MATERIALS FOR SUPERCONDUCTING ACCELERATORS: BEYOND BULK Nb - C.Z. Antoine - Indico
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
MATERIALS FOR SUPERCONDUCTING ACCELERATORS: BEYOND BULK Nb SRF 2021 Tutorials C.Z. Antoine + some material gathered by A.M. Valante-Feliciano
Preamble: - many slides will not be presented but will be available if you download the presentation. - if not mentioned in the slide, references are available in the comment page 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | 2
THOUSANDS OF SUPERCONDUCTORS … Thousands of SC exist, ~10 are currently used for applications, only bulk Nb works well for SRF !!! Source: © 2018 Macmillan Publishers Limited, part of Springer Nature 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 4
IDEAL SRF MATERIAL: TAILORED FOR APPS High RRR not required for superconductivity but for thermal stabilization in case of defects Today RF side Future ? Protective Layer e.g. oxide layer = diffusion barrier, or Natural oxide, not low secondary emission layer (↓ MP) stable @ T>150°C “RF” layer Optimized for superconductivity: Low surface resistance (medium m,f,p,) High RRR Nb, and/or High HP (high HC1 high HSH) optimized for thermal transfer, in case of dissipation “thermal conducting” layer Optimized for thermal transfer (high RRR, high m,f,p., not necessarily Nb) “External” layer Natural oxide, not optimized to reduce Optimized for thermal transfer (low Kapitza resistance Kapitza resistance, e,g, anodization) He side 1000 RBCS ( nW ) 4,5 K “RF” layer optimized for superconductivity: 900 800 Low surface resistance (medium m.f.p.) 700 and/or* 600 High HP (high HC1 or high HSH) 500 400 *Depends on the application 1 10 100 l ( nm ) 1000 (l RRR) 10000 2021/06/24 C.Z. Antoine -SRF 2021 |tutorial 5 - Beyond Nb
ULTIMATE LIMITS IN SRF-1 Niobium superconducting radiofrequency cavities Performances Eacc HRF HPeak Q0 ( 1/RS) TC => Nb3Sn, MgB2, NbN… (but not YBCO) Limit = magnetic transition of the SC material @ Hpeak HMax. Superconductivity only needed inside : Thickness ~10 l, ~ < 1 µm => thin films (onto a thermally conductive, mechanically resistant material, e.g. Cu) Today : Thin films exhibit too many defects Only Bulk Nb has high SRF performances (high Q0 and high Eacc) 0 Issues : getting “defect free” superconductors H field mapping in an elliptical cavity (Yes but not all defects are detrimental... See doping !) 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 6
HIGH Q0, EACC IN SRF => MEISSNER STATE ! Mixed state w.Vortex SC phase diagram (i.e. N. cond. flux line + screening currents) All SC applications except SRF: mixed state w. vortex ● Vortices dissipate in RF ! H SRF => Meissner state mandatory ! HC2 Limit ? HC1 = limit Meissner/mixed state HC ● Nb: highest HC1 (180 mT) Or HC1 HSH “Superheating field”: Metastable state favored by H // to surface ● Difficult to get in real life ! Surface resistance: 2 RBCS A(l , F , l, n ) 4 L e / kT Meissner state TC T T Screening current over l, High Tc is better no magnetic field deeper T
VORTEX PENETRATION WITH B // Surface barrier J (Bean & Livingston, 1964) Boundary condition. (J = 0) ≡ “image” vortices H0 Supercurrent tends to push Vx inside Image antivortex tends to pull it out Before entering the material Vx have to cross a surface barrier: Image Vortex Vx thermodynamic Potential : − J = 0 = 0 0 ൗ − 2 + 1 − 0 l Image G Meissner H0 < Hc1 “Ideal surface” Barrier disappears only at HSH~HC >HC1 H0 = Hc1 Rationale used to predict SRF limits BUT H0 > Hc1 If localized defect w.: H c ≪ Hc (or Tc ≪ H0 = Hc T c ) => early penetration of 1 or several Vx there ↗ 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 8
SUPERCONDUCTORS FOR SRF ? n µ0HC1 µ0HC2 µ0HC µ0HSH l Material TC (K) Type (µWcm) (mT)* (mT)* (mT)* (mT)* (nm)* (nm)* (meV) Pb 7,1 n.a. n.a. 80 48 I Nb 9,22 2 170 400 200 219 40 28 1.5 II NbN 17,1 70 20 15 000 230 214 200-350 11 000 100-200 80-160 210-420 5,4 NbTiN 17,3 35 30 150-200
WHAT IS THE ACTUAL LIMIT (HP/HC1/HSH) ? Avalanche penetration/flux jumps ~100 µm in 1 ns (~RF period) Compare with l (field penetration depth) Nb : ~ 40 nm MgB2 ~ 200 nm Avalanche : high RF dissipation MgB2: http://www.nature.com/srep/2012/121126/srep00886/full/srep00886.html?message-global=remove&WT.ec_id=SREP-20121127 In real world, cavities behavior is dominated by a few number of defects It is very important to measure the penetration field of samples in realistic conditions H H H// HSH HSH H// HC1 HC1 H H T2C T2 T2C T2 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 11
HSHNb3Sn (~ 400 mT @ 0 K) EFFECTS OF LOCAL DEFECTS Hays. "Measuring the RF critical field of Pb, Nb, NbSn". in SRF 97. 1997. Nb HC1 (170 to 200 mT) Nb3Sn Cornell, 1997 pulsed vs Nb Recent results from Cornell & FNAL CW Posen, 2019 (23 MV/m) 2015 (17 MV/m)) = 0 ቈ − ( ) HC1Nb3Sn (~27mT) (50 mT pour le bulk) Dissipations : ~ 2K for Nb 3,5 K for Nb3Sn Vortices enter more easily at lower temperature (counter intuitive !)? @ T~TC : H is low => low dissipations => easy to thermally stabilize @ T high dissipations => Favors flux jumps => We have to reduce defect density (yes but which ones?) 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 13
CHALLENGES TO FACE ON THE ROUTE TOWARD OTHER SUPERCONDUCTORS: GENERALITIES C.Z. Antoine -SRF 2021 |tutorial PAGE- 14 Beyond Nb
GENERAL ISSUES WITH SCs Needed: high TC, high HSH EX. : Grain boundaries (by defect high HC1) Advantages of niobium: pure metal. Highest TC of metallic SC, HC1 Easy to form Uniform composition, no phase transition in the domain of interest Very large : makes it less sensitive to small crystalline defects (e.g. GB) Issues with alloyed, metallic SC compounds (e.g. NbTi) Some nm Compare with Higher TCs, but smaller HC1, Still relatively easy to form (harder) Usually several phases, not all of them SC Risk of non homogeneity Issues with non metallic SC compounds Higher TCs, but smaller HC1, Brittle, no forming is possible, only films (OK for SRF, but a more complex fabrication route is needed) Usually several phases, not all of them SC Risk of non homogeneity http://www.phys.ufl.edu/~pjh/grain-boundry.html Sometimes local disorder => ≠ local composition, possibly non SC Weak links e.g. NC grain boundaries = main reason why HTC do not apply in SRF . 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 15
THIN FILMS DEPOSITION Nb : l~50 nm => only a few 100s nm of SC necessary (the remaining thickness= mechanical support) => Make thin films ! Advantages Thermal stability (substrate cavity = copper, Aluminum,… W) Cost Opens route to innovative materials Optimization of RBCS possible (e.g. by playing with m.f.p) Disadvantages Fabrication and surface preparation of substrate (at least) as difficult as for bulk Nb Steep Q0 decrease often observed by increase of RF field (sputtered niobium films, improved lately) Deposition of innovative materials is very difficult (large parameters space to be explored) Most of the known SC have been optimized for wire applications (low HC1, defects, pinning centers… ) => most of the literature recipes are not fitted for SRF application 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 16
DEPOSITION TECHNIQUES: 3 MAJORS FAMILIES Physical deposition techniques (PVD, MS, DS…) line of sight techniques issues: getting uniform thickness/structure internal stress and adhesion limited for complex geometry Thermal diffusion films limited compositions available non uniform composition issues (S shaped diffusion front, differential diffusion rate with substrate grain orientation) Chemical techniques CVD, ALD conformational even in complex shape very quick for large surfaces issues: get the proper crystalline structure 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | 17
SEARCH FOR BETTER STRUCTURE Generalized Structure Zone Diagram Structure zone model (from A. Anders) derived from Thornton’s diagram, 1974 Energetic deposition (HPIMS, CED, VAD…) => Bulk like films Magnetron sputtering => A lot of defects Cu limits annealing temperature © Andre Anders, 2010 11 recrystallization A. Anders, Thin Solid Films 518, 4087 (2010). Unfortunately, more “bulk-like” Nb Films gave disappointing RF results at the beginning . Not fully understood yet 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb 19
Nb/ Cu: example of the issues when dealing with thin fims 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | 21
SPUTTERED Nb FILMS The only Nb films deployed in accelerators were made by magnetron or diode sputtering (CERN). Reached relatively low surface fields => Eacc ~ 5 MV/m. Exponential slope in Rs and Q0 1.5 GHz Nb/Cu cavities, sputtered w/ Kr @ 1.7 K (Q0=295/Rs) Bulk Nb Nb films – sputtered 1.5 GHz Nb/Cu LEP II 350MHz Nb/Cu (4.2K) Possible origin of the slope Depinning of trapped flux Low HC1 Early vortex penetration due to roughness CERN 2000 Current concentration due to porosities (generating local electrical field) 2021/06/24 C.Z. Antoine -SRF 2021 22 tutorial - Beyond Nb
EXAMPLE OF QUALITY ISSUES OF FILMS Magn. Sput. Nb 0° 45° 90° Line of sight issues 400 µm => porosities [G. Rozas] Sensitivity to Cu roughness (the smoother, the better) Inverted pyramid crystalline growth Ra = 0.1 µm Internal stress Advantage: higher Tc (up to a certain Ra = 0.15 µm impurity concentration) Disadvantage: adhesion issues Ra = 0.025 µm (peeling) Ra = 0.01 µm High impurities content Ra = 0.015 µm Nb = getter material (nearly as good as Ti => high interstitial content) Carrier gas incorporation (Ar) [M. Ribeaudeau, PhD] 2021/06/24 C.Z. Antoine -SRF 2021 tutorial -| Beyond 23 Nb
SUBSTRATE ISSUES Cu and Nb not miscible (especially in presence of O) Advantage: low interdiffusion Disadvantage: adhesion issues (peeling) Issues on Cu welding arreas Best results are not always [data from CERN + where AM Valente-Feliciano] expected: Standard Oxide-free films films RRRmax =28 RRRmax= 40 Bulk like films did not perform better ! (but recent changes !!!) New approaches: Bulk machining Electroplating Laser surfacing … 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | 24
SEAMLESS ISOLDE CAVITY Local magnetometry on a sample deposited in the same condition First evidence of a bulk-like behavior for a thin film !!!! 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 25
DENSIFICATION /MITIGATION OF THE Q-SLOPE 2021/06/24 C.Z. Antoine -SRF 2021 tutorial -| Beyond 26 Nb
OTHER SUPERCONDUCTORS 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 28
MOST COMMON SUPERCONDUCTORS: - A2 (e.g. NbTi, Transition metal alloys, BCC structures) - B1 (e.g. NbN, NbTiN, Transition metal carbide or nitride, NaCl structures) - A15 (e.g. Nb3Sn, Compounds, NaCl structures) - 2-D SC (Compounds, anisotropic) - MgB2 - Cuprates, Pnictides - (others TaS2, organic…) - SPECIAL SRF: METAMATERIALS (Multilayers) 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 29
A2 SC ALLOYS: e.g. NbTi Ti BCC pure metal and solid solution alloy Ti precipitates ( ~0,4 µm) NbTi widely used in coils NC Metal => RF dissipation !!!! Available alloys range around 45-55 % Ti Ti is not fully miscible inside Nb (Ti precipitates ꓱ at low T when [Ti] > 5 W%) => no RF !!! Ti = Nb Ti precipitates in a niobium matrix (with a few Nb replaced by substitutional Ti) => ~ same TC, same HC as Nb, but not same l => high k http://www.dierk-raabe.com/titanium- alloys/biomedical-titanium-alloys/ 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | 30
A1 SC COMPOUNDS: e.g. NbN https://link.springer.com/content/pdf/10.1007%2F978-1-4757-0037-4.pdf BCC pure metal + smaller atoms (N, C) in interstitial location => NaCl structure NbN cubic phase : TC ~17-18 K NbTiN stabilization of cubic (SC) phase NbN not too sensitive to local variation of composition ! Solid solution => relatively easy fabrication (thermal diffusion, reactive sputtering…) Good model SC Widely used for JJ and SC electronics 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | 31
A15 COMPOUNDS : HIGH TC compound TC (K) compound TC (K) compound TC (K) compound TC (K) Ti3Ir 4.6 V3Os 5.15 Nb3Os 0.94 Cr3Ru 3.43 Ti3Pt 0.49 V3Rh 0.38 Nb3Rh 2.5 Cr3Os 4.03 Ti3Sb 5.8 V3Ir 1.39 Nb3Ir 1.76 Cr3Rh 0.07 V3Ni 0.57 Nb3Pt 10 Cr3Ir 0.17 Zr3Au 0.92 V3Pd 0.08 Nb3Au 11 Zr3Pb 0.76 V3Pb 3.7 Nb3Al 20.3 Mo3Re 15 V3Au 3.2 Nb3Ga 18.9 Mo3Os 11.68 V3Al 9.6 Nb3In 8 Mo3Ir 8.1 V3Ga 15.4 Nb3Ge 23 Mo3Pt 4.56 [after Due-Hugues] V3In 13.9 Nb3Sn 18.3 Mo3Al 0.58 V3Si 17.1 Nb3Bi 2.25 Mo3Ga 0.76 V3Ge 7 Mo3Si 1.3 V3Sn 4.3 Ta3Ge 8 Mo3Ge 1.4 V3Sb 0.8 Ta3Sn 6.4 Ta3Sb 0.72 Extreme brittleness !!! cannot be plastically formed Phases with proper stoichiometry thin/thick film route only ! (A3B) not stable in normal condition (RT to Cryogenic temp) nm µm => Quenching necessary 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 32
A15 COMPOUNDS : NARROW DOMAIN OF SC A3B Nb3Ga Nb3Al Nb3Ge B atoms occupy corners and center of BCC structure A atoms form orthogonal chains bisecting the faces of the BCC unit cell. Linear Chain Integrity is crucial for Tc (long-range order required) Narrow range of concentration for the SC phase: Highest TC area is even narrower Difficult to get uniform SC phase everywhere* Special issues at grain boundaries: “intrinsic” local deviation of stoichiometry* In Nb3Sn wires : GB exhibit degraded SC => weak links, pinning centers Compare with 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 33
Nb3Sn Special interest for SRF since the 1980’s 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | 34
Nb3Sn ON Nb (thermal way) Pioneer work: Wuppertal, Cornell [Lieppe, SRF 2013] @ 4,2 K: Q0 x 20 compare to Nb, @ 2K ~ the same Limited in Eacc , best results today ~20 MV/m Important developments: FNAL, JLAB, CERN, PKU…. 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | 35
HSHNb3Sn (~ 400 mT @ 0 K) EFFECTS OF LOCAL DEFECTS Hays. "Measuring the RF critical field of Pb, Nb, NbSn". in SRF 97. 1997. Nb HC1 (170 to 200 mT) Nb3Sn Cornell, 1997 pulsed Vs Nb Recent results from Cornell & FNAL CW Posen, 2019 (23 MV/m) 2015 (17 MV/m)) HC1Nb3Sn (~27mT) (50 mT pour le bulk) ~ 2K for Nb 3,5 K for Nb3Sn => We have to reduce defect density (yes but which ones?) 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 36
https://indico.jlab.org/event/405/contributions/7865/attachments/6613/8979/Zeming%20Sun_Thin%20film%20workshop%20slide.pdf 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | 37
OTHER APPROACHES Sputtered Nb3S films on copper Activities at Cern, STFC, Jlab, Cornell,,, RT deposited films : right composition but no A15 structure Heating of substrate (CERN) And/or post annealing Other issues Cracks on the layer (due to differential dilatation coef) Diffusion of copper in the layer Carrier gas incorporation (Ar, Kr) Sn evaporation at higher temperature (> 1000°C) 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 41
OTHER APPROACHES…2 Electrochemical deposition + diffusion through copper Proposed at FNAL Inspired from wire fabrication Not expensive !!!! Multilayer is heated => solid state diffusion Cu lowers the formation Tp° of A15 phase and suppresses the unwanted NbSn2 and Nb6Sn5 phases. 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 42
Nb3Sn: TENTATIVE DESCRIPTION OF THE LANDSCAPE Method Sn vapor/ Sputtering Co-sputtering Sequential Bronze routes liquid ! target sputtering + (electrochemical deposition + diffusion annealing annealing) substrate Nb only Nb or Cu Nb or Cu Nb or Cu Nb or Cu Activities at Cornell Cornell STFC* IMP IMP FNAL Cern Peking U. Jlab Cornell JLAB Jlab Darmstadt ODU FNAL INFN ASC(FSU) Wisconsin U.* FNAL KEK * In Multilayer structures NB : work on V3Si also in progress 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 43
2-D SC (Compounds, anisotropic) - MgB2 - Cuprates, Pnictides - Multilayers 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 44
MAGNESIUM DIBORIDE (MgB2) Graphite-type boron layers BCS type superconductor separated by hexagonal close- packed layers of magnesium TC ~40 K, two-gap nature Advantages: Very high TC (higher temp operation) Semimetal, cheap (fertilizer !) , l of high quality* MgB2 similar to Nb (~50 nm) (transparency of GB to current flow) Low n (lower RS) p=2.3meV, 2D, in-plane s-orbital Disadvantages: s=7.1meV 3D, out-of-plane p-orbitals Orientation issues (in polycrystalline materials !) RF dominated by lower gap ! Still better than Nb : ΔNb= 1.5 meV < ΔπMgB2= 2.3 meV < ΔNb3Sn= 3.1 meV
MAGNESIUM DIBORIDE (MgB2) Phase diagram: at low Mg pressure only extremely low deposition temperatures can be used Optimal T for epitaxial growth ~ Tmelt/2 For MgB2 Tmelt/2 = 540°C => PMg ~11 Torr Too high for UHV deposition techniques (PLD, MBE...) At PMg = 10-4-10-6 Torr, and Tsub ~ 400°C Compatible with MBE, and other deposition techniques MgB2 is stable, but no MgB2 formation: Mg atoms re-evaporate before reacting with B At PMg = 10-4-10-6 Torr, and lower T MgB2 is stable, If Tsub >250°C, free Mg is lost because the re- evaporation rate is higher than the impinging rate If Tsub < 250°C .Growth rate is very slow, Z.-K. Liu et al., APL 78(2001) 3678. Mg (kinetically limited by available Mg) MgB2 evaporation pressure of Mg from MgB2 < decomposition curve of MgB2 M. Naito and K. Ueda, < Mg vapor pressure Kinetically limited Mg SUST 17 (2004) R1 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 46
MgB2 – HPCVD ON METAL SUBSTRATES HYBRID PHYSICAL CHEMICAL VAPOR DEPOSITION MgB2 [X. Xi- TempleUniversity] Polycrystalline MgB2 films deposited: On stainless steel, Nb, TiN, and other substrates. Flat samples and tubes (conformational) interface SiC Fitted for SRF apps: RRR>80 low resistivity (
MGB2 – OTHER ROUTE S RF measurement @ MIT/Lincoln Lab In-situ reactive evaporation @ 550˚C High quality flat samples Difficult to apply to complex geometries [T. Tajima, LANL] Superconducting Technologies Inc. Plasma electrolytic oxidation (PEO) MgB2 particles in suspension in an electrolyte MgB2 Islands deposited on the surface Issues : homogeneity, purity To be further explored [R. Valizadeh, STFC] 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | 48
HPCVD MGB2 – RF MEASUREMENTS 11.4 GHz TE013 cavity @ SLAC 7.5 GHz sapphire-loaded TE011 cavity at JLab The MgB2 coatings were also characterized at 11.4 GHz at SLAC using a cryogenic RF system. B.P.Xiao et al., SUST 25 (2012) 095006. The samples showed a Q factor comparable to a Nb reference sample and higher than the Cu reference sample. The films showed a Tc of 37 K. Stripline resonator Scaled to 1,5 Ghz Lower surface resistance comparable to Nb film. [P. Welander, SLAC] Nb on sapphire MgB2 on LAO MgB2 on sapphire Oates et al., SUST 23, 034011 (2010)
THE IMPORTANCE OF MEAN FREE PATH, MgB2 EXAMPLE Most developments of HTc are done in view of magnet applications : small high l by playing with ℓ Small x makes the superconductors very sensitive to (usual) crystalline defects (wanted for magnets, not for SRF) Typical values found in literature for MgB2 : l ~100-150 nm ~ 5-10 nm ℓ mean free path if ℓ ↗ then: 1 1 1 ↗ = + 0 ℓ l↘ k↘↘ l~50 nm ~ some 10 nm ? 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | 52
YBCO FAMILY… NOT FOR SRF ! MonoXtal : Jc maximum for (a,b) planes and minimum when // c axis c (~0,03 nm) try to introduce preferential orientation (epitaxy): difficult to get on a cavity (but is applied to fabricate tapes for magnets) D-symmetry of the gap superconducting gap is also anisotropic = zero at 4 line nodes located at the diagonals of the Brillouin zone = 0 => power law for RS: RS α to T2-T3 For the recall: gaps of conventional SC have s symmetry: http://mason.gmu.edu/~grobert1/2014syl641.htm isotropic and RS α e-/T (BCS resistance) Gap () Fermi Fermi surface surface = 0 3.6 nm d-wave s-wave https://areeweb.polito.it/ricerca/ Twin boundary in YBCO superconductivity/melt.htm 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | 53
PNICTIDE FAMILY… MAYBE YES ? Oxypnictide base: ReOMPn M = Fe, Co, Ni Pn = As or P Re = La, Nd, Sm, Pr A lot of common with YBCO High TC (10-55 K up today) Layered structure Brittle material d-wave symmetry observed for some member of the family But most compounds exhibit s-wave gaps…? Opening for SRF application ? 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | 54
PNICTIDE FAMILY… MAYBE YES ? A lot of common with YBCO NaFe1−xCoxAs (x = 0.0175) High TC (10-55 K up today) = ferromagnetic Layered structure NaFe1−xCoxAs (x=0.045) Brittle material = SC but Most compounds exhibit s-wave gaps Very sensitive to impurities content (either magnetic or not) ~ 5-6 meV >> Nb3Sn (3,1meV) Ba0.6K0.4Fe2As2 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | 55
MULTILAYERS 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | 57
AFTER NIOBIUM : NANOCOMPOSITES MULTILAYERS Structures proposed by A. Gurevich in 2006, SRF tailored Dielectric layer Small vortex (short -> low dissipation) Quickly coalesce (w. RF) Blocks avalanche penetration => Multilayer concept for RF application Nanometric I/S/I/ layers deposited on Nb SC nanometric layers ( 100 nm) => HC1 => Vortex enter at higher field HNb Nb surface screening => allows high magnetic Happlied // field inside the cavity => higher Eacc Cavity's interior → SC w. high TC than Nb (e.g. NbN): R SNbN 1 R SNb dsc 10 => Q0multi >> Q0Nb S-I-S 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | 58
FIRST APPROACH: TRILAYERS Meissner state stable if: d Screening current @ both SC surface is < depairing current J(0) < Jd = Hs/λ and J(d) < Jd0 = Hs0/λ0 If d is small, HSHS is high, but most of the field reach S0 If d is thicker, HSHS is lower, but screening is more effective S I S0 ꓱ an optimum thickness and a maximum screening field!!! Maximum screening field Hm at the optimum S thickness dS = dm d/λ 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | 59
SIS OPTIMIZATION: IMPORTANCE OF MODELS A. Gurevich, T. Kubo First approach: trilayers Exp decay Boundary conditions implemented (including effect of an insulating layer. finite thickness)) w. boundary HSH determined initially in London approx., further cond. improved w. quasiclassical theory (valid @T
TRILAYER OPTIMIZATION (…) A. Gurevich, T. Kubo Go for realistic condition Layers present defects, non-negligible surface roughness, non-uniform thickness. → HSHS suppressed due to of the local screening current enhancement. Introducing material suppression factor η =f(defect size and aspect ratio, ξS ) η ~ 0.85 for typical electropolished Nb surface) HSHSIS and optimal S layer thickness dmS can be determined w. surface topographical data Ideal Nb substrate Nb with defects*, with BC1=170 mT with BC1=50 mT * e.g. morphologic defects that allow earlier vortex penetration See exp proof later on 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | 62
WHAT IS THE LIMIT (HP/HC1/HSH) ? Real world cavities behavior is dominated by a few number of defects It is very important to measure the penetration field of samples in realistic conditions H H H// HSH HSH H// HC1 HC1 H H T2C T2 T2C T2 Local magnetometry ~ Same geometry as cavities No shape/edge effect (vs DC/ Squid magnetometry) No demagnetization effect Measures actual penetration field wherever it is HP/HC1/HSH 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 63
EXPERIMENTAL DETAILS H maximum field without Vx (max Eacc) HC2 maximum pinning Vx ~ liquid HMelting Vx lattice HC1 Ia/Ha Meissner state TC T Low frequency ≡ DC : flux line moving in AC field 0 < Ha< HC1 => R=O, Meissner state Vx pinned in HC1< Ha< HM =>Vx are trapped, R=O, Campbell regime the SC HM< Ha< HC2 =>Vx are moving liquid like, R≠O, Flux flow regime Third harmonic signal arise from flux line tension (affects the e- inside the Cu coil), It does not depend on dissipation inside Nb, BUT depends on # of Vx trapped there (and length). 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 64
Nb – INSULATOR – NbN MODEL NbN coating by Magnetron Sputtering NbN single layers series NbN SL / “thick” Nb layer Magnetron sputtered MgO as dielectric layer Far from perfect… MgO (nm) NbN (nm) Nb (nm) TC (K) Calc(actual) Calc(actual) 250† 14 0 8.9 250† 14 25 15.5 500 10 (10.3) 50 (65) 15* 500 10 (8.4) 75 (72) 14.1* 500 10 (9.8) 100 (94) 14* 500 10 125 14.3* 500 10 (6.7) 150 (132) 15.9* 500 10 (10.4) 200 (164) 15* † Not same batch, deposited on the same conditions, but substrate = sapphire *As determined with magnetometry, see below. 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 65
SPUTTERED (DEFECTIVE) MATERIALS… Typical defects… NbN 100nm Low HC1 NbN Thickness ~ 105nm Thickness ≠ uniform … 250 nm Nb on saphire DC Squid magnetometry @ 4.5 K Nb Thickness ~ 425nm Rippled interfaces NbN 50nm ~ 18 mT As expected for MS samples 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 66
COMPARAISON WITH THEORY Theoritical predictions from T. Kubo (KEK) x x x x Ideal Nb substrate x Nb with defects*, x with BC1=170 mT with BC1=50 mT x x x * e.g. morphologic x defects that allow earlier x x vortex penetration See SST paper cited earlier T (K) 02 4 6 8 10 12 14 16 80 70 Nb NbN 25nm The enhancement of the field 60 NbN 50nm NbN 75nm penetration increases with thickness of Hc1 (mT) 50 NbN 100nm NbN 125nm NbN 40 NbN 150nm 30 NbN 200nm It reaches a saturation at thicknesses > 20 100 nm 10 0 0 40 80 120 160 200 240 280 T2 (K2) 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 68
COMPARAISON WITH THEORY Theoritical predictions from T. Kubo (KEK) x x x x Ideal Nb substrate x Nb with defects*, x with BC1=170 mT with BC1=50 mT x x x * e.g. morphologic x defects that allow earlier x x vortex penetration See SST paper cited earlier According to 2nd transition of V3 150 140 130 @ 4.5 K Nb ~ + 110 mT? 120 110 NbN 25nm The enhancement of the field ~25-30 MV/m NbN 50nm 100 NbN 75nm penetration increases with thickness of ILC shape 90 Hc1 (mT) NbN 100nm 80 NbN 125nm NbN 70 NbN 150nm 60 NbN 200nm It reaches a saturation at thicknesses > 50 40 100 nm 30 20 10 0 0 50 100 150 200 250 300 2 2 T (K ) 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 69
CLOSEUP OF 3rd HARMONIC SIGNAL V3(V) For a given Happl, we observe 3 0,00012 Phase(deg) ≠ transition temperatures 0,00010 115 Phase signal 110 0,00008 1 Phase(deg) V3(V) 105 V3(V) Phase(deg) 0,00006 0,00012 150 100 0,00004 0,00010 100 95 0,00002 0,00008 90 0,00000 Phase(deg) 50 V3(V) 0,00006 6 8 0 T~T Temperature(K) 0,00004 T >>T 3 V3(V) Phase(deg) 150 0,00002 0,00004 -50 0,00000 100 -100 4 6 8 10 12 14 16 18 Phase(deg) 50 Temperature(K) V3(V) 0 2 T~T: within noise level Voltage signal -50 T >>T: dramatic transition -100 6 8 Temperature(K) 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 70
ROLE OF THE DIELECTRIC LAYER ! Why do we have two transitions ? H app Defect Thin SC layer NbN Insulator MgO Thick SC layer Nb Field lines First transition H // surface => surface barrier† A defect locally weakens the surface barrier 1st transition, vortex blocked by the insulator ~100 nm => low dissipation. 2nd transition, propagation of vortex avalanches (~100 µm) => high dissipation. Dielectric layer = efficient protection !!! Second transition 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | PAGE 71
SIS : IRREVERSIBILITY => NO VORTEX PINNING
1st TEST RF @ 3,88 GHZ (4 25nm NbnN LAYERS) Accelerator cavities' Rs = f(T) operating range 8 7 Rs(ML4) 6 Rs(Nb) (scaling in w2) 5 Rs (µOhm) 4 3.88 GHz => 3 RBCS~ 9 x R (1.3 GHz) 2 1 0 0 1 2 3 4 5 T(K) Comparison is done with a high performance 1.3 GHz Nb cavity (scaling in 2) Indium gasket presents some defects measured with thermometric map => extra RF losses Residual resistance comes from NbN + bulk Nb substrate + indium gasket. Further investigations needed. 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | 77
[A.M.Valente-Feliciano] 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | 78
ML WITHOUT DIELECTRIC INTERLAYER µ-SR [Junginger, SRF 2017] The SS boundary provides an additional barrier to prevent penetration of vortices. It would not be as robust as the I layer of the SIS structure, but it also contributes to pushing up the onset of vortex penetration. [Kubo, SST] 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | 79
CONCLUSIONS AND PERSPECTIVES Superconducting cavities are dominated by their surface quality (Niobium AND other SC !) Niobium is close to its ultimate limits, but can be surface tailored (doping) HSH difficult to reach in real “accelerator cavities” (low T, large scale cavity fabrication, surface defects,…) ML structures seem to be a promising way to go toward realistic complex materials (+ Nb cavity upgrade) Renewed activity on bulk-like Nb films (cost issues) and high HSH SC e.g. Nb3Sn or NbN (higher performances) Look for higher Q0, not only Eacc ! WE ARE ON THE EVE OF A TECHNOLOGICAL REVOLUTION FOR SRF CAVITIES ! 2021/06/24 C.Z. Antoine -SRF 2021 tutorial - Beyond Nb | 80
You can also read