Advances in Development of Nb Coating Technology A-M Valente-Feliciano Acknowledgements JLab : G. Eremeev, O. Trofimova, J. Spradlin, C. Reece, L. Phillips College William & Mary : Prof. A. Lukaszew & Group NSU: J. Skusa, Q. Yang NCSU: F. Stevie & Group ODU: A. Gurevich CERN: S. Aull, S. Calatroni, G. Terenziani, A. Sublet LBNL: A. Anders ANL : T. Proslier TRIUMF: T. Junginger AASC: M. Krishnan A-M Valente-Feliciano - FCC Week /24/2015 State of the art of Nb/Cu Technology Thin films Growth Energetic Condensation Coaxial Energetic Deposition HiPIMS ECR Concluding Remarks Outline A-M Valente-Feliciano - FCC Week /24/2015 RRR Nb = R 300K /R 10K Gauge for mean free path & material quality, affected by scattering centers such as structural defects, impurities Thin Films for SRF - State of the Art CERN LEP x 353MHz Nb/Cu 4-cell cavities LHC16 x 400 MHz Nb/Cu 1-cell cavities INFN Legnaro 52 x 160 MHZ Nb/Cu QWR Standard films Oxide-free films Courtesy: P. Jacob - EMPA Columnar grains, size ~ 100 nm In plane diffraction pattern: powder diagram (110) fiber texture substrate plane Equi-axed grains, size ~ 1-5 m In plane diffraction pattern: zone axis [110] Heteroepitaxy Nb (110) //Cu(010), Nb (110) //Cu(111),Nb (100) //Cu(110) RRR max =28 RRR max = 40 A-M Valente-Feliciano - FCC Week /24/2015 High Q at low field BUT strong Q-slope Bulk Nb 1.5 GHz Nb/Cu LEP II 350MHz Nb/Cu (4.2K) LEP II 350MHz Nb/Cu (4.2K) CERN 2000 C. Benvenuti et al., Physica C, Vol. 351(4), April 2001, pp. 421428Vol. 351(4 1.5 GHz Nb/Cu cavities, sputtered w/ 1.7 K (Q 0 =295/R s ) 40 MV/m + Bulk Nb Thickness of interest for SRF applications = RF penetration depth, i.e. the very top 40 nm of the Nb surface. Bulk-like performance Nb film Minimize Rres, maximize Q critical for CW RF Potential major system simplifications Highest level of quality assurance and reliable performance. Use of substrates with higher thermal conductivity (Cu, Al) Next generation Nb films for FCC A-M Valente-Feliciano - FCC Week /24/2015 The thickness of interest for SRF applications corresponds to the RF penetration depth, i.e. the very top 40 nm of the Nb film. Identify and correct the cause(s) of anomalous Q-slope (not limited to Nb films alone, higher Tc films for multi layers also affected, especially with such issues as the entry of Josephson vortices driven by the RF field.) focused on the entire problem and all possible causes in order to understand, identify, and eliminate the causes. Higher density of grain boundaries Different spectrum of grain boundary energies than for bulk Nb surfaces Thermal diffusivity of the RF surface (from the thermal properties of the film itself in addition to the thermal impedance of the Nb/ substrate interface). Surface chemistry Presence of defects (dislocations, porosities, inclusions) Surface morphology. Magnetic field losses- several sources: field enhancing surface features, external stray fields captured in form of vortices or arising from thermoelectric currents as Nb transitions from normal to superconducting state. Crystallographic structure (orientation, grain size) What are the differences between bulk and thin-film Nb in reference to the RF surface? ALL film properties are a direct consequence of the film structure, defect/impurity content thus the technique, environment, substrate are key factors Next generation Nb films for FCC A-M Valente-Feliciano - FCC Week /24/2015 Careful characterization of the attained composition and microstructure (RHEED, STM, XRD, EBSD, AFM, optical profilometry, XPS, SIMS,TEM, FIB). Close association with resulting RF surface impedance & superconducting properties (, , T c, H c, RRR) UNDERSTANDING OF The chemistry of the involved species Reactivity Stoichiometric sensitivity Reaction process temperatures Crystal structure dependence on substrate structure Influence of deposition energy on resulting structure Sensitivity to the presence of contaminating species, defects Stabilization of desired film against subsequent degradation Full control of the deposition process & tailored SRF performance Full control of the deposition process & tailored SRF performance Thin Films Growth A-M Valente-Feliciano - FCC Week /24/2015 Control over the deposition process is exercised by only 3 first-order vapor parameters & 1 first-order substrate parameter. Vapor parameters Absolute arrival rates of film atoms Partial pressures of background gases in the chamber Energies of the deposition fluxes. Substrate parameter Substrate temperature T. Without energetic atoms, only the substrate temperature influences the processes of physi- and chemisorption, thermal desorption, nucleation, nuclei dissociation, surface diffusion, and formation of specific nucleation sites. nucleation & island growth island coalescence continuous film (subsequent growth). [C. Thompson] Energetic Condensation A-M Valente-Feliciano - FCC Week /24/2015 Additional energy provided by fast particles arriving at a surface number of surface & sub- surface processes changes in the film growth process: Generalized Structure Zone Diagram residual gases desorbed from the substrate surface chemical bonds may be broken and defects created thus affecting nucleation processes & film adhesion enhanced mobility of surface atoms stopping of arriving ions under the surface morphology microstructure stress As a result of these fundamental changes, energetic condensation allows the possibility of controlling the following film properties: Density of the film Film composition Crystal orientation may be controlled to give the possibility of low-temperature epitaxy Condensing (film-forming) species : hyper-thermal & low energies (>10 eV). Changes in A. Anders, Thin Solid Films 518 (2010) 4087 derived from Thorntons diagram for sputtering (1974) Energetic Condensation Techniques A variety of techniques with distinct technologies Vacuum Arc Plasma & Coaxial Energetic Deposition (CED) High Impulse Power Magnetron sputtering (HiPIMS) Electron cyclotron Resonance (ECR) A-M Valente-Feliciano - FCC Week /24/2015 Coaxial Energetic D eposition (CED TM ) The difference between 4K and 2K is smaller than expected: BCS resistance should be about 40x less if all the surfaces are Nb. suggests presence of areas not well coated well and lossy presence of macro-particles KEK06 hydro-formed copper cavity, CBP at FNAL (Cooper), coated at AASC, tested at LANL (Tajima, Haynes) CED coater uses welding torch technology Arc source is scalable to high throughputs for large scale cavity coatings Present version deposits ~1 monolayer/pulse ~0.2 ms Good structure, RRR but presence of macro-particles Coaxial Energetic Deposition (CED) - AASC very high purity Excellent adhesion better (normal) conductivity, Large crystal grains, low defect density Suppression of fiber structure Superior density Decreased roughness Homogeneous coating even on complex-shaped surfaces Phase composition tailoring Interface engineering E NERGETIC CONDENSATION VIA H I PIMS A-M Valente-Feliciano - FCC Week /24/2015 The target material is partly ionized such that the ions will reach the substrate with a higher energy and, in case of an applied bias voltage to the substrate, always normal to the surface. The intense pulse plasma density provides a large concentration of ions producing high-quality homogeneous films. The high ionization fraction allows fine control of the sputtered species during deposition High Power Impulse Magnetron Sputtering (HiPIMS) produces films with enhanced properties V. Kouznetsov, et al., Surf. Coat. Technol. 122 (1999) 290 HIPIMS= pulsed sputtering where the peak power exceeds the average power by typically two orders of magnitude. A-M Valente-Feliciano - FCC Week /24/2015 Energetic condensation with HiPIMS - CERN CERN (G. Terenziani, S. Calatroni, T. Junginger) Sheffield-Hallam University (A.P. Ehisarian) Seemingly larger surface features than in DCMS RRR ~ GHz Cavity A-M Valente-Feliciano - FCC Week /24/2015 HiPIMS cylindrical coating system for single cell under commissioning Energetic condensation with HiPIMS ion trajectory possible collision cavity sheath nearlyperpendicularincidence! plasma V bias magnetron 1 magnetron 2 high power mode (above runaway threshold) Dominated by Nb emission LBNL(A. Anders) HiPIMS dual magnetron system Most effective for Biasing & influencing Ion Energies & Trajectories R. VALIZADEH, 15:00 Thin films SRF studies in ASTeC HiPIMS & CVD No working gas Ions produced in vacuum Singly charged ions 64eV Controllable deposition energy with Bias voltage Excellent bonding No macro particles Good conformality Generation of plasma - 3 essential components: Neutral Nb vapor RF power 2.45GHz) Static B E RF with ECR condition Conformality of the ECR process : the film thickness along a 3GHz half-cell profile varies from 4 m (equator) to 6 m (iris) Note: the substrate is very rough, was only grossly mechanically polished Energetic condensation with ECR A-M Valente-Feliciano - FCC Week /24/2015 ECR Nb films on Al 2 O 3 substrates - Crystalline vs. distorted A-M Valente-Feliciano - FCC Week /24/2015 Amorphous Al 2 O 3 ceramic substrate RRR=89 Crystalline Al 2 O 3 (11-20) substrate RRR= 247 A. R. Wildes et al., Thin Solid Films, (2001) Fiber Growth Heteroepitaxy with continuously crystalline interface Al 2 O 3 (11-20) Nb film substrate Interface ECR Nb films on Al 2 O 3 substrates Influence of film growth in phases (dual energy) A-M Valente-Feliciano - FCC Week /24/2015 Nb (100)/(1-120) Al 2 O 3 always higher RRR and bulk-like T c than Nb (110) and (111) 500 C, 360 C, for all Al 2 O 3 substrates dual energy and thick films (interrupted growth) Nb on Cu substrates A-M Valente-Feliciano - FCC Week /24/2015 Nb/ single crystals Cu Nb/ large grain Cu Nb/ fine grain Cu Nb(100) typically higher RRR Nb/large grain Cu can achieve higher RRR due to underlying relaxed (heat treated) substrate For Nb/fin grain Cu, less variation in RRR as preferred orientation (110) RRR=169 RRR=82 RRR=88 RRR=242 RRR=76 Nb(110)/Cu(100) Nb(100)/Cu(110) Nb(110)/Cu(111) 150 x 150 m, 1 m resolution, CI Avg x 150 m, 1 m resolution, CI Avg x 75 m, 1 m resolution, CI Avg eV (Bias -120V), 2 m 64 eV (Bias 0V), 4 m Typical Cu substrate Tune the Nb Film Structure with Ion Energy A-M Valente-Feliciano - FCC Week /24/2015 Ab-normal growth vith bias vs. columnar growth without bias Nb on crystalline Cu substrate (360 C) Nb on Cu oxide (200 C) Amorphous interface Continuous crystalline interface =1.71 meV =1.56 meV Gap measurements performed by PCT (point contact tunneling spectroscopy- T. Proslier) Superconducting gap ( meV) similar to bulk Nb ( Nb bulk =1.55meV measured on the same setup) for hetero-epitaxial ECR Nb films on polycrystalline Cu. ECR Nb/Cu Structure, interface and superconducting gap A-M Valente-Feliciano - FCC Week /24/2015 Single crystal versus large grain film (360 C) If dense, grain boundaries not necessarily detrimental to RF performance Dual energy : Nucleation and early growth at 124 eV Subsequent growth at 64 eV When films deposited in 2 phases high energy, subsequent 64 eV ), some improvement of Rs ECR Nb /Cu Surface resistance A-M Valente-Feliciano - FCC Week /24/2015 Growth at continuous energy (360 C) Decrease of R s with increasing incident ion energy Tc= 9.36 0.12 K RRR = 179 (Nb/a-Al 2 O 3, witness sample) EBSD IPF map and XRD pole figure show very good crystallinity and grain sizes in the range of the typical Cu substrate Nb film in the clean limit * [nm]RRR 144 2053 7 * with L = 32 nm and 0 = 39 nm R res [n] (0K) [nm] 400 MHz46.6 MHz79 238 MHz156 1138 1 Bake & coating temperature: 360 C Coating with dual ion energy: 184 eV for nucleation/early growth 64 eV for subsequent growth Hetero-epitaxial film Nb on OFHC Cu A-M Valente-Feliciano, JLab S. Aull, CERN ECR Nb/Cu Surface resistance A-M Valente-Feliciano - FCC Week /24/2015 Complete RF measurement Sample to be stripped and coated with HiPIMS at CERN (Film quality highly dependent on substrate quality) More samples in preparation with different deposition conditions Complete RF measurement Sample to be stripped and coated with HiPIMS at CERN (Film quality highly dependent on substrate quality) More samples in preparation with different deposition conditions Tailored Nb films via energetic condensation Tune thin film structure and quality with ion energy and substrate temperature on a variety of substrates (amorphous, polycrystalline and single crystal) Achieve film structures and properties only achievable at higher temperature with classic coating methods Tune RRR values from single digits to bulk Nb values No intrinsic limitations Lower impurity (H) content than bulk Nb Good adhesion to the substrate (delamination threshold determined as function of ion energy and temperature) Grain boundaries not necessarily detrimental (if dense) to R s Tailoring interface with high energy and subsequent growth at energy minimizing defect creation can contribute to lower R s A-M Valente-Feliciano - FCC Week /24/2015 Engineering for optimum RF performance 3 sequential phases for film growth Film nucleation on the substrate (Nb, Al 2 O 3, Cu; single crystal, polycrystalline, amorphous) Growth of an appropriate template for subsequent deposition Deposition of the final surface optimized for minimum defect density. A-M Valente-Feliciano - FCC Week /24/2015 Nb Cu Conclusions Nb films deposited by energetic condensation (HiPIMS, ECR) Significant improvement in film quality (crystallinity, impurity content, RRR, superconducting gap) Under way RF characterization (TE 011, 013, QPR)of Nb/Cu samples deposited with various energetic condensation techniques currently the bottleneck Further tailor the interface to optimize structure for maximum SRF performance (ion stitching, interlayer) Study effect of top SRF surface doping/alloying Coating systems available or under development for elliptical cavities Results are widely the fruit of collaborations between institutions Enhancing already existing collaborations and developing new ones will speed up progress A-M Valente-Feliciano - FCC Week /24/2015