[PPT] - Status of 650MHz Cavities for PIP-II In partnership with: Martina PowerPoint Presentation

SLIDE 1

In partnership with: India/DAE Italy/INFN UK/STFC France/CEA/Irfu, CNRS/IN2P3

Status of 650MHz Cavities for PIP-II

Martina Martinello PIP-II β=0.90 & 0.92 Jacketed Cavity FDR 1 February 2019

SLIDE 2

Outline

Martina Martinello

High Q optimization:

– Intro on N-doping treatment – Early tests on 650 MHz – 5-cells cavity results – Processing optimization for higher Q in cryomodule – Trapped flux sensitivity measurements

Instrumentation for prototype cavities
Summary

2/1/2019 2

SLIDE 3

N-doping treatment: how is done

800 C (3 hours + duration of doping) 25 mTorr (2 minutes)

Example of a N-doping process (2/6 recipe):

Nb bulk EP cavity annealed

for 3 hours in vacuum (UHV furnace) at 800C

Nitrogen injected (25 mTorr)

at 800C for 2 minutes

Cavity stays for another 6

minutes at 800C in vacuum

Cooling in vacuum
5 um electro-polishing (EP)

Cavity after welding EP 140 μm 3 h at 800C UHV baking N2 injection 800C (2-30 minutes) EP 5-10 μm

Martina Martinello

800C w/o N2 (0-60 minutes)

2/1/2019 3

SLIDE 4

N-doping treatment: how is done

Example of a N-doping process (2/6 recipe):

Nb bulk EP cavity annealed

for 3 hours in vacuum (UHV furnace) at 800C

Nitrogen injected (25 mTorr)

at 800C for 2 minutes

Cavity stays for another 6

minutes at 800C in vacuum

Cooling in vacuum
5 um electro-polishing (EP)

Cavity after welding EP 140 μm 3 h at 800C UHV baking N2 injection 800C (2-30 minutes) EP 5-10 μm

Martina Martinello

800C w/o N2 (0-60 minutes)

Caps to avoid diffusion of furnace contaminations

2/1/2019 4

SLIDE 5

N-doping treatment: interstitial N in Nb

Martina Martinello

N Nb N Interstitial

Only Nb from TEM/NED spectra: N must be interstitial

Final RF Surface

Y. Trenikhina et Al, Proc. of SRF

2015

2/1/2019 5

SLIDE 6

N-doping treatment: performance improvement with field

Martina Martinello

𝑆𝑇 2 𝐿, 𝐶𝑈𝑠𝑏𝑞 = 𝑆𝐶𝐷𝑇 2 𝐿 + 𝑆0 + 𝑆𝐺𝑚 ( 𝐶𝑈𝑠𝑏𝑞, 𝑚 )

2 4 6 8 10 12 14 16 18 4 6 8 10

standard treatment standard treatment nitrogen treatment nitrogen treatment

R

2K BCS (n)

Eacc (MV/m)

5 10 15 20 25 30 35 40 10

9

10

11

Q0 Eacc (MV/m)

T= 2K

Anti-Q-slope emerges from the BCS surface resistance decreasing with field Anti-Q-slope

A. Grassellino et al, Supercond. Sci. Technol. 26 102001 (2013) - Rapid Communications
A. Romanenko and A. Grassellino, Appl. Phys. Lett. 102, 252603 (2013)

2/1/2019 6

SLIDE 7

Outline

Martina Martinello

High Q optimization:

– Intro on N-doping treatment – Early results on 650 MHz – 5-cells cavity results – Processing optimization for higher Q in cryomodule – Trapped flux sensitivity measurements

Instrumentation for prototype cavities
Summary

2/1/2019 7

SLIDE 8

HB650 Single-cell Early Test Results

Martina Martinello

PIP-II specs

120C baked cavities not always

meet specs

N-doping capable to double

the Q-factor at medium field, sometimes affected by early quench

World record Q-factor of 7e10

at 2K,17 MV/m and 650 MHz with N-doping

2/1/2019 8

SLIDE 9

HB650 Single-cell Early Test Results

Martina Martinello

PIP-II specs

120C baked cavities not always

meet specs

N-doping capable to double

the Q-factor at medium field, sometimes affected by early quench

World record Q-factor of 7e10

at 2K,17 MV/m and 650 MHz with N-doping

Results were very promising with N-doping, however:

1. large variability was observed with both

processing (N-doping and 120C baking)

2. No anti-Q-slope observed with N-doped

cavities

2/1/2019 9

SLIDE 10

Outline

Martina Martinello

High Q optimization:

– Intro on N-doping treatment – Early tests on 650 MHz – 5-cells cavity results – Processing optimization for higher Q in cryomodule – Trapped flux sensitivity measurements

Instrumentation for prototype cavities
Summary

2/1/2019 10

SLIDE 11

HB650 5-cells Tests Results (all N-doped + 20um EP)

Martina Martinello

PIP-II specs

B9A-AES-010 B9A-AES-009 B9A-AES-008

Light N-doping applied to

650 MHz cavities: 2/6 N- doping + 20um EP

3 N-doped 5-cells 650 MHz

cavities meet PIP-II specification

B9A-AES-010

will be dressed with He vessel, the

thers will be re-processed

to improve performance

2/1/2019 11

SLIDE 12

Light N-doping applied to

650 MHz cavities: 2/6 N- doping + 20um EP

3 N-doped 5-cells 650 MHz

cavities meet PIP-II specification

B9A-AES-010

will be dressed with He vessel, the

thers will be re-processed

to improve performance

HB650 5-cells Tests Results (all N-doped + 20um EP)

Martina Martinello

PIP-II specs

B9A-AES-010 B9A-AES-009 B9A-AES-008

Very light N-doping treatment was chosen

Pro: no early quench observed
Cons: very little doping effect remains

5um EP 20um EP

LCLS-II processing

2/1/2019 12

SLIDE 13

Outline

Martina Martinello

High Q optimization:

– Intro on N-doping treatment – Early tests on 650 MHz – 5-cells cavity results – Processing optimization for higher Q in cryomodule – Trapped flux sensitivity measurements

Instrumentation for prototype cavities
Summary

2/1/2019 13

SLIDE 14

Frequency dependence of RBCS(Eacc)

Martina Martinello

N-doped cavities at 650 MHz

do not show the RT reversal (anti-Q-slope) typically

bserved at 1.3 GHz

N-doping 120C baking

Also

for 120C baked cavities, the field dependence

f

RT is unfavorable at low frequencies

The

physical mechanism underneath the reversal of RBCS (here called RT) has a stronger effect at high frequencies

650 MHz cavities

M. Martinello et al., Phys. Rev. Lett. 121, 224801 (2018)
Optimization of processing specifically for 650

MHz is needed!!

2/1/2019 14

SLIDE 15

HB650 Single-cell R&D program

Martina Martinello

Intensive processing optimization is being pursue starting from

2018:

GOAL: reach the highest possible Q at medium/high field for

650 MHz cavities

Flux expulsion and trapped flux sensitivity will be also
ptimized for Q preservation in cryomodule
Surface treatments under studies: EP, BCP for baseline and

modified 120C (75-120C baking), N-doping for Q improvement

Trapped flux sensitivity will be acquired for each treatment to

understand magnetic flux shielding requisition in cryomodule

2/1/2019 15

SLIDE 16

All data acquired in 2018

Martina Martinello 2/1/2019 16

SLIDE 17

EP vs N-doping (B9AS-RRCAT-301)

Martina Martinello

EP N-doping (2/6 + 5um EP)

2/1/2019 17

SLIDE 18

EP vs N-doping and 75-120C baking (B9AS-PAV-104)

Martina Martinello

75-120C baking N-doping (3/60 + 5um EP) EP

2/1/2019 18

SLIDE 19

Outline

Martina Martinello

High Q optimization:

– Intro on N-doping treatment – Early tests on 650 MHz – 5-cells cavity results – Processing optimization for higher Q in cryomodule – Trapped flux sensitivity measurements

Instrumentation for prototype cavities
Summary

2/1/2019 19

SLIDE 20

Trapped Flux Sensitivity Measurements

Martina Martinello

N-doped cavities show significant larger sensitivity that other treatments

2/1/2019 20

SLIDE 21

Minimizing Losses in Cryomodules due to Trapped Flux

Martina Martinello

Even though final processing of 5-cells

650 MHz cavities is not finalized yet, it is necessary to minimize as much as possible flux trapping, especially knowing that N-doped cavities show larger sensitivity

In order to do that, we are:
Maximizing flux expelling efficiency: all our 5-cells are being

treated at 900C for 3 hours, this treatment is known to relief stress and dislocations and cause flux trapping during cooldown

Minimizing remnant magnetic field in cryomodule: design of

magnetic shielding is in process and will take into account sensitivity of production processing

Instrumenting cavities with thermometers and fluxgates in order

to monitor the magnetic field in-situ and the cooldown properties

2/1/2019 21

SLIDE 22

Outline

Martina Martinello

High Q optimization:

– Intro on N-doping treatment – Early tests on 650 MHz – 5-cells cavity results – Processing optimization for higher Q in cryomodule – Trapped flux sensitivity measurements

Instrumentation for prototype cavities
Summary

2/1/2019 22

SLIDE 23

Martina Martinello

Flux-gates location

In order to understand flux expulsion efficiency during cooldowns, simulations suggested that the flux-gates need to be placed as follow:

Longitudinally between irises (Bsc / Bnc = 0.18)
Vertically between irises (Bsc / Bnc = 0.85)

In this way the variation of the field (Bsc/Bnc) during the SC transition, after complete Meissner effect, is maximized and the fraction of field trapped/expelled can be estimated. The simulations, courtesy of Iouri Terechkine, take into account REAL magnetic field environment in cryomodule and integrate the results within the active length

f the fluxgate

2/1/2019 23

SLIDE 24

Martina Martinello

Flux-gates location In order to detect B generated by thermo-currents a transverse fluxgate is needed Example of magnetic field generated by thermo- current during a fast cooldown of an LCLS-II cryomodule

2/1/2019 24

SLIDE 25

Martina Martinello

Thermometers location Fluxgate sensor Thermometer Thermometers will be placed in different locations on cell #1, 3, 5 to monitor both vertical and longitudinal thermal-gradient during cooldown

2/1/2019 25

SLIDE 26

Outline

Martina Martinello

High Q optimization:

– Intro on N-doping treatment – Early tests on 650 MHz – 5-cells cavity results – Processing optimization for higher Q in cryomodule – Trapped flux sensitivity measurements

Instrumentation for prototype cavities
Summary

2/1/2019 26

SLIDE 27

Summary

Martina Martinello

The production processing for 650 MHz cavities is still under investigation, the

treatment with the best compromise between BCS, residual and trapped flux surface resistance at ~20 MV/m will be chosen;

A very light N-doping treatment was applied to 5-cell cavities giving mostly

good results. The best 5-cell cavity will be dressed within next days, the other have been reset and treated with 900C baking to maximize flux expulsion. They will be soon re-processed and will be dressed once specs are met;

Magnetic shielding design will be finalized after the production processing is

chosen since that will set the maximum magnetic field allowed in cryomodule;

Some of the 5-cell cavities will be instrumented, before being dressed, with

fluxgates and thermometers to monitor: remnant magnetic filed, flux expulsion efficiency, magnetic field due to thermo-currents and cooldowns dynamic directly in the cryomodule.

2/1/2019 27