Challenges for Polarimetry at the ILC

Challenges for Polarimetry at the ILC

Moritz Beckmann, Anthony Hartin, Jenny List

DESY - FLC

EUCARD Workshop“Spin optimization at Lepton accelerators”

Mainz, GermanyFebruary 12, 2014

Outline

• Introduction• Polarimetry in the ILC beam delivery system• Spin transport

• Results• Beamline simulation• Collision effects

• Polarization measurement at the disrupted beam• Beamline design in view of the polarization measurement• Impact of the laser spot size at the downstream polarimeter

Moritz Beckmann DESY-FLC EUCARD Workshop Mainz 2014 2 / 26

ILC Beam Delivery System (BDS)


Laser Compton Polarimeter (Principle)

dipole magnetspolarized laser

scattered e±

e± beam

detector

• Laser: Compton scattering e± γ → e± γ• Scattering cross section depends on Pz

• Magnet chicane separates scattered e± from beam• Pz is determined from scattered electrons


Polarimetry at the ILC

150 m~1 650 m

upstream polarimeter IP

downstream polarimeter

e± beamlines

• Two Compton polarimeters per beam to measure Pz• Upstream polarimeter undisturbed by collision effects• Downstream polarimeter assesses collision effects• 0.25 % systematic uncertainty (goal)

• What do these measurements tell us about thelongitudinal polarization at the IP?

→ spin transport simulation

• Aim to understand spin transport to 0.1 %


Polarimetry at the ILC

150 m~1 650 m

upstream polarimeter IP

downstream polarimeter

e± beamlines

• Two Compton polarimeters per beam to measure Pz• Upstream polarimeter undisturbed by collision effects• Downstream polarimeter assesses collision effects• 0.25 % systematic uncertainty (goal)

• What do these measurements tell us about thelongitudinal polarization at the IP?→ spin transport simulation

• Aim to understand spin transport to 0.1 %


Spin Precession

• Spin precession in electromagnetic fields:T-BMT equation

• For ~B⊥ only:

ϑspin = b(E) · ϑorbit

b(E ) = aγ + 1 = g−22 ·

Em + 1

≈ 568 for 250 GeV-electrons

• Dipole magnets, no beam energy spread:spin vectors precess uniformly, |~P| conserved

B


Spin Fan-Out in Quadrupole Magnets

PP'P

quadrupole magnets

For illustration purposes, the second quadrupole is stronger. Two-dimensional

betatron oscillations are not taken into account here.

• Different precession angles after first quadrupole⇒ polarization |~P| decreases

• |~P| recovered by second quadrupoleMoritz Beckmann DESY-FLC EUCARD Workshop Mainz 2014 7 / 26

Beam-Beam Collision Effects

e e Pairs+ _

BeamstrahlungA. Vogel

Bunches focus each other by their electromagnetic fields:

• Spin fan-out (like in quadrupole magnets)

• Spin flip by emission of beamstrahlung(Sokolov-Ternov effect)


Spin Transport Simulation Framework

UP IP DP

beam-beam collisionGuinea-Pig++

data analysis

polarimeter measurement

polarimeter measurement

beamline layout

particle / spin transport along the BDS Bmad

beam parameters

• Developed a beamline simulation (based on Bmad)

• Simulate 40 000 (macro)particles per bunch, generated frombeam parameters at the beginning of the BDS

• Interfaced directly to the simulation of the collisions(Guinea-Pig++)


Results

•√s = 500 GeV

• Beam parameters according to Reference Design Report(RDR, 2007)

• Collision effects also for beam parameters according toTechnical Design Report (TDR, 2013)


Spin Transport in the BDS: Basic Configuration

distance s along BDS [m]0 500 1000 1500 2000 2500 3000 3500

po

lari

zati

on

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0.8

UP IP DP

-ezP|P|

UP/DP: up-/downstream polarimeter


Spin Transport in the BDS: Basic Configuration

distance s along BDS [m]3400 3450 3500 3550

po

lari

zati

on

0.799

0.7995

0.8

]-3

rela

tive

ch

ang

e [1

0

-1

-0.5

0

IP DP

-e

zP|P|

extractionline

quadrupoles

dipole chicanes

DP: downstream-polarimeter

• Quadrupoles cause spin fan-out

• Changes in Pz well below 0.1 % without collisions


Factors affecting the spin transport (without collisions)

contribution uncertainty[10−3]

Beam and polarization alignment 0.72(∆ϑbunch = 50µrad, ∆ϑpol = 25 mrad)Random misalignments (10µm) 0.43Variation in beam parameters (few %) 0.03Bunch rotation (crab cavities) < 0.01Detector solenoid 0.01Synchrotron radiation 0.005

Total (quadratic sum) 0.85

Now: e+e− beam collisions


Spin Transport after Collision


zlo

ng

. po

lari

zati

on

P

0.76

0.77

0.78

0.79

0.8

]-3

rela

tive

ch

ang

e [1

0

-40

-20

0

IP DP

-e

no collisionlumi-weightedafter collisionmeasurable



• Luminosity-weighted (•): Pz of the colliding particles

• Larger angular divergence / energy spread after collision

• Large spin fan-out in extraction line quadrupoles




zlo

ng

. po

lari

zati

on

P

0.76

0.77

0.78

0.79

0.8

]-3

rela

tive

ch

ang

e [1

0

-40

-20

0

IP DP

-e




• Extraction line design: restore luminosity-weighted Pz(•) at the downstream polarimeter

• Employ spin fan-out: focus beam at downstreampolarimeter with half divergence angle w. r. t. the IP




zlo

ng

. po

lari

zati

on

P

0.76

0.77

0.78

0.79

0.8

]-3

rela

tive

ch

ang

e [1

0

-40

-20

0

IP DP

-e




θx θy ⇒ ∆Pz ∝ θx2

∆P lumz ≈ 1

4∆Pz ∝(θx

2

)2

Idea: |R22(IP→ DP)| = 0.5 ⇒ P lumz = PDP

z

Further reading: SLAC-PUB-4692, SLAC-PUB-8397


Laser and Particle Bunch at the Downstream Polarimeter

after collisionwithoutcollision

laser spote± beam

~mm ~cm0.04mm

~cm

• Without collision: entire beam exposed to laser

• After collision: center of beam exposed to lasersample of scattered electrons representative?


Downstream Measurement

• Downstream polarimeter located in magnet chicane⇒ particle position correlated with energy (dispersion)

-0.4 -0.3 -0.2 -0.1 0-0.02

-0.015

-0.01

-0.005

0

0.005

0.01

0.015

1

10

210

310

410

510

610

particle energy [GeV]160 180 200 220 240ve

rt. p

arti

cle

po

siti

on

y [

mm

]

-20

-10

0

10 Laser spot size

• Laser spot size at Compton-IP only ∼ 0.1 - 1 mm



• Beamstrahlung correlates energy and Pz⇒ Pz correlated with particle position⇒ Selective measurement, measurement bias

-0.015 -0.01 -0.005 0

zlo

ng

. po

lari

zati

on

P

0.75

0.76

0.77

0.78

0.79

0.8

vert. particle position y [mm]-15 -10 -5 0

laser spot size

• Measurable longitudinal polarization := average Pz ofparticles within a given (laser spot) radius


Spin Transport for Different Beam Parameters

0 20 40 60

zlo

ng

. po

lari

sati

on

P

0.79

0.795

0.8 ]-3

rela

tive

ch

ang

e [1

0

-15

-10

-5

0

no energy spread/lossRDR TDR TDR* RDR TDR TDR*

before collision lumi-weighted after collision DP DP measurable (r=0.1/0.2/0.5/1.0 mm)

DP: downstream polarimeter

• No energy spread/loss: no discrepancy between measurement() and average Pz () at downstream polarimeter

• RDR → TDR: stronger focussing ⇒ higher collision intensity⇒ larger spin fan-out in collision and afterwards



0 20 40 60

zlo

ng

. po

lari

sati

on

P

0.79

0.795

0.8 ]-3

rela

tive

ch

ang

e [1

0

-15

-10

-5

0




• No energy spread/loss: no discrepancy between measurement() and average Pz () at downstream polarimeter

• RDR → TDR: stronger focussing ⇒ higher collision intensity⇒ larger spin fan-out in collision and afterwards



0 20 40 60

zlo

ng

. po

lari

sati

on

P

0.79

0.795

0.8 ]-3

rela

tive

ch

ang

e [1

0

-15

-10

-5

0




Extraction line design: restore P lumz (•) at downstream pol. ()

• Design (|R22| = 0.5) assumes Dx 1DRDRx = 0.17 DTDR

x = 0.3

• More beamstrahlung (not accounted for by design)


Spin Transport for Different Spin Configurations

DP

TDR*

R22=-0.5

P'

P

P

P'

IP

TDRafter collision

T-BMT

T-BMT

assumingθ

spin= (aγ+1)·θ

orbit

at the beginning

assumingθ

spin= (aγ+1)·θ

orbit

at the beginning

For illustration only. All angles exaggerated. Beamstrahlung effects neglected.


Spin Transport for Different Spin Configurations

0 20 40 60

zlo

ng

. po

lari

sati

on

P

0.79

0.795

0.8 ]-3

rela

tive

ch

ang

e [1

0

-15

-10

-5

0




TDR* with respect to TDR:

• All spin vectors parallel before collision, bunch focussed(45µrad divergence angle)

• Mostly same behaviour in collision (N, •, H), but differentvalue at downstream polarimeter ()



0 20 40 60

zlo

ng

. po

lari

sati

on

P

0.79

0.795

0.8 ]-3

rela

tive

ch

ang

e [1

0

-15

-10

-5

0




• Polarization varies by several % along the extraction line

• Discrepancies between P lumz and Pz at the downstream pol.

(•, , ) in the range 0.1− 0.4 %;discrepancies cancel partially, but only coincidentally


Conclusions

• Cross-calibration (without collisions) to precision of < 0.1 %

• Polarization vector alone not sufficient anymore to describespin configuration of beam:• Spin fan-out becomes relevant due to higher measurement

precision, higher energy and more intensive collisions• How well do we know the initial spin configuration?→ “cradle-to-grave” simulation

• Extraction line design (restore P lumz at downstream pol.):

• Works as foreseen for low-intensity collisions X• TDR beam parameters: higher intensity → larger discrepancies• Beamstrahlung not taken into account; Dx no longer 1• Disrupted beam lets knowledge of the laser spot size/position

at the downstream polarimeter become crucial for themeasurement precision

• Larger laser-spot? Drawbacks: required laser power,low-energy tail undesired in polarimeter


Conclusions

• Cross-calibration (without collisions) to precision of < 0.1 %

• Polarization vector alone not sufficient anymore to describespin configuration of beam:• Spin fan-out becomes relevant due to higher measurement

precision, higher energy and more intensive collisions• How well do we know the initial spin configuration?→ “cradle-to-grave” simulation

• Extraction line design (restore P lumz at downstream pol.):

• Works as foreseen for low-intensity collisions X• TDR beam parameters: higher intensity → larger discrepancies• Beamstrahlung not taken into account; Dx no longer 1• Disrupted beam lets knowledge of the laser spot size/position

at the downstream polarimeter become crucial for themeasurement precision

• Larger laser-spot? Drawbacks: required laser power,low-energy tail undesired in polarimeter


Thanks for your attention!

Further reading:

• DESY-THESIS-13-053http://www-library.desy.de/preparch/desy/thesis/desy-thesis-13-053.pdf

• Publication in preparation


http://www-library.desy.de/preparch/desy/thesis/desy-thesis-13-053.pdf

Backup Slides


Differences RDR - TDR

Parameter symbol RDR TDR

Bunches per train 2 625 1 312Horizontal bunch size σx [nm] 639 474Vertical bunch size σy [nm] 5.7 5.9Beam energy spread (e−/e+) σE/E [10−3] 1.4/1.0 1.24/0.7

e+e− luminosity L [1038 m−2 s−1] 2 1.47incl. waist shift 1.8

Beamstrahlung parameter Υglobal 0.048 0.062


Thomas-Bargmann-Michel-Telegdi (T-BMT) Equation

d

dt~S =

(~ΩB + ~ΩE

)× ~S

~ΩB =− q

mγ

((1 + aγ) ~B − a ~p · ~B

(γ + 1)m2c2~p

)

=− q

mγ

((1 + aγ) ~B⊥ + (1 + a) ~B‖

)

~ΩE =q

mγ· 1

mc2

(a +

1

1 + γ

)~p × ~E


Compton Scattering

0 50 100 150 200 2500.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

σC

om

pto

n[m

barn

/ G

eV

]

Compton−scattered Electron Energy [GeV]

λPe = +1(same)

λPe = −1

(opposite)

0 100 200 300 400 5000

5

10

15

20

25

30

Com

pton

End

poin

t Ene

rgy

[GeV

]Beam Energy [GeV]


Polarimeter Chicane (upstream)

• Constant magnetic field• Dispersion (depending on beam energy): 1-11 cm• Scattering for every bunch per bunch train• Energy spectrum is polarization-dependent• Energy distribution → spatial distribution• Cherenkov gas detector counts electrons per channel

+e /e

+e /e IP

16.1m

8 m

16.1m

CherenkovDetector

125 GeV

25 GeV

50 GeV

Magnetic Chicane

250 GeV

24 c

m

45.6 GeV

inout

LaserIP

8.1m

Dipole 2 Dipole 3

8.1m

Dipole 4Dipole 1

P11P10 P12P1 P2 P3

P5P4 P6 P7 P8 P9

total length: 74.6 m


Polarimeter Chicane (downstream)

• Constant magnetic field• Dispersion (depending on beam energy): 1-11 cm• Scattering for 3 bunches per bunch train• Energy spectrum is polarization-dependent• Energy distribution → spatial distribution• Cherenkov gas detector counts electrons per channel


Polarimeter Detector

e

PMLED (calibration)

Cherenkov photons

2

10 c

m

10 x 10 mmcross section:

15 cm

Al−

tube

s

beam

aluminum tubes

xy

z

photodetectors

LEDs


Compton Polarimeters: Systematic Errors

Goal: relative systematic error on measurement < 0.25 %(SLD polarimeter: 0.5 %)

• Detector linearity: contribution of ∼ 0.1− 0.2 % (goal)

• Laser polarization: ∼ 0.1 % X• Analyzing power: ∼ 0.1 % (UP: X, DP: ?)

• Detector alignment: can be determined from data (X)0.5 mm precision sufficient

• Alignment of magnets negligible compared to detector XField inhomogeneities? to be investigated

• Disrupted electron beam at downstream polarimeter:• Dependence on laser-spot size and position: ??• Beam energy spread no concern for small laser-spot sizes

thanks to dispersion X


Polarization Measurement at the IP

-1Luminosity fb0 200 400 600 800 1000 1200

% e

rror

pol

ariz

atio

n

0

0.2

0.4

0.6

0.8

1 pol- pol 80% e+30% e

positron Blondel

electron Blondelpositron Fitelectron Fit

-1Luminosity fb0 200 400 600 800 1000 1200

% e

rror

pol

ariz

atio

n

0

0.2

0.4

0.6

0.8

1 pol- pol 80% e+60% e

positron Blondel

electron Blondelpositron Fitelectron Fit

I. Marchesini

Blondel scheme:

|P lumiz (e±)| =

√(σ−+ + σ+− − σ−− − σ++)(±σ−+ ∓ σ+− + σ−− − σ++)

(σ−+ + σ+− + σ−− + σ++)(±σ−+ ∓ σ+− − σ−− + σ++)



I. Marchesini



I. Marchesini


Quadrupole Magnet

S

NS

N

S

N S

N

Black arrows: magnetic field linesBlue arrows: forces on an incoming electron beam


Bunch Rotation at the IP

• Collision under crossing angle of 14 mrad

• Maximize luminosity: rotate bunches using crab cavities

• Time-dependent transverse deflection of particles


Downstream Pol.: Dispersion w/o Collision

-0.008 -0.006 -0.004 -0.002 0 0.002 0.004

-0.25

-0.2

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

0.2-310×

1

10

210

310

410

510

particle energy [GeV]248 248.5 249 249.5 250 250.5 251ve

rt. p

arti

cle

po

siti

on

y [

mm

]

-0.2

-0.1

0

0.1

0.2


Downstream Pol.: Dispersion after Collision

-0.4 -0.35 -0.3 -0.25 -0.2 -0.15 -0.1 -0.05 0-0.02

-0.015

-0.01

-0.005

0

0.005

0.01

1

10

210

310

410

510

610

particle energy [GeV]160 180 200 220 240ve

rt. p

arti

cle

po

siti

on

y [

mm

]

-20

-10

0

10

-0.05 -0.04 -0.03 -0.02 -0.01 0

-0.4

-0.2

0

0.2

0.4

-310×

1

10

210

310

410

510

particle energy [GeV]238 240 242 244 246 248 250ve

rt. p

arti

cle

po

siti

on

y [

mm

]

-0.5

0

0.5



Longitudinal polarization vs. energy at the downstreampolarimeter, after collision

-0.3 -0.2 -0.1 0

lon

g. p

ola

riza

tio

n

0.66

0.68

0.7

0.72

0.74

0.76

0.78

0.8

particle energy [GeV]160 170 180 190 200 210 220 230 240 250

(E)zP

zaverage P


0 50

zlo

ng

. po

lari

zati

on

P

0.79

0.795

0.8 ]-3

rela

tive

ch

ang

e [1

0

-15

-10

-5

0

UP before collision lumi-weighted after collision DP DP measurable (r=0.1/0.2/0.5/1.0 mm)

-e

no CCnoCCL no BS,SR

no SRref. WS WSL TDRϒ

0 20 40 60 80

|P

enti

re p

ola

riza

tio

n |

0.79

0.795

0.8 ]-3

rela

tive

ch

ang

e [1

0

-15

-10

-5

0

-e

no CCnoCCL no BS,SR



0 50

zlo

ng

. po

lari

zati

on

P

0.296

0.297

0.298

0.299

0.3 ]-3

rela

tive

ch

ang

e [1

0

-15

-10

-5

0

UP before collision lumi-weighted after collision DP DP measurable (r=0.1/0.2/0.5/1.0 mm)

+e

no CCnoCCL no BS,SR


0 20 40 60 80

|P

enti

re p

ola

riza

tio

n |

0.296

0.297

0.298

0.299

0.3 ]-3

rela

tive

ch

ang

e [1

0

-15

-10

-5

0

+e

no CCnoCCL no BS,SR



Why Polarization?

Electroweak processes: cross sections depend on Pze. g. W+W− pair production

I. Marchesini

Polarized beams

• provide new observables

• can be used to enhance/suppress processes


The International Linear Collider (ILC)

• e+e− collider as complement to LHC

•√s ≤ 500 GeV, upgradable to 1 TeV

• Longitudinally polarized beams: |Pz(e−)| = 80 %

Longitudinally polarized beams:

|Pz(e+)| = 30 to 60 %


Documents

Challenges for Polarimetry at the ILC