Electron Cloud in Positron Rings and - PowerPoint Presentation

Slide1

Electron Cloud in Positron Rings and Intra-beam Scattering

Mauro Pivi, MedAustron - work made while at SLAC -

i

n

collaboration

with: T.

Demma

(Frascati & LAL), the ILC / CLIC and

SuperB

Working

Groups

,

colleagues

at

LBNL,

Cornell

University

and SLAC.

Slide2

Intra-beam scattering

Intra-beam

scattering (IBS) is associated with multiple

small

angle scattering events leading to emittance growth.

In most

storage

rings,

typical radiation damping times are shorter than IBS growth times and IBS

effect is not

observed

.

However, for high population and ultra-low emittance bunches, IBS may lead to emittance increase



IBS important for future colliders.

Slide3

IBS Formalisms, models and simulation tools

M.

Boscolo

, USR Workshop, Oct.30th 2012

3

Piwinski

,

Bjorken and Mtingwa formalisms First formalisms ‘70/’80 for calculating IBS growth rates in storage rings based for Gaussian bunch distributionsK. Bane modelHigh energy approximation for Gaussian beamsA. Chao modelNovel analytical model, coupled differential eqs. valid for Gaussian beamsSemi-analitical model Using fit parameters from simulations iteratively, estimate emittanceMonte Carlo macroparticle tracking code (T. Demma et al.)6-D Monte Carlo, realistic studies for non-Gaussian beam distributions‘IBS-Track’ based on Zenkevich-Bolshakov Algorithmaims at exploring final equilibrium for non-Gaussian beams. Tails.ex, ey and ez evolution in time

Methods

are in good agreement

Slide4

During the

two particles

small angle collision, the

momentum

change for 1 particle

can be expressed as:

with the equivalent polar angle

eff and the azimuthal angle  distributing uniformly in [0; 2], the invariant changes caused by the equivalent random process are the same as that of the IBS in the time interval tsIBS – Monte Carlo code based on Zenkevich-Bolshakov AlgorithmM. Pivi, December 2015

Slide5

Intrabeam Scattering in SuperB LER

Parameter

Unit

Value

Energy

GeV

4.18

Bunch population

10

10

6.5

 

Circumference

m

1257

 

Emittances (H/V)

nm/pm

1.8/4.5

 

Bunch Length

mm

3.99

 

Momentum spread%0.0667 Damping times (H/V/L)ms40/40/20 N. of macroparticles-105 N. of grid cells-64x64x64 

Bane PiwinskiIBS-Track

Bane PiwinskiIBS-Track

T.Demma, INFN

December 2011

Slide6

Emittance Evolution in SuperB LER

SuperB V12 LER

Nb= 2x10

10

- 12x10

10

F=10 tx = 10-1 x 40 ms ty = 10-1x 40 ms ts = 10 -1x 20 msT.Demma, INFNDecember 2011

Slide7

Simulations for

SuperB

M.

Boscolo

, USR Workshop, Oct.30th 2012

7

The easy computable semi-analytical approach allows a quick scan of some key design parameters, such as the bunch population

Equilibrium horizontal emittance vs bunch currentEquilibrium longitudinal emittance vs bunch current

Slide8

Intra-Beam Scattering (IBS) Simulation Algorithm: CMAD

The Monte Carlo IBS routine was imported in the

CMAD (M.P.) code by

Theo

Demma

.

CMAD

parallel code: Collective effects & MADAccelerator lattice uploaded from MADX filesBunch particles are 6D tracked along the ring.The IBS scattering routine is called, at each element in the ring. IBS method:All beam particles are grouped in cells.Each 2 particles within a cell are coupled.Momentum of 2 particles is changed due to scattering.Radiation damping and excitation effects are evaluated at each turn. Code physics: Electron Cloud + IBS + Radiation Damping & Quantum Excitation

IBS applied at each element of the Ring

Mauro

Pivi

, CERN, CLIC

May 9-11, 2012

M.Pivi

,

A.Chao

,

C.Rivetta

,

T.Demma

,

M.Boscolo

,

F. Antoniou

, K.Li, Y.Papaphilippou, K.Sonnad, IPAC2012T. Demma, M. Pivi

Slide9

IBS modeling: animation

http://www-user.slac.stanford.edu/gstewart/movies/particlesimulation_animation/

Slide10

IBS - SuperB LER

Parameter

Unit

Value

Energy

GeV

4.18

Bunch population10106.5 Circumferencem1257

 

Emittances (H/V)

nm/pm

1.8/4.5

 

Bunch Length

mm

3.99

 

Momentum spread

%

0.0667

 

Damping times (H/V/L)

ms

40/40/20

 N. of macroparticles-105 N. of grid cells-64x64x64 Bane PiwinskiIBS-TrackT. Demma (INFN), M. Pivi (SLAC)IBS-Track C-MADDecember 2011

Slide11

Emittance

Evolution

in

SuperB

LER

M. Pivi (SLAC), T.

Demma (INFN)

Slide12

IBS Distribution

study

:

tails

Parameter

c

2

799ConfidenceZ 1857.56<1e-6X 1455.68<1e-6Y 778.2280.6920

T.

Demma

(

INFN)

,

M

. Pivi (SLAC

)

Slide13

SIRE: IBS Distribution

study

in

CLIC

DR:

tails

ParameterValueEq. ex (m rad)2.001e-10Eq. ey (m rad)2.064e-12Eq. sd 1.992e-3Eq. sz (m)1.687e-3Parameterc

2

999

Confidence

D

p

/p

3048.7

<1e-15

X

1441.7

<1e-15

Y

1466.9

<1e-15

A.

Vivoli

(CERN)

Slide14

IBS - Swiss Light Source (SLS)

IBS_Track

, T

.

Demma

(INFN)

Evolution of the emittances obtained by tracking with IBS for different bunch populations. Horizontal lines: Piwinski (full) and Bane (dashed) models for the considered bunch populations.Comparison with experimental data at SLS: F. Antoniou et al. IPAC 2012-- 6×109 ppb-- 60 ×109 ppb-- 100 ×109 ppb

Slide15

Summary IBS

Both tracking codes

(INFN/CERN

),

that implement the

Zenkevich-Bolshakov

algorithm, successfully benchmarked with conventional theories (i.e. K. Bane) and with the novel semi-analytical model.

Monte Carlo code was implemented in the parallelized CMAD codeComparison between theoretical models and multi-particle algorithms, give good agreement for IBS dominated regimes.IBS features that cannot be studied by analytical models such as the impact on the damping process and the generation of non- Gaussian tails can be investigated with multi-particles tracking codes.Code benchmarked with SLS real data [F. Antoniou et al., IPAC2012], planned also with CESR-TA data.Multi-particle codes are suited for studies of ultra-low emittance beams for future colliders.15

Slide16

In a positron or proton storage ring, electrons are generated by a variety of processes, and can be accelerated by the beam to hit the vacuum chamber with sufficient energy to generate multiple “secondary” electrons (multipacting).

Under certain conditions, the “electron cloud” density can reach high levels and can drive

the beam unstable

,

increase the beam

emittance, vacuum etc. decreasing

the collider performances.Electron cloud effect in proton and positron storage rings25 nsElectron cloud in the LHC25 ns

Slide17

Surface measurements at SLAC. F. Le

Pimpec

et al.

For convenience the SEY maximum value is often quoted.

The Secondary Electron Yield (SEY) on a surface: Key Parameter for Electron

M

ultiplicationThe secondary electron yield (SEY) is the number of electrons emitted per primary incident electron. It depends on:the energy of the incident electronSurface treatment and history

Slide18

Observations of Electron Cloud

KEK-B accelerator, Japan: the vertical bunch size increases along the train due to the build-up of the electron cloud density.

Electron cloud has been observed in several accelerators

including: PEP-II, DA

F

NE, CESRTA,

CERN

SPS and at LHC.H. Fukuma et al.J. Flanagan et al.

Slide19

Electron cloud assessment Linear Colliders working group: Worldwide Laboratory

E

ffort

Development of mitigation techniques:

e

-

conditioning, surface coatings, clearing electrodes, grooves, solenoids

SEY measured on samples placed in accelerator environments: SLAC, CERN, KEK, CesrTA, DafneInstability simulations: to determine instability thresholdBuild-up & evolution simulations: fed with measured SEY to evaluate level of electron cloud in the acceleratorConverge to recommendation for adoption of mitigationsJan 18, 2010 ILC BAW-2Global Design Effort19

Slide20

Electron Cloud Effect Mitigations

Electron “scrubbing” or “conditioning”:

electrons impinging on the surface.



decrease of SEY

linked to surface “graphitization”

Secondary electron Yield (

dmax) vs electron dose for different electron energies on LHC beam screen colaminated Cu. R. Cimino, T. Demma, M. Commisso, D. R. Grosso, V. Baglin, R. Flammini and R. Larciprete Phys. Rev. Lett. 109, 064801 – 2012

Slide21

“Clearing” electrodes capture electrons in the time between bunches.

Electron Cloud

Technical Mitigations

Surface coating

with low Secondary Electron Yield material

Solenoid magnetic field modifies electron dynamics

Surface with grooves

confine electrons e-

Slide22

Recommendation of Electron Cloud Mitigations

22

Clearing Electrodes

KEK

Grooves w/

TiN

coating, KEK/SLAC

CESRTAClearing ElectrodeGrooves on CuStable StructuresReliable Feedthroughs

Manufacturing Techniques

& Quality

amorphous-Carbon, CERN

Slide23

Evaluation of Electron Cloud Effect

Used two categories of simulation codes:

The

build-up codes

: follow the evolution in time of the electron cloud interacting with a stable (fixed)

beam.

Secondary electron yields as measured in accelerator environments and all technical mitigations are included in simulations.

The beam instability codes: assume already formed clouds with given density and track the beam particles.used to define the cloud density that results in an instability threshold.

Slide24

Beam Instability Code: CMAD

CMAD

(M. Pivi, Theo

Demma

, Kiran Sonnad, Claudio Rivetta

):The code simulates:

electron cloud instability. (Mauro P., Kiran S.)Intra-beam scattering (Theo D.)Feedback system to mitigate electron cloud instability (Claudio R.)The accelerator model is uploaded via MAD-X.The code tracks the beam for several turns and computes the electromagnetic interaction between particles in the beam and the electrons in the cloud. Parallelized code.CMAD has been used at a number of institutions including CesrTA Cornell University, Frascati Laboratory Italy and SLAC. M. Pivi, in the Proceedings PAC07 Conference THPAS066 (2007)

Slide25

Electron

Cloud

instability threshold

4.4e11

3.9e11

3.5e11

Cloud density (e/m

3)3.2 km ILC Damping RingInstability simulations for the International Linear Collider Positron Damping Ring. Instability threshold ~2×1011 e/m3

Slide26

Build up simulations: Quadrupole in wiggler section

Electron cloud density (e/m

3

) Electron energies (

eV

)

J. Crittenden, Cornell U.

Slide27

Sextupole in TME arc cell

Electron cloud density (e/m

3

) Electron energies (

eV

)

J. Crittenden, Cornell U.

Slide28

Build-up simulations: Model of clearing electrode in wiggler magnets

Modeling of clearing electrode: round chamber is used

Clearing Field (left) & potential (right)

L. Wang, SLAC

Slide29

detail

+600V

0v

+600V

+400V

+100V

-300V

-600VL. Wang, SLACBuild-up simulations: Electrodes with negative (above) or positive (below) potential

Slide30

Mitigations: Wiggler Chamber with Clearing Electrode

Thermal spray tungsten electrode and Alumina insulator

0.2mm thick layers

2

0mm wide electrode in wiggler

Antechamber full height is 20mm

Joe Conway – Cornell U.

Slide31

Mitigations: Dipole Chamber with Grooves

20 grooves (19 tips)

0.079in (2mm) deep with 0.003in tip radius

0.035in tip to tip spacing

Top and bottom of chamber

Joe Conway – Cornell U.

Slide32

Electron Cloud Mitigation Recommendation

Global Design Effort

32

Efficacy

Photoelectric yield (PEY)

Secondary emission yield (SEY)

Ability to keep the vertical

emittance growth below 10%CostDesign and manufacturing of mitigationMaintenance of mitigationEx: Replacement of clearing electrode PSOperationalEx: Time incurred for replacement of damaged clearing electrode PSRiskMitigation manufacturing challenges: Ex: ≤1mm or less in small aperture VC

Ex: C

learing

electrode

in

limited

space or in presence of BPM buttons

Technical uncertainty

Incomplete evidence of efficacy

Incomplete experimental studies

Reliability

Durability

of mitigation

Ex:

Damage of clearing electrode

feed-through

Impact on Machine Performance

Impact on vacuum performanceEx: NEG pumping can have a positive effectEx: Vacuum outgassingImpact on machine impedanceEx: Impedance of grooves and electrodesImpact on opticsEx: x-y coupling due to solenoidsOperationalEx: NEG re-activation after saturationNov 3-4, 2011 CLIC coll. meetingDedicated ILC DR Workshop at Cornell University, NY, USA 2010 on Recommendation on electron cloud mitigations

Slide33

Structured Evaluation of EC Mitigations

Nov 3-4, 2011 CLIC coll. meeting

Global Design Effort

33

Criteria for the evaluation of mitigations: Working Group rating

Efficacy of Mitigation

Costs

RisksImpact on MachineRating10144Normalized Weighting0.530.050.210.21

Slide34

C

ESR

TA

results and simulations suggest the possible presence of

sub-threshold emittance growth

Further investigation required

May require reduction in acceptable cloud density

a reduction in safety marginAggressive mitigation plan is required to obtain optimum performance from the 3.2km positron damping ring and to pursue the high current option ILC Working Group Baseline Mitigation RecommendationDrift*DipoleWigglerQuadrupole*Baseline Mitigation ITiN CoatingGrooves with TiN coatingClearing ElectrodesTiN CoatingBaseline Mitigation IISolenoid WindingsAntechamberAntechamberAlternate MitigationCarbon coating/ NEG CoatingTiN CoatingGrooves with TiN CoatingClearing Electrodes or Grooves*Drift and Quadrupole chambers in arc and wiggler regions will incorporate antechambers

Summary of Electron Cloud Mitigation Plan

Global Design Effort

34

Mitigation Evaluation conducted at ILC DR Working Group Workshop meeting

M. Pivi, S

.

Guiducci

, M. Palmer,

J

. Urakawa on behalf of the ILC DR

Electron Cloud Working

Group

Slide35

Completing evaluation for ILC

With recommended mitigations

the ring-average cloud density

is

4

×10

10 e/m3, well below the instability threshold of 2

×1011 e/m3.ILC Technical Design Report 2012: implemented technical mitigations allowed reducing the size of the ILC damping rings to 3.2km (17km in 2004).Mitigations adopted also at SuperKEKB and Daphne.

Slide36

e-

Cloud

@ DAFNE: Clearing

Electrodes

D.

Alesini

,

T. Demma et al. , in Proc. of IPAC 2010.Electric Field as computed by POISSON Clearing electrodes are installed in the vacuum chambers of wigglers

and

dipoles

of DAFNE positron ring

.

Electron

cloud

evolution

with clearing

electrodes

(POSINST)

0 50 100 150 200 250 300 350

Time (ns)

Theo

Demma

, INFN

Slide37

Clearing electrodes in DaF

ne

Clearing electrodes ON/OFF: horizontal tune shift of 0.0065.

D.

Alesini

, A. Drago, A. Gallo, S. Guiducci, C.

Milardi

, A. Stella, M. Zobov, S. De Santis, T. Demma, P. Raimondi, Phys Rev. Letters 110,124801 (2013)Simulated evolution of the cloud density for different electrode voltage.

Slide38

Horizontal

fractional tune as a function of bunch

number along train of bunches.

Electrodes OFF

Electrodes ON

Effect of clearing electrodes on beam in

Da

FneBeam dimension (um) at the Synchrotron light monitor turning all electrodes progressively off.Growth rate of horizontal instability for different clearing electrode voltages.D. Alesini, A. Drago, A. Gallo, S. Guiducci, C. Milardi, A. Stella, M. Zobov, S. De Santis, T. Demma, P. Raimondi, IPAC 2012

Slide39

Electrod cloud mitigations in Da

F

ne

: clearing electrodes

Clearing electrodes installed in

Da

F

ne dipoles and wigglersExperimental measurements have shown an impressive effectiveness of these devices in mitigating the e-cloud.Electrodes ON indicate an evident reduction of the electron cloud density.Electrodes allowed reducing the beam size, increase the instabilities growth rate, increase the beam current and luminosity.

Slide40

40

The interaction between the beam and the cloud is evaluated at 40 Interaction Points around the SuperB HER (LNF option) for different values of the electoron cloud density.

The threshold density is determined by the density at which the growth starts:

Beam energy E[GeV]

6.7

circumference L[m]

1200

bunch population N

b

5.7x10

10

bunch length

σ

z

[mm]

5

horizontal emittance

ε

x

[nm rad]

1.6

vertical emittance εy [pm rad]4hor./vert. betatron tune Qx/Qy40.57/17.59synchrotron tune Qz0.01hor./vert. av. beta function25/25momentum compaction 4.04e-4Input parameters (LNF conf.) for CMAD=5x1011=4x1011=3x1011T. Demma: Preliminary results for SuperBVertical emittance growth induced by e-cloud

Slide41

41

Snapshot of the electron (x,y) distribution

Density at center of the beam pipe is larger then the average value.

Buildup in Free Field Regions

Snapshot of the electron (x,y) distribution 50G solenoids on

Solenoids reduce to 0 the e-cloud density at center of beam pipe

Slide42

42

Snapshot of the electron (

x,y

) distribution

“

just before

”

the passage of the last bunch

LER Arc quadrupole vacuum chamber (CDR)

dB

y

/dx=2.5T/m,



=99%



max

=1.2

center

average

Buildup in the SuperB arcs: Quadrupoles

Cloud density average in chamber and near beam center

Slide43

43

Single Bunch Instability Threshold

June 2008

January 2009

March 2009



int

[10

15

m

-2

]

solenoids



int

[10

15

m

-2

] no solenoids



int

[1015m-2] solenoidsint [1015m-2] no solenoidsint [1015m-2] solenoidsint [1015m-2] no solenoidsSEY=1.195%0.062.10.092.50.222.799%0.020.250.040.30.04 0.7SEY=1.295%0.222.80.273.20.456.599%0.0450.710.060.820.072.4SEY=1.395%2.720.22.925.75.42599%0.943.21.34.14.513SuperB V12center 1012 [e-/m3]0.10.070.60.22.00.7th= 1012 [e-/m3]

Slide44

44

T.Demma

: Summary of electron cloud evaluation for

SuperB

Single bunch instability simulations for

SuperB

HER V12 taking into account the effect of solenoids have been performed using CMAD. They indicate a threshold density of ~10

12 e-/m3 (roughly 2 times previous estimates).Build-up simulations Indicate SEY<1.2, eta < 0.05 as safe region for the single-bunch instability..But:what is our confidence level in reaching these safe SEY values even including countermeasure such as antechambers, coatings, grooves, clearing electrodes…?Do we have reliable estimates (from measurements) of parameters such as PEY, photon reflectivity…?December 2011

Slide45

Multi-particle code simulations

Mauro Pivi

IBS in Super-B; Theory compared with C-MAD and IBS-Track codes.

CODE DEVELOPMENT

SUMMARY

Evaluation of electron cloud build-up and instability in LHC, Super-B, ILC

Evaluation of IBS in ultra-low emittance rings: CLIC, Super-B

Validation of mitigationsElectron cloud: emittance growth with cloud density in Super-BC-MADT. Demma and M. Pivi, Collective effects in Super-B, IPAC 2010Electron cloud: clearing electrodes in Super-B+100V+1000V0V