The SuperB Accelerator

M. Biagini for the SuperB Accelerator Team

Epiphany 2012 ConferenceKrakow, January 9-11, 2012

SuperB AcceleratorSuperB is a 2 rings, asymmetric energies (e- @ 4.18, e+ @ 6.7 GeV) collider with: large Piwinski angle and “crab waist” (LPA & CW) collision scheme ultra low emittance lattices longitudinally polarized electron beam target luminosity of 1036 cm-2 s-1 at the (4S) possibility to run at /charm threshold with L = 1035 cm-2 s-1

Criterias used for the design: Minimize building costs Minimize running costs Minimize wall-plug power and water consumption Reuse of some PEP-II B-Factory hardware (magnets, RF)

SuperB can be also a good “light source”: there will be some Sinchrotron Radiation beamlines (collaboration with Italian Institute of Technology)

2

3

World e+e- colliders luminosity

B-Factories-FactoriesFuture Colliders

1027

1029

1031

1033

1035

0.1 1 10 100 1000

Lu

min

os

ity

(cm

-2 s

-1)

c.m. Energy (GeV)

ADONE

DCI

ADONE

VEPP-2M

VEPP2000

DANE

BEPC

SPEAR2

VEPP-4M PETRAPETRA

PEPDORIS2

BEPCII CESR

PEP-II

KEKB

LEP

LEP

LEP

LEP

ILC

CLIC

SUPERKEKB

SuperB

BINP c-

CESR -c

SuperFactories

Factories

Linear collidersSuperB

SuperB: highest world luminosity collider ever

Ultra-low emittance

Very small at IP

Large crossing angle

“Crab Waist” transformation

Small collision area

Lowerispossible

NO parasitic crossings

NO x-y-betatron resonances

Principle: beams more focused at IP + “large” crossing angle (LPA) + 2 sextupoles/ring to “twist” the beam waist at the IP (CW)

Large Piwinski Angle & Crab Waist:a novel idea for Luminosity increase

Proved to work at upgraded DANE -Factory

2008-2009

P.Raimondi, 2° SuperB Workshop, March 2006

P.Raimondi, D.Shatilov, M.Zobov, physics/0702033

4

NEW COLLISION SCHEME

DANE Peak Luminosity

Des

ign

Go

al

5

6

SuperB main features

Goal: maximize luminosity while keeping wall power low

2 rings (~4 GeV and ~7 GeV) with flexible designUltra low emittance optics: 7x4 pm vertical emittanceBeam currents: comparable to present Factories LPA & CW scheme used to maximize luminosity and minimize beam size blow-upNo “emittance” wigglers used (save power)Design based on recycling PEP-II hardware (save costs)Longitudinal polarization for electrons in the LER (unique feature)Possibility to push the cm energy down to the -charm threshold with a luminosity of 1035 cm-2 s-1

Parameter Table

Baseline + other 2 options:•Lower y-emittance•Higher currents (twice bunches)

Tau/charmthreshold runningat 1035

Baseline: •Higher emittance due to IBS•Asymmetric beam currents

RF power includes SR and HOM

7

SuperB layout

Site chosen @ Tor Vergata University (Rome II) campus

Sinchrotron Light (SL) beamlines are becoming part of the layout (HER preferred at the moment)

One tunnel will host both rings, which will probably have a tilt one respect to the other, to allow for easier crossing and SL beamlines from both HER and LER (if needed)

The position of the Linac complex has still to be finalized, depending on the injection requirements

The rings layout has been recently improved to accomodate Insertion Devices (ID) needed for SL users

8

SiteSite

LNFLNF

About 5 Km

Tor Vergata University campusTor Vergata University campus

9

Ground measurementsGround motion measurements performed on site in April show very «solid» grounds in spite of the vicinity of the highway, just 100 m away

The highway is at higher level with respect to the site, and the traffic vibrations («cultural noise») are very well damped

Ground x-sectionVolcanic soil

10

SuperB @ Tor Vergata

SR Beamlines

Injection complex11

SR Beamlines

Tomassini

12

Rings Lattice

The two rings have similar geometry and layout, except for the length of dipolesThe arcs cells have a design similar to that of Synchrotron Light Sources and Damping Rings in order to achieve the very low emittancesIn the latest version of the lattice some cells for Insertion Devices have been insertedRings are separated about 2m in horizontal and 1m in vertical

Spin Rotator

FF

3 ID cells 3 ID cells

Injection section

RF section

IP•Circumference 1195 m

•Horizontal separation of arc ~2 m

•Vertical separation of RF section 0.9 m

•Dimension sizes of rings 416 m x 342 m

Siniatkyn

Collider Layout

14

αtilt=2.6 mrad

Vertical separation 0.9 m

Vertical rings separation

Rings tilt at IP (by small solenoids not vertical bends) provides ~1 m vertical separation at the opposite point: (a) e+e- beams separation, (b) SR beamlines from both rings, (c) better equipment adjustment

Polarization in SuperB 90°spin rotation about x axis 90°about z followed by 90°about y

“flat” geometry no vertical emittance growth

Solenoid scales with energy LER more economical

Solenoids are split & decoupling optics added

The SR optics design has been matched to the Arcs and a similar (void) insertion added to HER

This design poses severe constraints on the FF bending angles of LER and HER in order to achieve the “right” spin dynamics

A polarimeter has been designed to measure polarization

IP HER

HER LER

LER

S.R. solenoids (90° spin)

S.r. dipoles(270° spin)

15

Polarization resonances

ELER

Beam polarization resonances do constraint the beam Energy choice

Plot shows the resonances in the energy range of LER

Beam polarization computed assuming 90% beam polarization at injection 3.5 minutes of beam lifetime (bb limited)

From this plot is clear that the best energy for LER should be 4.18 GeV HER must be 6.7 GeV

16

Interaction RegionThe Interaction Region must satisfy both machine and detector requirements: Final Focus elements as close as possible to the IP Small detector beam pipe Enough beam stay clear small emittance helps Control Synchrotron Radiation backgrounds Have an adequate detector solid angle Magnet vibrations need to be damped (at the level of 10nm) A state-of-the-art luminosity feedback is needed

With the large crossing angle the beam is off-axis in the first quadrupoles, hence it is not only focused but also bent, producing unwanted SR backgrounds and emittance growth

For SuperB a new design of the first doublet with «twins» quadrupoles was developed

17

Final Focus sections• “Spin rotator” optics is replaced with a simpler matching section

IP

Y-sext

X-sextMatchCrab

HER

• Matching section is shorter than HER to provide space for spin rotator optics.• ±33 mrad bending asymmetry with respect to IP causes a slight spin mismatch between SR and IP resulting in ~5% polarization reduction.

IP

Y-sext

X-sextMatch & Spin RotatorCrab

LER

* = 26 / 0.25 mm

* = 32 / 0.21 mm

18

QD0 Design: 2 possible choices

-3 -2 -1 0 1 2 3

0

100

-100

-200

200

mm

m M. Sullivan March, 13, 2010 SB_RL_V12_SF8A_3M

QF1 QF1

QD0

PM

SolenoidsHER QF1

HER QD0

PEP-II Support tube

-3 -2 -1 0 1 2 3

0

100

-100

-200

200

mm

m M. Sullivan March, 13, 2010 SB_RL_V12_SF8A_3M

QF1 QF1

QD0

PM

SolenoidsHER QF1

HER QD0

Vanadium Permendur “Russian” Design

Air core “Italian” QD0, QF1 Design

19

Air-core QD0 is a SC iron free septum double quad

20

Field generated by 2 double helix windingsin a grooved Al support

Construction of a model coil for addressing quench issues

The coil has been constructed at ASG Superconductors and now is at INFN Genova for testing at 4.2K. The results of this test are crucial for the design.

Test this month

Collettive effects

Stored beams are subject to effects that can produce instabilities or degrade the beam quality, such as: Intra-Beam-Scattering (IBS) inside the bunch produces

emittance and energy spread growth (not important in Damping Ring)

Electron-cloud instability limits the current threshold of the positron beam needs mitigation methods (ex. solenoids, beam pipe coating, clearing electrodes...)

Fast Ions Instability is critical for the electron beam CSR (Coherent Synchrotron Radiation) degrades beam

quality (not important in Damping Ring)

These effects have been studied and remediation techniques chosen 22

23

Snapshot of the electron (x,y) distribution

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

E-cloud build-up 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

24

e-cloud buildup in HER Dipoles

By=0.3 T; =95% SEY=1.1

th= 1012 [e-/m3]

e- density at center of the beam pipe

e- density averaged over the beam chamber

10 X beam sizes

Snapshot of the electron (x,y) distribution “just before” the passage of the last bunch

Demma

e-cloud clearing electrodes in DANE

Very positive results: vertical beam dimension, tune shift and growth ratesclearly indicate the good behaviour of these devices,which are complementary to solenoidal windings in field free regions

Drago

Beam loss above this current if no feedbacks

25

Low emittance tuningThe extremely low design beam emittance needs to be tuned and minimized careful correction of the magnet alignment and field errors

These errors produce emittance coupling with transfer of some horizontal emittance to the vertical plane this needs to be minimized

Beta-beating (ring -functions are not as in the model machine, but are perturbed by the magnet errors) also needs minimization

Vertical dispersion at IP needs to be corrected to the lowest possible value not to compromise luminosity

26

LET ToolThis tool has been successfully tested at Diamond (RAL) and SLS (PSI) synchrotron light sources, which have similar emittances as SuperB.

This work allows to set tolerances on magnet alignment and once the machine is running is able to detect such errors for correction

27

First tolerance tests for HER V16misalignments ARCS FF

QUAD SEXT DX,DY 50 μm 30 μm

QUAD SEXT DPHI 100 μrad 50 μrad

Monitor resolution 1 μm 1 μm

Monitors OFFSETs 50 μm 50 μm

DIPOLE DPHI and DTHETA

50 μrad 50 μrad

50 random sets, correcting with LET for 2 iterations after 3 orbit pre-correction iterations

beforebefore4.4 pmrad4.4 pmrad

Injection System

Injection in top-up requires a very stable, reliable injection complexLatest design:

Only e+ beam is stored in Damping Ring (DR) while e- beam is directly accelerated and injected e+ stored in DR for the time between two injection pulses, achieving same emittance damping factor at twice the repetition frequency possible with a 100 Hz Linac to inject at 50 Hz in each ring using a single bunch per pulse to make the current per bunch very uniform along the bunch trains

Proposal: use SLAC gun (high charge, 10 nC) for the e+ line and have a custom made polarized, low charge, low emittance gun for the e- line

R&D in progress at LAL/Orsay for the positron sourceContacts with SPARC group in Frascati for the low emittance gunR&D at SPARC on C-band Linac maybe useful also (shorter)

Boni, Guiducci,Preger, Variola et al) 29

THERMIONICGUNSHB0.6 GeVPC

BUNCHCOMPRESSOR

5.7 GeV e+ 3.9 GeV e-

POLARIZED SLAC GUN

SHB50 MeV

CAPTURESECTION

e+

e-combinerDC dipole

0.2 GeV

CAPTURESECTION

positron linac

1 GeV e+

≈ 350 m

L-band

Positron Source

Polarized Electron Source

Main Linac

e+ Damping Ring

Transfer linesPositron Source to Damping RingDamping Ring to Main LinacElectron Source to Main LinacMain Linac to HERMain Linac to LER

Transfer linesPositron Source to Damping RingDamping Ring to Main LinacElectron Source to Main LinacMain Linac to HERMain Linac to LER

Injection system layout

Details: “SuperB Progress Reports – Accelerator”, (Dec. 2010) – Chapter 15 http://arxiv.org/abs/1009.6178v3. “Updated Design of the Italian SuperB Factory Injection System”, IPAC’11

Injection Complex

Present status Parameters and site layout selected Layout and parameters of the system components defined Beam dynamics evaluation started

Remaining work: Baseline decision on electron source: direct injection or damping

ring Baseline decision on positron source: conversion at low energy

(.6 GeV), L-band linac for capture and acceleration up to 1 GeV (or a combination of S and L band)

Transfer lines layout and composition follows

Systems ready for TDR Damping ring Main linac

31

Injection tracking with bb

No beam-beam Crab = 1 Crab = 0.5 Crab = 0

Average over (1 ÷ 100) turns

No beam-beam Crab = 1 Crab = 0.5 Crab = 0

Average over (4001 ÷ 4100) turns

32

No beam-beam Crab = 1 Crab = 0.5 Crab = 0

Average over (30001 ÷ 30100) turns

SuperKEKBSuperKEKB

33

FeedbacksDrago

34

R&D on Controls (!CHAOS) This activity attracted interest from several other INFN structures and Universities

Bisegni

35

SuperB not «just a collider»

36

Brilliance SuperB vs ESRF (and ESRF upgrade)Brilliance SuperB vs ESRF (and ESRF upgrade)

Used U23 of ESRF ID27

ESRF parameters (4nm) 200 mA0.7% coupling2 m undulator

ESRF upgrade (4nm) 300 mA0.3% coupling4 m undulator

SuperB (2nm)500 mA0.7% coupling2m undulator

SuperBESRF upgradeESRF

Bartolini

37

Compatibility with collider operation (current, orbit stability, heat load management,… ) needs to be studied

MC simulations of different targets for different particles production

Quintieri

38

Summary on Accelerator work

Lattice «close» to be frozen, some more work needed on beam dynamics issues

We do have some systems «close» to TDR phase

Some strategical design choices still to be taken (ex. in injection system)

Most important issues to solve in the next months have been identified

R&D on control system started

R&D on new bunch-by-bunch feedback very positive (test at DANE)

Tests on e-cloud suppression electrodes at DANE successfull

39

Organization

The Cabibbo Laboratory, in charge of building and operating the SuperB Accelerator and Detector, has been founded on October 7th 2011 as a Consortium between INFN and University of Tor Vergata

Systems (almost) ready for technical design: Magnets, vacuum chamber, support structure of the main rings Beam diagnostic and control Power supplies Damping ring Linear accelerator Polarized electron source and positron source

40

List of present partnersCountry Partner Location Short name Contact

personWP

Italy Istituto Nazionale Fisica Nucleare 

Frascati INFN-LNF M. Biagini 1,2,3,4,5,6,7,8,9,10,14,16,17,18,19,20

Tor Vergata INFN-ROMA2 L. Catani 7,10,15

Pisa + Genova + Napoli

INFN-PI, GE, NA

E. Paoloni 8

Padova INFN-PD M. Bellato 10

Bari INFN-BA G. Iaselli TBD

Italy Italian Institute of Technology

Genova IIT E. Di Fabrizio 11

USA DOE – Stanford Linear Accelerator

Stanford SLAC J. Seeman 20

France Centre National de la Recherche Scientifique

Orsay IN2P3-LAL A. Variola 1,2,3,4,5,6

Grenoble IN2P3-LPSC M. Baylac 15

Annecy IN2P3-LAPP A. Jeremie 7,8

UK John Adams Institute

Oxford JAI A. Seryi 7,8

DIAMOND Rutherford DIAMOND R. Bartolini 6,11,13

Russia Budker Institute Novosibirsk BINP E. Levichev 5,6,15,18

Poland Institute of Nuclear Physics PAS

Krakow  PAS T. Lesiak TBD

Conclusions IOrganization of the accelerator structure is progressing

We have a draft organization of the accelerator work structure in Work Packages, with milestones and deliverables, needs refinement

We plan to commission to other laboratories/Institutions parts of the accelerator, taking into account their expertise in the field

Some examples at present:France for the positron source, FF vibration control, ground

measurements,… England for the Final Focus, IP feedback, SL beamlines,…BINP for DR, special magnets, vacuum pipe,…SLAC for PEP-II components (RF, magnets,…)Poland contribution to be discussed 42

Conclusions IIAn MOU with SLAC for procurement of PEP-II equipment is being prepared

Synchrotron Light Italian community started to consider SuperB properties for SL users needs more thoughts on maximum current, operation mode, experiments

MC simulations on the possibility to have a «SuperBeamTestFacility» started interest from users

With the Cabibbo Laboratory now in place we will be ready soon to hire personnel and reinforce the collaboration in order to finish TDR and start digging the tunnel before the end of this year

Thank you for your attention !43