Q-Slope Studies at Fermilab: New Insight From Cavity and Cutouts - PowerPoint PPT Presentation

Q-Slope Studies at Fermilab: New Insight From Cavity and Cutouts Investigations A. Romanenko Fermilab

Outline • New experimental findings on Q slopes – Decomposition of the components of surface resistance (R BCS and R res ) • Shows which Q slope is due to what component • New superconducting measurements – Low energy muon spin rotation • Baked/unbaked cutouts • N doped • New proximity effect model of the high field Q slope – Evidence from cryogenic TEM investigations in cutouts • New model of the 120C baking – Vacancy-based 120C baking mechanism and supporting evidence from cutouts – Suppression of the second phase of hydrides in direct observations • Conclusions

Decomposition of Rs into components • Using different temperature dependence to deconvolute the components of average surface resistance at ALL fields R s (T) = R BCS (T) + R res Non-T-dependent, Due to thermally excited saturation value at quasiparticles T -> 0 3 September 30, 2013 Alexander Romanenko

Rs(B) decomposition Fit a set of R s (T) 3 10 6.0x10 5 20 curves to extract 7 9 18 11 R res at each E acc 13 16 15 10 4.0x10 17 14 19 R s (nOhm) 21 12 Q 0 23 25 10 27 10 2.0x10 29 8 6 4 0.0 1.5 1.6 1.7 1.8 1.9 2.0 0 5 10 15 20 25 30 35 Temperature (K) E acc (MV/m) Measure Q(E acc ,T) at Can be fitted using both approximate formula R BCS (T)=A/T exp(- ⊗ /kT), and by many different T<2.17K more precise BCS calculation based on Halbritter’s program – virtually no difference and E acc in the results A. Romanenko and A. Grassellino, Appl. Phys. Lett. 102, 252603 (2013) 4 September 30, 2013 Alexander Romanenko

Residual resistance A. Romanenko and A. Grassellino, Appl. Phys. Lett. 102, 252603 (2013) Contributes to the BCP EP medium field Q BCP+120C slope High field Q EP+120C slope is clearly a 10 residual R 0 resistance effect For some treatments decreases at lower fields 1 0 5 10 15 20 25 30 Eacc (MV/m) 5 September 30, 2013 Alexander Romanenko

BCS resistance A. Romanenko and A. Grassellino, Appl. Phys. Lett. 102, 252603 (2013) Unbaked 14 12 Typically Baked at 120C cited effect 10 Rbcs2K (nOhm) of 120C A strong change baking on the BCS in the field surface resistance 8 dependence due to 120C bake More on the medium BCP 6 field Q slope – hot topic EP BCP+120C session on Thursday EP+120C 0 5 10 15 20 25 30 Eacc (MV/m) 6 September 30, 2013 Alexander Romanenko

SC gap change with field mfp << ξ Field dependence of R BCS may be explained by the expected changes of pairing potential Δ = Δ (H) in clean (unbaked) and dirty mfp >> ξ (120C baked) limits 7 September 30, 2013 Alexander Romanenko

Role of thermal “feedback” Instead of modeling the full temperature EP+120C, 2K - heating effect 7 EP+120C, 1.66K - heating effect transfer with only BCP+120C, 2K 6 EP+120C, 2K R s =G/Q 0 as an input EP+800C+EP+120C, 2K use temperature 5 mapping to measure the outside wall 4 ∆ R BCS (n Ω ) temperature 3 2 Negligible effect on 1 R BCS at T <= 2K 0 More – hot topic 0 20 40 60 80 100 120 session on Thursday B (mT) 8 September 30, 2013 Alexander Romanenko

Correlation between medium and high field Q slopes in unbaked cavities T-map data shows that local surface resistance in HFQS regime is highly correlated to Rs at lower fields (MFQS) More info – please see [A. Romanenko et al, TUP101] 9 September 30, 2013 Alexander Romanenko

New cavity data allows to “filter” models • High field Q slope is due to residual – Not SC gap closing, thermal feedback etc. • Medium field Q slope is a combination of R BCS and R res – Not due to the difference in Trf and Tbath – Correlation between high and medium fields in unbaked cavities • Low field Q slope is likely due to residual 10 September 30, 2013 Alexander Romanenko

New superconducting measurements • Bulk muon spectroscopy – A. Grassellino et al, TUP031 • Low energy muon spectroscopy – A. Romanenko et al, TUP038 • Bitter decoration – F. Barkov et al, TUP016 11 September 30, 2013 Alexander Romanenko

Muon spin rotation B µ Contains physics aG ( t ) ~ Muon Spin Polarization Frequency – field amplitude Damping – field non-uniformity

Muon spin rotation – measure B(z) 1.0 0.8 Muon Spin Polarisation 0.6 0.4 B(z) 0.2 0.0 -0.2 -0.4 Superconductor -0.6 -0.8 in the Meissner State -1.0 0 1 2 3 4 5 6 7 8 9 10 Time ( µ s) 1.0 0.8 Muon Spin Polarisation 0.6 B ext 0.4 0.2 0.0 -0.2 -0.4 -0.6 -0.8 -1.0 0 1 2 3 4 5 6 7 8 9 10 Time ( µ s) 1.0 0.8 Muon Spin Polarisation 0.6  0.4 0.2 0.0 -0.2 -0.4 z -0.6 -0.8 0 -1.0 0 1 2 3 4 5 6 7 8 9 10 Time ( µ s)

LEM – data on EP baked/unbaked Use variable energy muons, which EP 120 um + BCP 10 um finish stop in the first ~100nm EP 120 um 3.3 keV ~15 nm - no 0.07 EP 120 um + 120C bake -1 ) Normalized stopping distribution (nm 5 keV screening Nitrogen treatment 0.06 7.5 keV 1.0 mfp ~ 2 nm at the surface, 0.05 10 keV 12.5 keV increasing deeper 15 keV 17.5 keV 0.04 20 keV 25.3 keV 0.03 0.02 mfp ~40 B/Ba 0.5 0.01 mfp > nm 0.00 400 nm 0 10 20 30 40 50 60 70 80 90 100 110 120 130 Depth (nm) BCP and EP unbaked -> strong 0.0 screening, excellent fit provided by the clean limit Pippard/BCS model Ba = 25 mT EP+120C bake-> strongly suppressed m.f.p., gradient of the m.f.p. from the 0 20 40 60 80 surface, dirty limit Average depth (nm) N-doped -> intermediate m.f.p., no Fit by Gaussian model for the field at the muon site – gradient approximate, qualitative comparison

New model of the HFQS • Main element: presence of small proximity effect coupled nanohydrides within the penetration depth – Q disease “in miniature” • Consistent with all experiments, provides quantitative description • Falsifiable – Testable predictions A. Romanenko, F. Barkov, L. D. Cooley, A. Grassellino, Supercond. Sci. Technol. 26 (2013) 035003 15 September 30, 2013 Alexander Romanenko

Neither standard 800C degassing nor “fast” cooldown help Near-surface H-rich layer is still there Typical fast cooldown of a cavity (FNAL) after standard H degassing treatments C. Antoine et al, SRF’01 T. Tajima et al, SRF’03 Integrate the H diffusion over the time spent in the precipitation temperature range T < 160K => L > 1 um All free near-surface H will precipitate into hydrides 16 September 30, 2013 Alexander Romanenko

Nanohydrides upon cooldown Not 120C baked sample Oxide Oxide ~50 nm Interstitial hydrogen Niobium hydrides “fast” cooldown T= 300K T= 2K Note drastic change in the hydrogen-related m.f.p.

Proximity effect model • Normal conducting F H Cumulative distribution function of proximity- controlled breakdown fields of hydrides R s ~ R 0 + R n * F H (H a ) hydrides of size d are Shape is superconducting by determined by the proximity effect up to distribution of hydride critical fields the field H b ~ 1/d H b H • Excellent fits High field Q slope Q disease A. Romanenko, F. Barkov, L. D. Cooley, A. Grassellino, Supercond. Sci. Technol. 26 (2013) 035003 18 September 30, 2013 Alexander Romanenko

• So what happens with 120C bake? 19 September 30, 2013 Alexander Romanenko

Positron annihilation on cavity cutouts A. Romanenko, C. J. Edwardson, P. G. Coleman, P. J. Simpson, Appl. Phys. Lett. 102 , 232601 (2013) Fine grain Large grain EP BCP • Positron annihilation spectroscopy: 120C baking results in “doping” of the first ~50 nm from the surface with defects, most likely vacancies – EP itself introduces some vacancies in ~1 um – may be the reason for more efficient 120C baking in EP cavities

Effect of 120C baking Oxide Oxide ~50 nm Free interstitial hydrogen Hydrogen is trapped by vacancies 120C baking T= 300K T= 300K A. Romanenko, C. J. Edwardson, P. G. Coleman, P. J. Simpson, Appl. Phys. Lett. 102 , 232601 (2013) 21 September 30, 2013 Alexander Romanenko

Effect of 120C baking Cooling down of 120C baked niobium Oxide Oxide No/smaller hydrides are formed due to significant portion of hydrogen trapped “fast” cooldown T= 300K T= 2K Note no change in the hydrogen-related m.f.p. – remains low 22 September 30, 2013 Alexander Romanenko

TEM evidence for nanohydrides • Direct imaging of the cross-sections of cavity cutouts in cryo-TEM [see Y. Trenikhina et al, TUP043] Look at this area with subnanometer resolution in TEM at room AND T<100K temperatures TEM See also R. Tao et al, J. Appl. Phys. 114 , 044306 (2013) and TUP042 for cryoimaging of H-reach Nb samples 23 September 30, 2013 Alexander Romanenko

Direct evidence for nanohydrides Y. Trenikhina et al, TUP043 24 September 30, 2013 Alexander Romanenko

Direct observation of large hydrides F. Barkov et al, TUP014 Growing of hydrides at T=160K in a mechanically polished sample t=0 1 min 2 min 5 min 15 min 45 min 100 min 3 hr

Further evidence: 100K and 120C baking effect T=110K T=100K • Second phase (lower concentration, lower temperature) forms at 100K – NOT observed on 120C baked samples 26 September 30, 2013 Alexander Romanenko