5 T at 1.9 K for Nb-Ti16 T at 4.5 K for Nb3Sn

(or ~18 T at 1.9 K)

Needed for 20+T

1 flux lineper pinning site(core pinning)

10 flux linesper pinning site(collective pinning)

12May 23, 2014 A. Godeke – Critical Surfaces – WAMHTS-1 workshop – DESY Hamburg

Three paths towards higher dipole fields…1. Use High Tesla Superconductors for inserts

• Requires compatible magnet technology

2. Improve high field pinning in Nb3Sn• Requires clever material science; might reduce Hc2

3. Increase Hc2 in Nb3Sn• Is 35 T the limit of Nb3Sn Hc2(0)?

Cooley, et al., Appl. Phys. Lett. 88 142506 (2006)

Mentink, Ph.D. Thesis Univ. of Twente (2014)Godeke, Supercond. Sci. Techn. 19 R68 (2006)

?

Still more to expect from Nb3Sn!

13May 23, 2014 A. Godeke – Critical Surfaces – WAMHTS-1 workshop – DESY Hamburg

What dipole fields are possible using HTS?

Bi-2212 is available as a round wire…with significant potential!

Hc2(0) [T] Tc(0) [K]

NbTi 14 9.5

Nb3Sn 30 18

MgB2 3.5-35 32-40

YBa2Cu3O7 >100 93

Bi-2223 >100 108

Bi-2212 >100 95No field limitations

7 m of cable in a coil

14May 23, 2014 A. Godeke – Critical Surfaces – WAMHTS-1 workshop – DESY Hamburg

What about intrinsic limitations: Hc2(T) or Hirr(T)?Irreversibility field?

Irreversible magnetizationi.e. no more pinning!

Godeke, Ph.D. thesis, Univ. of Twente (2005)

Thermal de-pinningcan be severe

Hirr swamped byinhomogeneities

SC butno pinning

15May 23, 2014 A. Godeke – Critical Surfaces – WAMHTS-1 workshop – DESY Hamburg

Pinning defines practical phase boundaryActual phase boundaries HTS compared to LTS

Larbalestier, et al., Nat. Mater. 13 375 (2014)

Severe lackof pinning?

Saved by BZO?MacManus-Driscoll, et al., Nat. Mater. 3, 439 (2004)“Spontaneous” nano-distribution of nano-dots and nano-rods (Nb3Sn…?)

Large CPLarge T-margin Low NZPLow NZP

Bi-2212(Bi-2223)inserts

Do we really want good pinning for Bi-2212 inserts at 4.2 K?(We do want good pinning for Nb3Sn at 4.2 K!)

16May 23, 2014 A. Godeke – Critical Surfaces – WAMHTS-1 workshop – DESY Hamburg

So: Are there enough pinning sites in Bi-2212 at 4.2 K?

Miao, et al., IEEE Trans. Appl. Supercond. 15 2554 (2005)

1 bar reaction, 4 cm sample, 2005!

1 flux lineper pinning site(core pinning)

10 flux linesper pinning site(collective pinning)

Weak links

Linearity at high field suggests core pinning for strong links at 4.2 K=> Sufficient density of pinning sites?

17May 23, 2014 A. Godeke – Critical Surfaces – WAMHTS-1 workshop – DESY Hamburg

How do 4.2 K LTS and HTS pinning strengths compare?

2005 1 bar Bi-2212:• JE(45 T, 4.2 K) = 266 A/mm2

• 25 % Bi-2212: FP-max = 48 GN/m3 in Bi-2212

2012 100 bar (OP) Bi-2212• FP-max ≈ (650/450)*48 = 70 GN/m3 at 45 T

NbTi and REBCOTrociewitz, et al., 2005 NHMFL report 297

Nb3Sn ~5000 A/mm2 at 12 T– FP-max = 93 GN/m3 at 5 T

• “APC” Nb3Sn: FP-max = 215 GN/m3 at 12 T (?)

Applied magnetic field [ T ]

Nor

mal

ized

pin

ning

for

ce

4.2 K = 1700 GN/m3

(“flat” from 8 – 32 T)

OP Bi-2212 at 4.2 K seems comparable to present Nb3Sn but at much higher field

REBCO is still far better

Xu, et al., Appl. Phys. Lett. Mater. 2 046111 (2014)

18May 23, 2014 A. Godeke – Critical Surfaces – WAMHTS-1 workshop – DESY Hamburg

Comparison of superconductor fractionsNbTi: 50% of wire is SC

Nb3Sn (PIT): 20% of wire is SC

Bi-2212: 25% of wire is SC• NB: Price Ag = 630 $/kg, Nb (G1) = 350 $/kg

REBCO: 1% of tape is SC

25%40%10%25%

Non-Cu: Ternary 64h/675°C

C.M. FischerMS thesisUW-Madison

Larbalestier, et al., Nat. Mater. 13 375 (2014)

P.J. Lee, Applied Superconductivity Center (2003)

Larbalestier, et al., Nat. Mater. 13 375 (2014)

Larbalestier, et al., Nat. Mater. 13 375 (2014)

A lot to gain in HTS SC fraction

19May 23, 2014 A. Godeke – Critical Surfaces – WAMHTS-1 workshop – DESY Hamburg

Uniaxial strain (e.g. CTE differences in accelerator magnets)Generic plot of axial strain sensitivity• Not much happens for NbTi until >1% strain• Nb3Sn, YBCO are reversible in compression

– Cracks when loaded too far into tension• Bi-2212, Bi-2223 are crack dominated

– Limited reversible linear region

YBCO under strain depends on direction…

• …which is beneficial for certain cable layouts

Bottura and Godeke, Rev. Accel. Sci. Techn. 5 25 (2012)

Van der Laan, et al., Supercond. Sci. Techn. 24 115010 (2011)

20May 23, 2014 A. Godeke – Critical Surfaces – WAMHTS-1 workshop – DESY Hamburg

Uniaxial strain sensitivity of porous and dense Bi-22121 bar reacted, porous Bi-2212 wires• Generic behavior since mid 1990’s

100 bar reacted, dense Bi-2212 wiresGodeke, et al, submitted to Appl. Phys. Lett. (2014)

Godeke, et al, submitted to Appl. Phys. Lett. (2014)

Cheggour, et al., Supercond. Sci. Techn. 25 015001 (2011)

Densification does not improve strain margin

Will densification reduce buckling?

21May 23, 2014 A. Godeke – Critical Surfaces – WAMHTS-1 workshop – DESY Hamburg

Cables and stress (e.g. Lorentz load in accelerator magnets)Nb3Sn: 200 MPa

Bi-2223: Cables?REBCO:• 45 MPa (Roebel)• …• 1 GPa (CORC)

Dietderich, et al., IEEE Trans. Appl. Supercond. 11 3577 (2001)Hasegawa, et al., IEEE Trans. Appl. Supercond. 11 3034 (2001)

Transverse pressure measurements on modern Bi-2212 cables are urgently needed!(Experiments at 65 K and self-field are probably sufficient)

Bi-2212

?

about 0.3% strain ⊥

22May 23, 2014 A. Godeke – Critical Surfaces – WAMHTS-1 workshop – DESY Hamburg

SummaryNb3Sn is far from exhausted!• 20 T dipoles might be possible with engineered pinning center Nb3Sn

– Huge benefit: Utilization of proven Nb3Sn magnet technology!• R&D in academia needed for development (bring on your Ph.D. students!)

If only REBCO was a multi-filamentary round wire available in km length• Crazy pinning strength• Low strain sensitivity at 4.2 K• No reaction!• BUT…

If only Bi-2223 was available as cable without crack dominated strain dependence

Bi-2212 is a multi-filamentary round wire available in km length• Comparable pinning to Nb3Sn, but at 10x the field• Magnet designers need high JE cables and these have been proven• W&R: One order more complex than Nb3Sn, but proven• 0.3% strain margin is small but workable, transverse pressure still unknown