Update on the potential of Bi-2212 for high field magnet use · Nexans was the benchmark producer...
Transcript of Update on the potential of Bi-2212 for high field magnet use · Nexans was the benchmark producer...
April 10, 2018 1
D.C. LarbalestierWith major collaborators Eric Hellstrom, Jianyi Jiang, Fumitake Kametani and Ulf Trociewitz at
the NHMFL, Tengming Shen at LBNL, multiple PhD students and staff and many industrial
partners noted on the next slide
Update on the potential of Bi-2212 for high field magnet use
This work is supported by the UD Department of Energy for Conductor Development underGrant DE-SC0010421, by the National Science Foundation under NSF/DMR-1644779, by a grant from the National Institute of
Health under 1 R21 GM111302-01, by the state of Florida, and is also amplified by the U.S. Magnet Development Program (MDP).
1
April 10, 2018 22David Larbalestier FCC Workshop Amsterdam, April 9-14, 2018
2212 conductor development has been a Collaboratorium
E. S. Bosque, M. Brown**, D. S. Davis**, C. L.
English, G. Bradford, I. S. Hossain**, Jianyi
Jiang*, Y. Kim, E. Miller, G. E. Miller, Y. Oz**, J.
Lu, F. Kametani*, U. P. Trociewitz*, E. E.
Hellstrom*, and D. C. Larbalestier*
*PIs of the DOE-OHEP supported conductor
effort and the internal NSF solenoid-driven
development
**PhD students
At ASC-NHMFL-FSU Our external
collaboratorsTengming Shen (LBNL)
Yibing Huang and Hanping Miao (B-OST)
Marvis White, Riley Nesbit, Aixia Xu and
Andrew Hunt (nGimat)
Alex Otto (SMS)
Suvankar Sengupta and Rao Revur
(MetaMateria)
An integrated effort by DOE-OHEP, SBIR
program, NSF through NHMFL and some
key early support by CERN through
EUCARD2 – and now USMDP
April 10, 2018 3
Outline of the talk
• Conductor– Good powder– Good wires
• Twisted with low hysteretic loss made in single pieces with 1-2 km length at 1 mm and 0.8 mm diameter
• Coils– Reliable HT under 50 bar is now routine– Double insulation scheme valid on 1 km scale– We can reinforce conductor in winding pack to support >300 MPa in
tension – higher stress at 0.4% strain in view– A new reaction chamber (25 cm dia./1 m long) and magnet test bed (14
T/160 mm bore) is under active MDP discussion
David Larbalestier FCC Workshop Amsterdam, April 9-14, 2018
April 10, 2018 4
Key messages• No one is going to use any HTS
conductor if they can use Nb-Ti or Nb3Sn………..
• All HTS conductors are 7-15 times the (volumetric cost) of Nb3Sn.
• High Field or High Temperature use are key
Magnet builders much prefer round, twisted, isotropic
conductors with high conductivity normal metal around the
filaments
This is exactly what Bi-2212 offers – in many architectures
with Jc well exceeding Nb3Sn from about 12 T up
Nb-Ti
Nb3Sn
Bi-2212
David Larbalestier FCC Workshop Amsterdam, April 9-14, 2018
April 10, 2018 5
Two small US companies now making powder
MetaMateria
Use co-precipitation
nGimat
Use spray combustion to form 2212
Replicating the Nexans 521 composition
New 2212 powders are better than the best of Nexans*
No need for grinding agglomerated bulks
Particle size and uniformity better than NexansImportant contribution of CERN in context of EUCARD-2 for enabling Nexans final powder
production from which we learned much due to care of Mark Rikel in its manufacture
David Larbalestier FCC Workshop Amsterdam, April 9-14, 2018
April 10, 2018 6
Nexans was the benchmark producer for more than 10 yrs
JC(4.2K, 15T) = 4200 A/mm2
JE(4.2K, 15T) = 825 A/mm2
Still some small hard particles, but
less impact on filament quality in
drawn wires (filament merging).
Note for FCC magnet builders: Jc far exceeds the FCC specification (~1850 A/mm2 at 15 T) and with only
20% superconductor cross-section, very high conductivity Ag without any diffusion barrier that can break
during cabling and then degrade the RRR
Data by Bonura and
Senatore on our OP wires
RRR > 100
David Larbalestier FCC Workshop Amsterdam, April 9-14, 2018
April 10, 2018 7
Recent nGimat and MetaMateria powders established a new benchmark (2017) in standard B-OST wires
Powder LXB-43
Some Bi-2201 (white), AEC-14/24
(black), and Cu-free (next to 2201)
Powder LXB-52
No Cu-free and Bi-2201, with
large AEC-14/24 particlesv
New level of phase and particle size control with “simpler” and cheaper processes
David Larbalestier FCC Workshop Amsterdam, April 9-14, 2018
Jianyi Jiang
April 10, 2018 8
New record Jc for 0.8 mm wire from nGimat LXB 5270% increase in Jc over the last and best Nexans lot 87
2 3 4 5 6 7 8 9 10 11 12 13 14 15 163000
4000
5000
6000
7000
8000
9000
10000
11000
Nexans lot 87
nGimat LXB-52
Nexans lot 87
50 bar OP-HT
JC,
A/m
m2
B, T
nGimat LXB-52
IC(4.2K, 15T) = 658 A, JE(4.2K, 15T) = 1365A/mm2,
JC(4.2K, 15T) = 6860 A/mm2
Ic (0.8 mm wire) = 660 A at 15 T and 475 A at 30 T
with only 19% superconductor in the wire
Most recent optimized HT
June 2017 high field data
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32
400
600
800
1000 pmm170123-4
pmm170123-10
pmm170123-3
Critica
l C
urr
en
t, I
C [A
]
Field, B [T]
Reacted state
sub-bundles
(Some
interfilament
connections
but with low
hysteretic loss)
David Larbalestier FCC Workshop Amsterdam, April 9-14, 2018
Jianyi Jiang
April 10, 2018 9
Round wire Jc goodness plot
102
103
104
0 5 10 15 20 25 30 35
Wh
ole
Wir
e C
riti
cal C
urr
en
t D
en
sity
(A/m
m²,
4.2
K)
Applied Magnetic Field (T)
Bi-2212: 50 bar OP
Nb₃Sn: Internal Sn RRP®
Nb-Ti: LHC 1.9 K
Nb-Ti: LHC 4.2 K
Nb-Ti: High Field MRI 4.22 K
2212
Nb3Sn: High Jc
Maximal Je at 1.9 K for entire LHC NbTi
strand production (CERN-T. Boutboul '07). Reducing the temperature from 4.2 K
produces a ~3 T shift in Je for Nb-Ti
Compiled from
ASC'02 and ICMC'03 papers
(J. Parrell OI-ST)
Nb-Ti
4.2 K LHC insertion
quadrupole strand (Boutboul et al. 2006)
4.22 K High Field
MRI strand(Luvata)
August 30th 2017
Nb-Ti1.9 K
37×18 filament B-OST strand with NHMFL
50 bar Over-Pressure HT. J. J. Jiang et al.
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
1.5
1.6
1.7
pmm170123-4
pmm170123-10
pmm170123-3
I C / I
C(1
5 T
)
Field, B [T]
Normalized IC at (15 T)
Jc(H) falls off only slowly because Hirr >> 50 T: Jc (20T) is 83% and Jc (30T) 72% of Jc (15 T)
Michael Brown
Peter Lee
Best measured is 970 A/mm2 at 30 T
David Larbalestier FCC Workshop Amsterdam, April 9-14, 2018
April 10, 2018 10
High Field Data shows excellent Jc(H) scaling out to at least 31 T –connectivity is the dominant multiplier for Jc(H)
Michael Brown (PhD 2018)-2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34
200
400
600
800
1000 pmm170123-4
pmm170123-10
pmm170123-3
pmm160524-2
pmm160913-1
pmm160913-2
pmm160913-3
Critica
l C
urr
en
t, I
C [A
]
Field, B [T]
Fp (4K)
Field [T]
10 K/ 37 T15 K/ 17 T20 K/ 9 T
Viewed globally, wires of last few years with different powders have large range of JcNormalized Jc(H) relations are all very similar
Fpmax does not peak at 31 T and 4 K
David Larbalestier FCC Workshop Amsterdam, April 9-14, 2018
April 10, 2018 11
Twisting reduces coupling losses and does not degrade Jc
Yavuz Oz et al., ICMC-2017
27x7
38x18
Yavuz Oz et al., ICMC-2017
Hysteretic
losses are
comparable to
ITER Nb3Sn
wires
Jc is much
higher Physically uncoupled filaments show greater benefits from twisting
Samples Qs kJ/m3 QSC MJ/m3
Bi-2212 450-600 2.00-3.00
Nb3Sn 200-500 1.20-2.30
Bi-2223 1500 5.00
REBCO 3400 320.00
David Larbalestier FCC Workshop Amsterdam, April 9-14, 2018
April 10, 2018 12
50 bar Over Pressure (OP) removes filament size dependent properties(1.5-1.0 mm wire dia.) – and widens the HT window
OP removes filament size dependent
properties
(1.5-1.0 mm wire dia.) 1.0 mm/9.3 mm wire
Densified 1.4 mm wire,
12.9 µm filament dia.
Jiang et al., IEEE Trans. Appl. Supercond., 27, 4, 6400104 (2017) )
9 10 11 12 13 143000
3500
4000
4500
5000
5500
6000
JC(4.2 K, 5 T)
JC(4
.2 K
, 5
T),
A/m
m2
Densified Filament Diameter (mm)
1.0 1.1 1.2 1.3 1.4 1.5 mm
5
10
15
20
25
30
35
40
45
50
55
n value
n V
alu
e
Very similar filament size
distribution
Nexans powder, 2014
David Larbalestier FCC Workshop Amsterdam, April 9-14, 2018
April 10, 2018 13
Getting the conductor into magnets
• Must react coil at almost 900C, not just short samples – and at 20-100 bar overpressure (OP) – we use 50 bar now (1 bar O2, balance Ar), almost exclusively
• Coil insulation must withstand reaction after winding
• Strengthening methods must be available for low E (70 GPa) and low s0.2 (~100 MPa) and be applicable to coil manufacture
Overpressure increases Jc by 5-8 times and prevents internal pressure leakage
David Larbalestier FCC Workshop Amsterdam, April 9-14, 2018
DCL et al, Nature Mat 2014
April 10, 2018 14
Bi-2212 Conductor Strengthening is mimicking 2223
David Larbalestier FCC Workshop Amsterdam, April 9-14, 2018
425 MPa at 0.4% strain versus 118 MPa for 2212/Ag~3.5 fold benefit with 33% strip area (similar to SEI 2223 NX).
FSU Measurements on SMS tapes given OPHT at FSU
• Low aspect ratio applied to preserve electro-magnetic isotropy of wire
• Shown here is a wire bonded with reinforcement tape made by Solid Materials Solutions (SMS, Alex Otto)
• Work on reinforced conductor in coils is ongoing – as also is round wire strengthening
Alex Otto (SMS) and Michael Brown (FSU)
April 10, 2018 15
Insulation for 2212: hybrid TiO2 and mullite fiber
David Larbalestier FCC Workshop Amsterdam, April 9-14, 2018
• Dip coating process with TiO2 particles dispersed in a polymer
• Polymer burns off leaving strong ceramic layer
NHMFL Insulation line, Jun Lu, J. Levitan
• >1 km lengths now being coated reliably• Added alumino-silicate braid to improve
stand-off and winding pack integrity
Alumino-silicate braid
TiO2
coating
• Tests at LBNL of their coated and braided race-track coils have also shown higher leak resistance along with an improvement of transport properties
April 10, 2018 16
2212 test coils: thick (pups) and large diameter (Riky) for high stress
About 1 km of conductor each!
• We build Coils of various sizes from small bobbins to fully instrumented larger coils for in-field testing applying analysis lead design
• Models coils are confidence builders for larger high-field coils like Platypus and beyond (testing new concepts, models, manufacturing procedures)
(All to scale)cm70
60
50
40
30
20
10
8.7 cm
10.5 cm
13.3 cm
8.7 cm
David Larbalestier FCC Workshop Amsterdam, April 9-14, 2018
Youngjae Kim, Ernesto Bosque and Ulf Trociewitz
April 10, 2018 17
Modeling Coil Behavior
David Larbalestier FCC Workshop Amsterdam, April 9-14, 2018
LTS Outsert at 8 T, HTS (Pup) coil at 4 T
(400 A)
Source:Jphi
. Bz. R [MPa]
AzimuthalStress Map [MPa]
AzimuthalPercent Strain
[%]
Goals: Understand and control performance limits of our 2212 coils • Model coils with conductor resolution• Compare with experiments
Riky concept: test coils with large bore size to generate high hoop stresses, TiO2 and braid insulated wire, varying reinforcement
Pup concept: test coils with same ID and OD as Platypus to allow observation of strain gradients across winding pack, TiO2 and braid insulated wire, varying reinforcement
Pup coil in 8 T outsert
Ernesto Bosque, Youngjae Kim,and Ulf Trociewitz
April 10, 2018 18
RIKY 3 test coil
David Larbalestier FCC Workshop Amsterdam, April 9-14, 2018
• FEA modeling is predicting coil performance well: 348 A predicted for onset of damage (Bjoerstad data confirmed by us (Michael Brown)
• The first trip was observed at 350 A compared to expected of 348 A!
• Very clearly not an Ic transition but rather a strain induced trip
• In no-reinforced condition the outer layer would have experienced a stress of ~300 MPa: reinforcement is effective
• Showed saturation of Ic retention on subsequent quenches
• Below onset of thermal runaway the coil could be load-cycled many times
Predicted peak strain expected for given Riky-3 operating currents are shown as blue line plotted against Ic/Ic0 retention curve based on data by C. Scheuerlein*, CERN
*R. Bjoerstad, C. Scheuerlein, M. Rikel et al., Supercond. Sci. Technol. 28 (2015) 062002
Predicted critical strain trip at 350 A:
Youngjae Kim, Ernesto Bosque and Ulf Trociewitz
April 10, 2018 19
A new HTS test bed
David Larbalestier FCC Workshop Amsterdam, April 9-14, 2018
Platypus coil
300 mm
• Is ideal test bed to explore performance limits of HTS inserts for improving performance predictions
• Magnet provides sufficient space to run 2212, 2223 or REBCO coils up to 30 T
14 T Magnet by Cryogenic LTD to be commissioned soon at ASC:• Magnet bore 161 mm• 128 ppm homogeneity (1 cm DSV)
without compensators or shims
Lamar English, Ernesto Bosque, Eric Hellstrom and Ulf Trociewitz
April 10, 2018 20
• First-of-its-kind pressurized 6-zone furnace built by Deltech
• Currently largest OP furnace available for Bi-2212 heat treatment
• Design parameters: 900 °C, 100 atm total pressure, 500 mm hot zone – currently use 50 atm and the hot zone is ~430 mm (130 mm dia.)
• This furnace required extensive in-house engineering to make it a robust, reliable equipment
• Successful heat treatment of many coils
Our High Pressure Furnace is the Work Horse for Bi-2212 Coil Heat Treatment
Outer wall dia: 324 mmWorking Hot Zone dia: 130 mmWorking Hot Zone height: 430 mm
Lamar English, Ernesto Bosque, Eric Hellstrom and Ulf Trociewitz
April 10, 2018 21
Hot Zone is Very Uniform Even with Large Thermal Mass – Platypus-II Dummy Coil
Zone1.5SS|707A|988A/mm^2
Zone2.4SS|697A|973A/mm^2
Zone3.0SS|697A|972A/mm^2
Zone1.5 – 707 A, 988 A/mm2
Zone2.5 – 697 A, 973 A/mm2
Zone3.0 – 697 A, 973 A/mm2
Placement and Ic
,Jc values of short samples
Stack of metal rings to simulated thermal mass of Platypus –II coil
Thermocouples monitoring locations on dummy
Dummy coil with three 18 layer, 3 turn test sections to verify calibration and OPHT procedure
21
Ernesto Bosque
April 10, 2018 22
OPHT Yields Uniform Transport from Top to Bottom and from Inside to Outside
988 973 972
721690 684
677 676 661
500
600
700
800
900
1000
1100
1 2 3 4
J_e[A/m
m^2]
DeltechFurnaceZone#
PYP-IIDummyShortSampleJ_e
TestIDJ_e
TestODJ_e
WiresPYP-II dummy: Pmm 140606SS: Pmm 160301
ID:
OD:
SS:
Zone1.5SS|707A|988A/mm^2
Zone2.4SS|697A|973A/mm^2
Zone3.0SS|697A|972A/mm^2
Zone 1.5 – 707 A, 988 A/mm2
Zone 2.5 – 697 A, 973 A/mm2
Zone 3.0 – 697 A, 973 A/mm2
Placement and Ic ,Jc values of short samples
Four important outcomes:• Coils performed consistently throughout the dummy stack• No oxygen deficiency in inner layers• Bottom windings survived compressive force of (simulated)
winding pack• We can handle the thermal mass and weight of Platypus-II
22
Ernesto Bosque and Lamar English
April 10, 2018 2323
LBNL HTS (2212) subscale magnet program topped with new RC-05 results
LBNL 17-strand Rutherford cable
Mullite braided insulation
Subscale coils allow fast-turnaround test of cable and magnet-relevant technologies.
23
LBNL RC-1,2,3,5 in FSU OP furnace
2212
RC5 – 3.4 T, Jcable = 740 A/mm2, Jwire = 940 A/mm2.Made possible by OPHT at FSU, wire by Bruker OST, and powder by nGimat.
Tengming Shen, Ernesto Bosque
and Lamar English
April 10, 2018 24
RC5 – E-J characteristics defined with a stair-case I(t) run. Many resistive, easy to detect signals revealed before the quench
(global superconducting transition driven by continuous joule heating)
100 200 300 400 500 600
1E-7
1E-6
1E-5
1E-4
1E-3
Co
il v
olt
ag
es
(V
)
Time (sec)
Vete
VL1
VL2
100 200 300 400 500 600
2000
4000
6000
8000S
hu
nt
cu
rre
nt
(A)
Time (sec)
Current (A)
100 200 300 400 500 600
0
50
100
150
200
250
300
350
400
Tu
rn v
olt
ag
es (m
V)
Time (sec)
L1-T1'
L1-T2'
L1-T3'
L1-T4'
L1-T5'
L1-T6'
Ramp-turn
L2-T6'
L2-T5'
L2-T4'
L2-T3'
L2-T2'
L1-T1 having voltage rises due to heating from splices.
Quench
RC5
5500 6000 6500 7000 7500 8000
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
L1-T6'
ramp turn
Fit, Ic=8282 A (1.0 mV/cm), n=28
Ele
ctr
ic f
ield
(m
V/c
m)
Current (A)
7925 A (DCCT), 180 s no quench.
8128 A (DCCT), a few sec quench.
David Larbalestier FCC Workshop Amsterdam, April 9-14, 2018
Tengming Shen LBNL
April 10, 2018 25
The global superconducting transition - Multiple turns turning normal makes possible easy quench detection.
500 502 504 506 508 510 512 514 516 518 520
-10
0
10
20
30
40
50
60
70
80
90
100
Tu
rn v
olt
ag
es
(m
V)
Time (sec)
L2-T6'
L2-T5'
L2-T4'
L2-T3'
L2-T2'
L1-T1'
L1-T2'
L1-T3'
L1-T4'
L1-T5'
Ramp-turn
L1-T6'
500 502 504 506 508 510 512 514 516 518 520
1E-6
1E-5
1E-4
1E-3
Co
il v
olt
ag
es
(V
)
Time (sec)
Vete
VL1
VL2
Quench
RC5
David Larbalestier FCC Workshop Amsterdam, April 9-14, 2018
Tengming Shen LBNL
April 10, 2018 26
Plans for a Larger OP Furnace
• Current furnace:– Working Hot Zone height: 430 mm– Working Hot Zone dia: 130 mm
• Due to a demand of heat treating larger coils (race-track, CCT coils with LBNL, and solenoids) we started planning for a bigger OP furnace
• Pressure range will be 50 bar• New furnace specs:
– 1 m working hot zone height (900oC)– 266 mm working dia.
• Major components to be sourced through commercial suppliers, assembly at the Lab, Labviewcontrolled
26
1 m long homogeneous zone
Lamar English, Ernesto Bosque, Eric Hellstrom and Ulf Trociewitz
April 10, 2018 2727David Larbalestier FCC Workshop Amsterdam, April 9-14, 2018
The conductor technology is mature at present scale
Single 10 kg piece lengths: 2.4 km at 0.8 mm, ~1 km at 1 mm
Twisted, multifilament, isotropic, low loss conductor with high RRR (> 100) and no need for diffusion barrier
Insulation technology for single wires is good
Reaction under 50 bar to 13 cm dia. x 40 cm length
Strain control at ~300 MPa demonstrated in coils with strengthened wires at >400 MPa
We have successfully replaced the classical powder producer Nexans with two new suppliers, nGimat and MetaMateria
Present status: 2212
April 10, 2018 28
Thank You!
28