Post on 30-Apr-2021
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 20041
Recent Results of High Gradient Superconducting
Cavities at Cornell
RongRong--Li Li GengGeng
Brown Bag Accelerator Physics SeminarOctober 8, 2004
Cornell University
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 20042
ContentsBackgroundBackground
A brief review of high gradientsA brief review of high gradients
Gradient envelopeGradient envelope
New geometry approach for higher New geometry approach for higher EaccEacc
Reentrant cavity: concept and performanceReentrant cavity: concept and performance
New fight against field emissionNew fight against field emission
SummarySummary
AcknowledgementAcknowledgement
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 20043
DisclaimerContext: High gradient cavities for ILC.Context: High gradient cavities for ILC.
Bulk niobium cavities only.Bulk niobium cavities only.
Mainly LMainly L--Band (1300Band (1300--1500 MHz). 1500 MHz).
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 20044
Background
ITRP executive summary:The machine will be designed to begin operation at 500 GeV,with a capability for an upgrade to about 1 TeV, as the physicsrequires. This capability is an essential feature of the design.Therefore we urge that part of the global R&D design effort be focused on increasing the ultimate energy to the maximumextent feasible.
August 19, 2004: ITRP recommended SC RF for ILC
800 GeV: 35 MV/m (Achieved)1 TeV: 44 MV/m (R&D needed)
Longer linac Higher gradientTESLA type linac
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 20045
Accelerating Gradient
Enter Exit
d
v = c
Accelerating voltage
Accelerating gradient
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 20046
A brief history
Multipacting
Thermal breakdown
Field emission
1974
1986
1991
2000
Hi field Q-slope
Time
Elliptical cavity shape
Hi purity Nb
Hi pressure water rinseHi power pulsed proc.Padamsee
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 20047
Eacc Envelope Trend
Envelop defined by single-cell
Flat since 1995! Why?
First TESLA Workshopat Cornell
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 20048
Multi-cell Cavities catching up quick
Multi-cell cavity gradient is aboutto reach the envelop!
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 20049
• Eacc important for accelerator• Cavity performance determined by EM field on surface
Surface electric fieldhigh at iris, Epk
Surface magnetic field high at equator, Hpk
No fundamental limit to surface electric field
210 MV/m
Nb
145 MV/m
Nb
113 MV/m
Graber et al.,1993
3GHz Nb
Knobloch
Moffat
Delayen, Shepard
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200410
Surface magnetic field has physical limit
0
500
1000
1500
2000
1970 1980 1990 2000 2010
TE011TM010Electropolished
HR
F[O
e]
Year
HPR
Nb theoretical Hcrit, RF ~ 2000 Oe
Highest achieved Hpk 1880 Oe
State-of-the-art 1750±100 Oe
Kenji SaitoHcrit,RF
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200411
Paths toward higher EaccThe maximum feasible Eacc is determined by the RF critical magnetic field Hcrit,RF. When the surface magnetic field exceedsHcrit,RF, superconductivity breaks down into normal conductivity.
EaccH
HEacc
pk
RFcrit
/,max =
Determined bymaterial
Determined bycavity geometry
Two paths to raise Eacc: Raise Hcrit, RF or decrease Hpk/Eacc
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200412
Solution: New geometry for a reduced Hpk/Eacc.
Advantage: Cavity is based on well established Nb technology.Disadvantage: Eacc improvement is inherently small (<30%).
Nb3Sn is the only candidate having potential of higher Hcrit,RF.but the new material perspective is still remote. (remind: 40 years R&D on Nb)
Hpk in Nb cavities is close to theoretical limit, it is apparentnew material is required to significantly increase Eacc.
The high gradient need is imminent.
Bottom line: A better Hpk/Eacc alone would reach 50 MV/m;And 55 MV/m is feasible with final push in Hcrit,RF of Nb.
situation
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200413
An idea – reentrant cavity
42=accEpk
H
A reentrant cavity geometry offers a reduced Hpk/Eacc,preserving at the same time the large bore diameter at the iris.Large bore diameter is the inherent advantage of SRF.
Regular TESLA shape Reentrant shape
70 mm
8.37=accEpk
H
Hpk/Eaccreducedby 10%
Hpk/Eacc in Oe/(MV/m)
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200414
TESLA type Reentrant type
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200415
3.063.06+50+50--12.9912.99
2.882.88+40+40--12.3012.30
2.642.64+30+30--11.3611.36
2.382.38+20+20--10.0010.00
2.102.10+10+10--7.927.92
cellcell--cell cell couplingcoupling
increase in increase in surface electric surface electric field ratiofield ratio
reduction in reduction in surface magnetic surface magnetic field ratiofield ratio
(%)eδ(%)hδ (%)k
42
/2
/
accpk
accpk
EHh
EEe
=
=
1
1
==
h
eRegular TESLA
Shemelin, Padamsee, Geng, NIM A 496 (2003) 1-7.
fabricated Multi-cell: Hpk/Eacc=37.8Epk/Eacc=2.4
Single cell:Hpk/Eacc=37.9Epk/Eacc=2.19A higher Epk is the price for a lower HpkBut no fundamental limit to surface electric field
100
80
60
40
20
0 10080604020
r, m
m
Z, mm
RF optimization
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200416
Regular deep drawing/coining sufficient to build the reentrant shape
Half-cell heat treatment with Yt1200 °C 4 hoursto improve Nb thermal conductivity
RRR increased from 250 to 400-500
Fabrication
Male die
Female die
reentrancy
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200417
E beam
half-cell
Nb rod
Electropolish half-cell/beam tube
water
HF+H2SO4
spin barhalf-cell
cathode
10-15 V
magnetic stirrer
Equator EBW with a central rodto shield RF surface from Nb vapor
Fabrication
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200418
Treatment• Standard BCP after equator weld to remove residual Nb deposit.
• Flip cavity between HPR cycles for thorough rising of reentrant surface.
• Dry cavity horizontally to avoid trapped water in reentrant pocket.
• Vacuum bake out at 100 – 120 °C for 48 hour.
• Vertical electropolish single cell cavity.
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200419
Why we must have EP ?
High-field Q-slopeRapid Q decline >20MV/mFollowed by quench ~ 30MV/m
BCP = HF + HNO3 + H3PO4
Dunk etch of CEBAF cavity
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200420
Why we must have EP ?
EP + 120°C bakecures high-fieldQ-slope!
E. Kako
Reproducible 40 MV/m by usingEP plus bake
KEK horizontal EP
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200421
Acquiring EP Capability at LEPP
Half-Cell EP
Single-CellVertical EP
H2SO4(96%):HF(48%)=10:1(vol.), ~0.5 µm/min
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200422
EP: Process and results
CCO-EPContinuous current oscillation electropolish
Low surfacehydrogen50 µm
Suppressed grain boundaries(120µm BCP+30 µm CCO-EP)
Geng et al., the 11th SRF Workshop, Travemunder/Lubeck, Germany,September 8 - 12, 2003
Tajima
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200423
First EP trials produced high gradients
TESLA cavity1.3 GHz single cell
35 MV/m
Cornell/CEBAF cavity1.5 GHz single cell
36 MV/m
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200424
Reentrant Cavity Vertical RF test
Tests run at 1.8 – 2.0 K, 9 tests to date
First test reached Eacc = 27 MV/m
Soft multipacting barrier near Epk = 55 MV/m
Gas helium processing at Epk > 90 MV/m
Highest Eacc = 44.4 MV/m
Test stand has MWcapability for HPPwith short RF pulse
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200425
Achieved performance
109
1010
0 5 10 15 20 25 30 35 40 45
Q0
Eacc [MV/m]
Q0>1E10 at 30 MV/m
Eacc = 44.4 MV/mQuench limit
2KX-rays start
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200426
Reproducible quench fieldindependent of field emission level
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200427
Challenges to materialize the benefit of a lower Hpk/Eacc
109
1010
0 5 10 15 20 25 30 35 40 45
Q0
Eacc [MV/m]
X-rays start
Epk=97 MV/mX-ray=60 R/h
Field emission !
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200428
Field Emission Theory
QM tunneling theory predicts exponetialFowler-Nordheim emission current ⎟
⎠⎞
⎜⎝⎛−=
E
CECjFN
221 exp
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200429
Field Emission
• Acceleration of electrons drain cavity energy.• Impacting electrons produce line heating.• Impact also produces bremsstrahlung x rays.
Nearly all emission is associated withmicroscopic foreign particles.
Knobloch
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200430
Avoiding Field Emission
High Pressure water rinsing
Pressurized (100 bar) DI water jetto remove surface particles
Kneisel
Reschke
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200431
Avoiding Field Emission
TTF 9-cell cavity
Clean Room Assembly
Pumping and venting withoutre-contaminating surface
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200432
Desired low Hpk/Eacc results in a higher Epk/Eacc(remind: traditional wisdom is to lower Epk/Eacc!)
Reentrant cavity faces FE challengesin untested regime
Eacc[MV/m] Epk[MV/m]45 10050 11055 120
DESY: max. Epk achieved 77 MV/mLEPP: max Epk achieved 70 MV/m
⎟⎠⎞
⎜⎝⎛−=
E
CECjFN
221 expToday
Future
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200433
Reentrant cavity faces FE challengesin untested regime
Larger Hi field surface areahigher chances to catch particles
TTF
reentrant
Shemelin
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200434
Reentrant cavity faces “pocket complexity”
Solution:• Longer HPR time• Shake and horizontal drying
• water jet shadowing• restricted drain
Need R&D:• understand jet/particle interaction• improve nozzle design• more radical ideas?
Easy drain
elliptical
reentrant
Puddle after HPR
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200435
Field Emission – in situ processing • RF processing • Helium processing• High Power Processing (HPP)
Destroy emitting particles by inducing RF sparking
Knobloch Knobloch
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200436
In-situ processing
1.3GHz, 1 MW 200µs pulse
• RF (CW and pulsed ) processing at low field effective.• But RF processing at high field activates new field emitters.• Helium processing at high field is effective, 17% gain obtained.• To get best result, He proc. should be followed by partial warm up.
• LEPP has HPP capability at 1.3 GHz.• Past HPP was very successful.
Successful HPP requires much higher Epk in short pulse.
Max. Epk(pulse) limited by Hcrit,RF
(Hpk already reach 1700 Oe) It is worth trying!
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200437
High gradient cavity comparison
800
1000
1200
1400
1600
1800
2000
40 50 60 70 80 90 100 110
Epk [MV/m]
Hpk
[Oe]
LE1-35
1B9 Cornell test
LR1-2, Feb-04
LR1-2 MLP
Eacc = 27
Eacc = 49
Eacc = 37
Eacc = 40
Un-labeled points refer to KEK and DESY cavitiesEacc in MV/m
July-04
Eacc = 37
July-14
Eacc = 40
July-16
Eacc = 42.6
1B9 DESY test
Eacc = 44
July-21
Eacc = 44.4
World high gradient data (regular elliptical shape)
Reentrant shape with δh=-10%
Reentrant cavitypotential Eacc=49MV/m
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200438
Gradient Status UpdatedThis is a small step
buttrend line is changed!
Ultimate gradient ?
Let’s explore !First TESLA Workshop
SRF chosen for ILC
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200439
SummaryAchieved Eacc in excess of 44 MV/m in Nb cavity.
Established in-house EP procedure providing reproducible Hi Eacc.
Demonstrated usefulness of helium processing at Hi gradient.
Next step:Keep pushing Eacc by more EP to ~ 50 MV/m.
Must reduce field emission: improve HPR nozzle; HPP.
Identify/control sources of particulate contamination (facility!).
Rong-Li GengBrown Bag Accelerator Physics
Seminar, October 8, 200440
Acknowledgement
RF design of reentrant cavity byValery Shemelin
In building/testing reentrant cavity,crucial contributions made by
Curtis CrawfordJoe KirchgessnerAndy SeamanJames Sears
This work is supported by the National Science Foundation.