Beyond Niobium SRF materials Thin films and new

  • Slides: 27
Download presentation
Beyond Niobium SRF materials: Thin films and new multilayer superconductors Claire Antoine SACM, Centre

Beyond Niobium SRF materials: Thin films and new multilayer superconductors Claire Antoine SACM, Centre d'Etudes de Saclay 91191 Gif-sur-Yvette, France Eu. CARD-2 is co-funded by the partners and the European Commission under Capacities 7 th Framework Programme, Grant Agreement 312453

Ultimate limits in SRF-1 • Superconductivity only needed inside : – Thickness ~ <

Ultimate limits in SRF-1 • Superconductivity only needed inside : – Thickness ~ < 1 µm => thin films » (onto a thermally conductive, mechanically resistant material, e. g. Cu) • Issues : getting “defect free” superconductor • Q 0 1/RS TC • Eacc HRF • Limit = transition of the SC material @ Hpeak – Which transition should we consider!? ? ? CZ. Antoine HMax. Hpeak 0 B field mapping in an elliptical cavity 2

Ultimate limits in SRF-2 Mixed state w. Vortex • SC phase diagram – All

Ultimate limits in SRF-2 Mixed state w. Vortex • SC phase diagram – All SC applications except SRF: mixed state w. vortex H (i. e. N. cond. flux line + screening currents) HC 2 » Vortices dissipate in RF ! – SRF => Meissner state mandatory ! – HC 1 = limit Meissner/mixed state – Nb highest HC 1 (180 m. T) – « Superheating field » (? ) : Metastable state favorized by H// to surface » Difficult to get in real life ! CZ. Antoine HC 1 Meissner state TC T Screening current over l, no magnetic field deeper 3

Vortex penetration • – – Ideal case Field // surface, => surface barrier (boundary

Vortex penetration • – – Ideal case Field // surface, => surface barrier (boundary conditions) Field // surface start to enter SC @ HSH > HC 1 @ H HSH Vortex oscillate in RF dissipations Most favorable SC : Nb 3 Sn, Mg. B 2 (high TC, high HSH) CZ. Antoine 4

HSHNb 3 Sn (~ 400 m. T @ 0 K) Nb 3 Sn: reaching

HSHNb 3 Sn (~ 400 m. T @ 0 K) Nb 3 Sn: reaching HSH ? Nb HC 1 (170 à 200 m. T) Cornell, 1997 pulsed Hays. "Measuring the RF critical field of Pb, Nb. Sn". in SRF 97. 1997. Recent results from Cornell CW (Posen, 2015 (17 MV/m)) Hi HC 1 Nb 3 Sn (~27 m. T) (50 m. T pour le bulk) TC Dissipations : • Vortices enter more easily at low temperature – @ T~TC : H low=> low dissipations => easy thermal stabilisation – @T<<TC : H high=> even small defects => vortex penetration and high dissipation • Reduce defect density (but which ones !? ) CZ. Antoine 5

Vortex penetration • – – Field // surface, => surface barrier (Bean Livingston) Vortex

Vortex penetration • – – Field // surface, => surface barrier (Bean Livingston) Vortex // surface start to enter @ HSH > HC 1 Vortex oscillate in RF dissipations Most favorable SC : Nb 3 Sn, Mg. B 2 (high TC, high HSH) • – – – Ideal case Real life: defects at surface Early vortex penetration (bundle) @ HC 1 (or less !) Formation of current loops Avalanche Oscillations in RF => dissipations What kind of defects do we fear ? ? ? CZ. Antoine 6

 Vortices: Avalanche penetration • ~100 µm in 1 ns (~RF period)= • Compare

Vortices: Avalanche penetration • ~100 µm in 1 ns (~RF period)= • Compare with l penetration depth) – – (field Nb : ~ 40 nm Mg. B 2 ~ 200 nm Mg. B 2 example http: //www. nature. com/srep/2012/121126/srep 00886/full/srep 00886. html? messageglobal=remove&WT. ec_id=SREP-20121127 CZ. Antoine 7

Vortex penetration • – – Field // surface, => surface barrier (Bean Livingston) Vortex

Vortex penetration • – – Field // surface, => surface barrier (Bean Livingston) Vortex // surface start to enter @ HSH > HC 1 Vortex oscillate in RF dissipations Most favorable SC : Nb 3 Sn, Mg. B 2 (high TC, high HSH) • – – – Defect at surface Early vortex penetration (bundle) @ HC 1 (or less ? ) Formation of current loops Avalanche Oscillations in RF => dissipations What kind of defects do we fear ? ? ? • – – – Ideal case Dielectric layer Small vortex (short -> low dissipation) Quickly coalesce (w. RF) Blocks avalanche penetration => Multilayer concept for RF application Most favorable SC : Nb 3 Sn, Mg. B 2, Nb. N… CZ. Antoine 8

Multilayer concept • Surface screening and low Rs – Thin SC films. d< l

Multilayer concept • Surface screening and low Rs – Thin SC films. d< l => Artificial enhancement of HC 1* » Thin layers stand high fields without vortex nucleation » Partial screening of Happlied – Niobium surface screening: allows higher field in the cavity – TCNb. N >> TCNb => RSNb. N << RSNb => Q 0 multi >> Q 0 Nb Happlied B(m. T) 100 Q 0 1 E+12 200 HNb 1 E+11 Outside wall 1 E+10 Nb 1 E+09 0 CZ. Antoine Cavity's internal surface → 20 40 Eacc (MV/m) 60 I-S- ** * In theory 20 nm Nb. N : HC 1 x ~200 ** Simplified model from Gurevich 9

ML developments in WP 12 • Optimization of ML structure – Series of Nb.

ML developments in WP 12 • Optimization of ML structure – Series of Nb. N single layer/Mg. O/Nb samples » Deposited by reactive magnetron sputtering on silicon substrate (collaboration Grenoble INP and CEA INAC) » Nb. N chosen because well mastered /SC electronics – Comparison w. recent theoretical developments • Development of specific sample measurement tools – Properties of thin films in SRF operation conditions cannot be done w. conventional techniques CZ. Antoine 10

 Nb – Insulator – Nb. N model Advanced model: only for single layer;

Nb – Insulator – Nb. N model Advanced model: only for single layer; includes: X * SC substrate Nb 250 m. T X Insulator – Boundary conditions – Role of Nb sublayer Thin SC layer Nb. N X X Thin SC layer thickness nm • X • I layer thickness nm Prediction for Nb. N (TC ∼ 15 K, l = 200 nm) – T. Kubo (2014)3 ~140 nm – A. Gurevich (2015)4 ~160 nm • WP 12. 2 (subtask 2) experimentals: – Nb. N not most favorable on paper… – but “easy” to make (cf SC electronics) Hmax optimum ∼ 250 m. T which is higher than of thick Nb (170 m. T) 2 C. Z. Antoine, et al. APL 102, 102603 (2013). 3 T. Kubo et al, Appl. Phys. Lett. 104, 032603 (2014). 4 A. Gurevich, AIP Advances 5, 017112 (2015). CZ. Antoine 11

Hc 1 Measurement: need for Local Magnetometer • • Develop new SCs multilayers at

Hc 1 Measurement: need for Local Magnetometer • • Develop new SCs multilayers at higher fields => Need for specific characterization tools Conventional Magnetometer (SQUID) gives ambiguous results: Happ – Uniform field around the sample – Demagnetization (orientation, edge, shape) effects – Exact local field configuration not known Sample SQUID magnetometer principle Happ Sample Local magnetometer principle CZ. Antoine 12

3 rd harmonic measurement of HC 1 /HSH • Building a setup ~operating conditions

3 rd harmonic measurement of HC 1 /HSH • Building a setup ~operating conditions for SRF (2 K-20 K; H >> 150 m. T) – (tbc existing facilities 6 : > 4, 5 K or 70 K and Bmax ~15 -20 m. T) – Magnet size << sample size (infinite plane approx. ) – Field decreases quickly away from the coil – Measurement of HC 1 on sample without edge/demagnetization effect – Exploring new SCs /multilayers at accelerator operating condition Coil multiturns T↑ Superconductor Meissner state Mixed state (Magnetic mirror) 5 J. H. Claassen, et al. Rev. Sci. Instrum, Vol. 62, 4 (1991). CZ. Antoine (non linear behavior) 6 M. Aurino, et al. , Journal of Applied Physics, 98. 123901 (2005). 13

Screening Power of Nb. N • Series of Nb. N single layer/Mg. O/Nb –

Screening Power of Nb. N • Series of Nb. N single layer/Mg. O/Nb – Deposited by reactive magnetron sputtering on silicon substrate (collaboration Grenoble INP and CEA INAC) – Insulator = Mg. O – Thick Nb layer to mimic bulk Nb Nb. N 200 nm H (m. T) Increasing H Nb. N 200 nm CZ. Antoine 14

What do we measure ? H = 7 m. T • Determination of Hc

What do we measure ? H = 7 m. T • Determination of Hc 1 – Low field => one transition – High field => two transitions » 1 st transition with low dissipation » 2 nd transition very strong dissipation • Why do we have two transitions ? H = 25 m. T 1 st transition 2 nd transition CZ. Antoine Nb. N 200 nm 15

Screening Power of multilayers • Why do we have two transitions ? H app

Screening Power of multilayers • Why do we have two transitions ? H app Defect Nb. N 200 nm, H = 53 m. T § Thin SC layer Nb. N § Insulator Mg. O § Thick SC layer Nb Field lines First transition Vortex avalanches progressive vortex penetration Second transition 7 B. CZ. Antoine Bean and J. D. Livingston, Phys. Rev. Lett. 12, 14 (1964). – H // surface => surface barrier 7 – A defect locally weakens the surface barrier – 1 st transition, vortex blocked by the insulator ~100 nm => low dissipation. – 2 nd transition, propagation of vortex avalanches (~100 µm) => high dissipation. – Dielectric layer = efficient protection !!! 16

Thin films, multilayer R&D • Scientifically: – Very challenging upstream, discovery, R&D – Efficiency

Thin films, multilayer R&D • Scientifically: – Very challenging upstream, discovery, R&D – Efficiency of multilayer concept demonstrated » Field enhancement => higher SRF performances » Results close from theory => will help optimization – Protection against “avalanches” => can accommodate defective (realistic) material • Future : – EUCARD 3 (ARIES) covers only a little of the work to be done – Other funding sources needs to be found • Practically: – Many difficulties to (propose and) follow a realistic schedule within EUCARD framework – The foreseen program and the collaborations started will be pursued beyond the end of EUCARD 2 CZ. Antoine 17

Team at CEA Saclay Thank you for your attention Claire ANTOINE Muhammad ABURAS Special

Team at CEA Saclay Thank you for your attention Claire ANTOINE Muhammad ABURAS Special CZ. Antoine thanks go to everyone participated in this work Aurelien FOUR 18

 Costs elements Cu SC • coûts ↑ avec C. U. – Optimum Eacc

Costs elements Cu SC • coûts ↑ avec C. U. – Optimum Eacc » Low Eacc => longer accelerator => fab. costs↑ » High Eacc => RF cost ↑ for Cu & cryogenics ↑ for SC Cost (investment + use) – Other example: CLIC vs ILC (e+/e- collider) (given E, I) » ILC : C. U. = 0, 5 % @ 1, 3 GHz » CLIC : C. U. = 0, 001 % @ 12 GHz – Linac costs » ~ 1/3 tunnel, building » ~ 1/3 niobium, cryo » ~ 1/3 RF, beam control Eacc(MV/m) 100 50 Duty cycle 0, 1 % CZ. Antoine 1% 100 % Log scale !!! 19

 elements de coûts Cu SC • coûts ↑ avec C. U. – champ

elements de coûts Cu SC • coûts ↑ avec C. U. – champ accel. Optimum » Faible champ => accélérateur + long => coûts ↑ » Fort champ => coûts RF ↑ pour cuivre et coûts cryogéniques ↑ pour supra Cost (investment + use) (given E, I) – Autre exemple : CLIC vs ILC (collisionneurs e+/e- usines à Higgs) » ILC : C. U. = 0, 5 % @ 1, 3 GHz » CLIC : C. U. = 0, 001 % @ 12 GHz – Coûts linac Eacc(MV/m) 100 » ~ 1/3 tunnel, BTP » ~ 1/3 niobium, cryo » ~ 1/3 RF, contrôles faisceau 50 Duty cycle 0, 1 % CZ. Antoine 1% 100 % Log scale !!! 20

Hc 1 Measurement, a Local Magnetometer copper rod (thermalization of electrical wires) spring coil

Hc 1 Measurement, a Local Magnetometer copper rod (thermalization of electrical wires) spring coil support (high conductivity copper) glass bead coil sample thermal braid High conductivity copper plate heating wire temperature sensor steel rods sample support (high conductivity copper) Schematic of local magnetometer CZ. Antoine 21

How this magnetometer works ? Hc 1 Measurement, a Local Magnetometer q Works have

How this magnetometer works ? Hc 1 Measurement, a Local Magnetometer q Works have been beginning in 2010 2014 2016 Experimental setup CZ. Antoine 22 8/15

How this magnetometer works ? Hc 1 Measurement, a Local Magnetometer q Works have

How this magnetometer works ? Hc 1 Measurement, a Local Magnetometer q Works have been beginning in 2010 Insert CZ. Antoine Cryostat Measurement devices 23 8/15

Behind every success, a lot of failures Hc 1 Measurement, a Local Magnetometer q

Behind every success, a lot of failures Hc 1 Measurement, a Local Magnetometer q Many efforts were achieved to overcome some difficulties q End of 2016, first successful measurement Finally, a measurement done correctly until ~100 m. T HC 1(T) 100 SL 25 nm 160 70 SL 100 nm 140 60 Nb Ref 80 50 Nb mono. X Support 3 30 60 20 40 10 20 0 10 T(K) 15 First acceptable results CZ. Antoine Nb support 5 : 1+ SD plane HC 1 (T) 100 80 5 Nb mono. X support 4 120 40 0 Hc 1 (T) 200 180 90 Field (m. T) HC 1(m. T) 20 0 0. 00 5. 00 T(K) 10. 00 15. 00 Calibration with a monocrystalline Nb 24 11/15

Behind every success, a lot of failures Hc 1 Measurement, a Local Magnetometer q

Behind every success, a lot of failures Hc 1 Measurement, a Local Magnetometer q Many efforts were achieved to overcome some difficulties Problems ! CZ. Antoine Modifications Thermal stabilizations Add some copper braids Calibration (important shift) The sample holder 25 10/15

Conclusion and Perspectives q A local magnetometer has proven to be effective at measuring

Conclusion and Perspectives q A local magnetometer has proven to be effective at measuring vortex penetration in conditions close to cavities operating condition. q We have shown a very promising behavior of Nb. N layers q S-I-S multilayers provide best protection of cavities against local penetration of vortices q Overcome Nb monopoly by higher Hc 1 superconductors multilayers is possible q Sample gives results close to theory : optimization can be done theoretically q Deposition methods inside cavities needs to be developed Perspectives q Enhancement of the maximum magnetic field applied on the sample, we hope to reach > 250 m. T by: § Replacement the coil by a ferrite core inductor § Novel thermal design of the experimental setup Coil Ferrite q Study other superconductors multilayers at higher fields. Superconductor CZ. Antoine 26 15/15

 • Click to edit Master text styles – Second level • Third level

• Click to edit Master text styles – Second level • Third level – Fourth level » Fifth level CZ. Antoine 27