View
1
Download
0
Category
Preview:
Citation preview
Alexey Kochemirovskiy
The University of Chicago/Fermilab
MuCool Test Area Experimental
Program Summary
Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016)
Outline
• Introduction
– Motivation
– MTA facility
• 201 MHz MICE cavity program
• Gas filled cavity program
• Study of vacuum RF breakdown in 805MHz
cavities
• Conclusion
August 23, 2016 Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016) 2
Why are we interested in Muons?
August 23, 2016 Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016) 3
Muon accelerator R&D is focused on developing a
facility that can address critical questions concerning
two frontiers:
• The Intensity Frontier with Neutrino Factory
generating well-characterized neutrino beams for
precise high sensitivity study
• The Energy Frontier with a Muon Collider capable
of probing multi-TeV energies
Muon Ionization Cooling in called for
August 23, 2016 Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016) 4
Buncher
PhaseRotator
InialCooling
CaptureSol.
ProtonDriver FrontEnd
MW-ClassTarget
Accelera on
DecayChannel
µ StorageRing
ν
281m
Accelerators:Single-PassLinacs
0.2–1GeV
1–5GeV
5GeV
ProtonDriver Accelera on ColliderRing
Accelerators:Linacs,RLAorFFAG,RCS
Cooling
µ+
6DCooling
6DCooling
FinalCooling
Bunch
Merge
µ−
µ+ µ−
Share same complex
n Factory Goal: 1021 m+ & m- per year within the accelerator
acceptance
NeutrinoFactory(NuMAX)
MuonCollider
m-Collider Goals: 126 GeV
~14,000 Higgs/yr
Multi-TeV Lumi > 1034cm-2s-1
ECoM:
HiggsFactoryto
~10TeV
Cool-ing
InialCooling
ChargeSep
arator
ν µ+
µ−
Buncher
PhaseRotator
CaptureSol.
MW-ClassTarget
DecayChannel
FrontEnd
SCLinac
SCLinac
Accumulator
Buncher
Accumulator
Buncher
Combiner
Muon cooling concept
August 23, 2016 5
Short lifetime of muon(2.2µs) require accelerating gradients of tens MV/m
Gradient is restricted by RF breakdown, rate of which should be kept small
It was experimentally shown that strong magnetic fields aggravates problem
Study of RF breakdown in strong magnetic field is needed
Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016)
Fermilab’s Mucool Test Area (MTA)
Facility built specifically for muon cooling hardware R&D
• Capacity to test 201 and 805MHz
cavities in strong magnetic field
• H- beamline passes through the
center of magnet bore
• Infrastructure for clean room
assembly and inspection
• Extensive instrumentation for BD
characterization
• Run control system to detect
breakdown events and
record relevant data streams
August 23, 2016 6
MagnetCavity Beamline
Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016)
Data acquisition and run control system
August 23, 2016
• Trigger system for breakdown detection and run control
• Fast oscilloscopes to record time sensitive signals <- Labview
– Cavity pickups
– Light signal from optical fibers
– Scintillators ( X ray detection)
– Forward and reflected power
– Radiation detectors
• Vacuum pressure data
• Temperature sensors
• Acoustic spark localization
• Etc.
7
Run control station at the linac gallery
Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016)
MTA experimental programs
• 201MHz MICE cavity
• Gas filled cavity program
• 805MHz vacuum program
August 23, 2016 Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016) 8
MTA experimental programs
• 201MHz MICE cavity
• Gas filled cavity program
• 805MHz vacuum program
August 23, 2016 Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016) 9
International Muon Ionization Cooling
Experiment (MICE)
August 23, 2016 Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016) 10
• 201 MHz cavity module very similar to that of the full ionization cooling
demonstration in MICE tested in MTA
• Design gradient of 10.3 MV/m, with cavities operated in fringe fields of multi-
Tesla solenoid magnets
• MTA has a 5-T superconducting solenoid that provides operational conditions
similar to MICE channel
201MHz MICE cavity
August 23, 2016 Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016) 11
• Pillbox geometry (~1m in
diameter, 40cm RF gap)
• External vacuum vessel
• Mechanical tuners
• Extensive instrumentation
MICE cavity program: operation results
August 23, 2016 Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016) 12
• The experimental program complete
• Cavity has been tested under operating conditions similar to MICE
• Stable performance is demonstrated in both B=0T and B=3T external
field configurations with copper and beryllium windows
• Zero breakdown rates at MICE design peak gradient (~10.3MV/m)
• Measured radiation rates are within limits for tracker backgrounds
• Gained a lot of experience in surface preparation, design and operation
MTA experimental programs
• 201MHz MICE cavity
• Gas filled cavity program
• 805MHz vacuum program
August 23, 2016 Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016) 13
Helical Cooling Channel
August 23, 2016 Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016) 14
• The main idea is to combine absorber
and RF cavity together
• Gas filled RF cavities operating inside
helix of solenoids
• Emittance exchange from dispersion
in magnetic field – 6D cooling
• Pressured gas inside the cavities acts
both as an absorber for beam cooling
an as RF breakdown mitigator
• Hydrogen is an ideal absorber due to
large radiation length and stopping
power
Gas breakdown vs metal breakdown
August 23, 2016 Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016) 15
Maximum stable gradient in hydrogen filled cavity for different electrode materials
• Maximum gradient increases with gas pressure (Paschen curve)
• Gradient behavior gets dominated by metallic RF breakdown at high pressures
• Baseline 20MV/m gradient is demonstrated for gas pressure as low as 20 atm
• Gradient is not affected by external magnetic field (green vs magenta)
Pressured gas cavity – beam considerations
August 23, 2016 Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016) 16
• Beam passing through the cavity must not trigger breakdown –
experiments at MTA showed it does not
• Resultant plasma loads the cavity which leads to degradation of
accelerating gradient - needs to be minimized
Plasma loading effect in gas filled cavity
Factors effecting plasma loading:
• Gas pressure
• Dopant concentration
• Beam intensity
Dielectric Loaded Pressurized Cavity
August 23, 2016 Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016) 17
HCC design challenge: 325 & 650 MHz pillbox cavities do not fit in bores of
present magnet technology
Possible solution – load cavity with dielectric to decrease resonant frequency
Alumina - ideal dielectric candidate due to low losses
Specially designed alumina donut insert to study effect on RF breakdown
Dielectric Loaded High Pressure Cavity with alumina donut insert
Dielectric Loaded High Pressure cavity -
performance
August 23, 2016 Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016) 18
• Gas breakdown observed up to ~10atm pressure
• Comparison with baseline “no-insert” performance shows gradient
limited by alumina
• No clear dependence of gradient on alumina purity
High pressure gas cavities: summary
August 23, 2016 Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016) 19
• Demonstrated that high pressure gas filled cavities can be
operated in external multi-Tesla magnetic fields
• Baseline 20MV/m gradient in multi-Tesla field is reachable with
hydrogen pressure as low as 20atm
• Gas breakdown and metallic breakdown curves investigated
for different gas species/dopants and metal material
• Established breakdown limits for alumina of 11-17MV/m
depending on alumina purity and gas composition
MTA experimental programs
• 201MHz MICE cavity
• Gas filled cavity program
• 805MHz vacuum program
August 23, 2016 Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016) 20
Why do we see the deterioration of cavity
performance in strong B fields?
August 23, 2016
• Dark current: electron field emission from surface
imperfections
• B field focuses dark current into “beamlets”
• Beamlets cause pulsed heating of the surface
• Pulsed heating leads to surface degradation
• Breakdown is triggered
Potential mitigations:
• Surface treatment
• Use higher radiation length materials ( for ex. Be)
• Decrease impact energy density of electrons
D.Stratakis , J.Gallardo, R.Palmer, Nucl. Inst. Meth. A 620 (2010), 147-154
21Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016)
Mitigation technique: surface treatment
August 23, 2016
201 MHz MICE cavity - electro
polished, clean room environment
Conditioned to design gradient of
10.3MV/m with no breakdowns
All Seasons Cavity – stainless
steel with copper coating
Modular Cavity - chemically
polished, clean room environment.
45MV/m in 0T field reached
Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016) 22
Mitigation technique: geometry
• Box cavity: magnetic insulation
August 23, 2016
E
B
Magnetic field deflects
dark current electrons
to low E field regions
Works well only for angles very close to 90o
Small shunt impedance compared to pillbox cavities
Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016) 23
Mitigation technique: geometry
August 23, 2016
All Seasons Cavity: longer RF gap length
15 cm
Electron energy as a function of
cavity length
Θ = 0o
Θ = 30o
Θ = 60o
Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016) 24
Mitigation Technique: geometry
August 23, 2016
Grid windows
• Allow the beam to pass though
• Increase shunt impedance of the cavity
• Allow dark current beamlets to exit cavity volume
Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016) 25
Performance of various 805MHz cavities
August 23, 2016
Factors that affect fit quality:
• Condition history
• Local field enhancement around coupler regions
• Surface treatment
Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016) 26
No surface prep
for this cavity
Pillbox 805MHz “modular” cavity
August 23, 2016 27
• End walls can be un-mounted easily
• Allows for end wall material swap
• Low E fields in the coupler region
Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016)
• Allows for careful control over experimental conditions
• Evaluate different materials
• Perform frequent inspections to track surface state
Goal: to build a coherent picture of processes inside the cavity during breakdown
Unique design features:
Surface inspection after B=3T run
August 23, 2016 28
All clearly visible damage was inflicted during B=3T run (!)
Endplate after B=0T run with peak gradient of 50MV/m
Endplate after B=3T run with peak gradient of 12MV/m
First time inspections were carried out separately after run at zero
magnetic field and run at high magnetic field
Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016)
Inspection after B=3T run: damage
microstructure
August 23, 2016 29
Typical
breakdown
damage
• Characteristic diameter of a BD damage ~1.5mm
with melted core up to ~70um deep
• Traces of splashing
• Damage is much more “violent” after B=3T run,
although stored energy is 16 times lower than in
B=0T run
Splashing traces around BD damage
5mm
2mm
Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016)
0.5mm
0.2mm Crater depth ~ 30µm
Inspection after B=3T runs: damage pattern
August 23, 2016
30
Endplate Center
Coupler
Map of “volcanos”
• Perfect 1-to-1 correspondence between
354 “volcanos” on each endplate (new
result) - supports the model of BD being
induced by focused dark current
• Damage distribution is denser in high E field
region
• Mystery: detected 136 sparks, but observed
354 damage sites
Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016)
Next set of measurements will be with
Beryllium endplates
August 23, 2016 31
• Radiation length of Beryllium (~35cm) for electrons
is higher than of copper (~1.4cm)
• Mitigation of breakdown triggers on the surface
• Better gradient performance
There are several measurements enabled by Beryllium:
– Direct measurement of dark current (Faraday
Cup)
– Measurement of transverse emittance of dark
current beamlets (film/glass)
Inner surface of Beryllium endplate
2.6 mm-thick window
What we expect to observe:
*Beryllium – copper configuration is also being discussed
Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016)
Conclusion
• We have gained a lot of hands-on experience operating cavities in external magnetic fields at MTA facility
• Applying advanced surface preparation techniques resulted in meeting design gradients for 201MHz MICE cavity
• We have demonstrated > 20 MV/m operation of 805 MHz vacuum cavities at B = 5 Tesla. That is an encouraging result towards demonstrating the feasibility of vacuum cavity-based ionization cooling channel
• Using RF cavities with high-pressure gas, we have demonstrated a general solution to the cooling problem
• A lot of interesting results coming from studying the problem of RF breakdown in strong magnetic field
• The work is still ongoing, including data analysis and high-power tests
August 23, 2016 Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016) 32
“Thank you” slide
August 23, 2016 Alexey Kochemirovskiy | NuFact'16 (Quy Nhon, August 21-27, 2016) 33
Thank you!
Recommended