C L I CC L I C
Ultra-low emittance for the CLIC
damping rings using super-conducting
wigglersOctober 8th, 2007
Yannis PAPAPHILIPPOU
ANKA Seminar
C L I CC L I C
ANKA Seminar, Y. Papaphilippou
208/10/2007
Outline Overview of the CLIC Project
CLIC damping rings design Design goals and challenges Energy Lattice choice and optics optimisation Circumference Wiggler design and parameter scan Final emittances including Intra-beam
scattering Chromaticity correction and dynamic
aperture Low emittance tuning in the presence of
coupling Summary and open issues
M. Korostelev (PhD thesis, EPFL
2006)
C L I CC L I C
ANKA Seminar, Y. Papaphilippou
308/10/2007
The CLIC Project
Compact Linear Collider : multi-TeV electron-positron collider for high energy physics beyond today's particle accelerators
Center-of-mass energy from 0.5 to 3 TeV RF gradient and frequencies are very high
100 MV/m in room temperature accelerating structures at 12 GHz
Two-beam-acceleration concept High current “drive” beam, decelerated in
special power extraction structures (PETS) , generates RF power for main beam.
Challenges: Efficient generation of drive beam PETS generating the required power 12 GHz RF structures for the required
gradient Generation/preservation of small emittance
beam Focusing to nanometer beam size Precise alignment of the different
components
C L I CC L I C
4
Injector complex
DC gunUnpolarized e-
3 TeV
Base line configuration
LaserDC gunPolarized e-
Pre-injector Linac for e-
200 MeV
e-/e+ Target
Pre-injector Linac for e+
200 MeV
Primary beam Linac for e-
2 GeV
Inje
ctor
Lin
ac
2.2
GeV
e+ DR
e+ PDR
2.424 GeV360 m
Boo
ster
Lin
ac
6.6
GeV 3 GHz
e+ BC1 e- BC1
e+ BC2 e- BC2e+ Main Linac e- Main Linac
12 GHz, 100 MV/m, 21 km 12 GHz, 100 MV/m, 21 km
1.5 GHz
e- DR
e- PDR
1.5 GHz 1.5 GHz 1.5 GHz
3 GHz162 MV
3 GHz162 MV
12 GHz2.3 GV
12 GHz2.3 GV
9 GeV48 km maximum
30 m 30 m
10 m 10 m
360 m
150 m
15 m 15 m 150 m
2.424 GeV360 m
2.424 GeV 2.424 GeV
L. Rinolfi
C L I CC L I C
Damping ring design goals
Ultra-low emittance and high beam polarisation impossible to be produced by conventional particle source: Ring to damp the beam size to desired
values through synchrotron radiation Intra-beam scattering due to high
bunch current blows-up the beam Equilibrium “IBS dominated”
emittance should be reached fast to match collider high repetition rate
Other collective effects (e.g. e--cloud) may increase beam losses
Starting parameter dictated by design criteria of the collider (e.g. luminosity), injected beam characteristics or compatibility with the downstream system parameters (e.g. bunch compressors)
PARAMETER NLC CLIC
bunch population (109) 7.5 4.1
bunch spacing [ns] 1.4 0.5
number of bunches/train 192 316
number of trains 3 1
Repetition rate [Hz] 120 50
Extracted hor. normalized emittance [nm] 2370 <680
Extracted ver. normalized emittance [nm] <30 < 20
Extracted long. normalized emittance [eV m] 10890 <5000
Injected hor. normalized emittance [μm] 150 63
Injected ver. normalized emittance [μm] 150 1.5
Injected long. normalized emittance [keV m] 13.18 1240
C L I CC L I C
ANKA Seminar, Y. Papaphilippou
608/10/2007
Ring energy
Choice dictated by spin tune (half integer) for maintaining high-spin polarisation
Frozen on early design stage Advantage of lower
energies: For same equilibrium
emittance
i.e. smaller circumference and radiated power (cost), high momentum compaction (longitudinal stability).
Advantages of higher energy For fixed damping fraction due
to wigglers and wiggler peak field,
i.e. easier magnetic design (lower main field) and smaller total wiggler length
0 5 10 150
1
2
3
4
5
6
7
n0 5 10 15
0
1
2
3
4
5
6
7
n0 5 10 15
0
1
2
3
4
5
6
7
Ene
rgy
[GeV
]
n0 5 10 15
0
1
2
3
4
5
6
7
n
E = E0 (n + 1/2)/
NLC DR
CLIC DR
ILC DR
C L I CC L I C
708/10/2007
Lattice choice
Usually racetrack configuration with Theoretical Minimum Emittance (TME) arcs and damping wigglers in the straights
NLC DR was based on lattice with 32 TME arc cells and wigglers of 62m total length
ILC has a large ring of more than 6km for accepting large number of bunches with reduced e-cloud effect
TME and FODO lattice considered
(A. Wolski et al. 2003)
(A. Wolski et al. 2007)
C L I CC L I C
ANKA Seminar, Y. Papaphilippou
808/10/2007
CLIC damping ring layout
C L I CC L I C
TME arc cell
TME cell chosen for compactness and efficient emittance minimisation over Multiple Bend Structures (or achromats) used in light sources
Large phase advance necessary to achieve optimum equilibrium emittance
Very low dispersion Strong sextupoles
needed to correct chromaticity
Impact in dynamic aperture
C L I CC L I C
ANKA Seminar, Y. Papaphilippou
1008/10/2007
Phase advance choice
Optimum horizontal phase advance of cells for minimising zero current emittance is fixed (284o for TME cells)
Vertical phase advance is almost a free parameter
First iteration based on lattice considerations, i.e. comfortable beta functions and relaxed quadrupole strengths and chromaticity
Low horizontal phase advance gives increased momentum compaction factor (high dispersion) but also chromaticity
C L I CC L I C
ANKA Seminar, Y. Papaphilippou
1108/10/2007 11
Phase advance with IBS Horizontal phase advance for minimum horizontal emittance with IBS, is found in an area of small horizontal beta and moderate dispersion functions (between 1.2-1.3π, for CLIC damping rings)
Optimal vertical phase advance quite low (0.2π)
The lowest longitudinal emittance is achieved for high horizontal and low vertical phase advances
The optimal point has to be compromised due to chromaticity considerations and dynamic aperture optimisation
C L I CC L I C
ANKA Seminar, Y. Papaphilippou
1208/10/2007
Circumference
Usually chosen big enough to accommodate number of bunches
Drift space increase essential for establishing realistic lattice, reserving enough space for instrumentation and other equipment
For constant number of dipoles (TME cells), zero equilibrium emittance is independent of circumference
Normalised emittance with IBS increases with circumference (no wigglers) When dipole lengths increase with drifts,
emittance grows due to increase of damping time (inversely proportional to radiation integral I2 which decreases with length)
When only drifts increase, smaller emittance growth due to increase of optics functions
Impact on chromaticity + dynamic aperture Compensation may be achieved due
to increase of bunch length with circumference (momentum compaction)
Drifts + dipoles
Only Drifts
Only Drifts
Drifts + dipoles
C L I CC L I C
ANKA Seminar, Y. Papaphilippou
1308/10/2007
Damping wigglers
Damping wigglers are used to increase radiation damping and reduce the effect of IBS in order to reach target emittances
The total length of wigglers is chosen by its dependence with the peak wiggler field and relative damping factor
Damping factor increases for higher fields and longer wiggler occupied straight section
Relative momentum spread is independent of total length but increases with wiggler field
C L I CC L I C
ANKA Seminar, Y. Papaphilippou
1408/10/2007
Wigglers effect in emittance
For fixed value of wiggler period, equilibrium emittance minimum for particular value of wiggler field
By reducing total length, optimal values necessitate higher fields and lower wiggler periods
Optimum values change when IBS included, necessitating higher fields
Damping rings cannot reach 450nm with normal conducting wigglers
C L I CC L I C
ANKA SCwiggler
BINP SCwiggler
BINP PMwiggler
Wigglers’ effect with IBS For higher wiggler field and
smaller period the transverse emittance computed with IBS gets smaller
The longitudinal emittance has a different optimum but it can be controlled with the RF voltage The choice of the wiggler parameters is finally dictated by their technological feasibility
The choice of the wiggler parameters is finally dictated by their technological feasibility. Normal conducting wiggler of 1.7T
can be extrapolated by existing designs
Super-conducting options have to designed, built and tested
C L I CC L I C
ANKA Seminar, Y. Papaphilippou
1608/10/2007
Two wiggler prototypes 2.5Τ, 5cm period, built by BINP 2.7Τ, 2.1cm period, built by ANKA
Aperture reduced for the more challenging design
Current density can be increased by using different conductor type
Short version to be installed and tested at ANKA (energy of 2.5GeV)
Lifetime of 8-10h for lower gap, enough for the beam tests
Parameters BINP
ANKA
Bpeak [T] 2.5 2.7λW [mm] 50 21Beam aperture full height [mm] 12 5
Conductor type NbTi NbSn3
Operating temperature [K] 4.2 4.2
Wiggler prototypes
C L I CC L I C
ANKA Seminar, Y. Papaphilippou
1708/10/2007
RF voltage and frequency
The smallest transverse emittance is achieved for the lowest RF frequency and higher voltage, while keeping the longitudinal emittance below 5000 eV.m
Reversely the longitudinal emittance is increased for small RF frequency
C L I CC L I C
Average horizontal β function should be small enough for the wiggler period not to exceed the value producing efficient damping
FODO cell structure chosen with phase advances close to 90o giving average β’s of around 4m and reasonable chromaticity
Quad strength adjusted to cancel wiggler induced tune-shift 18
Wiggler FODO cell
C L I CC L I C
Non-linear dynamics
Two sextupole schemes considered Two families / 9 families of sextupoles
Dynamic aperture is 9σx in the horizontal and 14σy in the vertical plane (comfortable for injection)
Wiggler effect should be included and optimised during the design phase
C L I CC L I C
20
Coupling correction
Coupling effect of wigglers should be included in simulations
Correction with dispersion free steering (orbit and dispersion correction)
Skew quadrupole correctors for correcting dispersion in the arc and emittance minimisation
Iteration of dynamic aperture evaluation and optimisation after correction
In CLIC damping rings, the effect of vertical dispersion is dominant (0.1% of coupling and 0.25μm of dispersion invariant)
C L I CC L I C
ANKA Seminar, Y. Papaphilippou
2108/10/2007 21
Approximate scaling laws can be derived for a given damping ring design For example, for the CLIC damping rings, the horizontal
normalized emittance scales approximately as The above relationship is even more exact when the longitudinal
emittance is kept constant (around 5000 eV.m, in the case of the CLIC damping rings)
Vertical and longitudinal emittance are weakly dependent on bunch charge, and almost linear with each other
Bunch charge
C L I CC L I C
ANKA Seminar, Y. Papaphilippou
2208/10/2007 CLIC PWG Y. Papaphilippou 22
Damping rings’ parameters
2005: original ring
2006a: super-conducting wiggler considered
2006b: vertical dispersion included
2007a: 12GHz structure
2007b: reduced bunch population
2007c: CLIC_G structure
C L I CC L I C
ANKA Seminar, Y. Papaphilippou
2308/10/2007
Concluding remarks
Robust design of the CLIC damping rings, delivering target emittance with the help of super-conducting wigglersPrototype to be built and tested in ANKA
Areas needing further optimisationPre-damping ring optics designCollective effects including electron cloudRealistic cell length and magnet designSextupole optimisation and non-linear
dynamics including wiggler field errors