Pawel Nadel-TuronskiPawel Nadel-Turonski
Jefferson LabJefferson Lab
55thth Int'l Workshop on High-Energy Physics in the LHC Era Int'l Workshop on High-Energy Physics in the LHC Era
UTFSM, Valparaiso, Chile, December 16–20, 2013UTFSM, Valparaiso, Chile, December 16–20, 2013
The Electron-Ion Collider at JLab
for the MEIC groupfor the MEIC group
2
Outline
The Electron-Ion ColliderThe Electron-Ion Collider
Some physics highlightsSome physics highlights
Accelerator and detectorsAccelerator and detectors
3
EIC@JLab – physics program
• The EIC at JLab supports the full physics program for a generic EIC– INT report, white paper
• The JLab implementation of the EIC offers some unique accelerator and detector capabilities (focus of this talk)
4
EIC – consensus on global EIC requirements
• Stage I energy: √s = 20 – 70 GeV (variable)
• Stage II energy: √s up to about 150 GeV
• Polarized electron, nucleon, and light ion beams– Electron and nucleon polarization > 70%
– Transverse polarization at least for nucleons
• Ions from hydrogen to A > 200From base EIC requirements in the INT report
The EIC project is pursued jointly by BNL and JLab
• Luminosity reaching 1034 cm-2s-1
(EIC)(EIC)
((MMEEIICC))
5
EIC – staging at BNL and JLab
Stage I Stage I Stage II Stage II eRHIC @ BNLeRHIC @ BNL
MEIC / EIC @ JLabMEIC / EIC @ JLab √s = 13 – 7013 – 70 GeV
Ee = 3 – 12 GeV
Ep = 15 – 100 GeV
EPb
= up to 40 GeV/A
√s = 34 – 7134 – 71 GeV
Ee = 3 – 5 (10 ?) GeV
Ep = 100 – 255 GeV
EPb
= up to 100 GeV/A
√s = up to ~180180 GeV
Ee = up to ~30 GeV
Ep = up to 275 GeV
EPb
= up to 110 GeV/A
√s = up to ~140140 GeV
Ee = up to 20 GeV
Ep = up to at least 250 GeV
EPb
= up to at least 100 GeV/A
(EIC)(EIC)((MMEEIICC))
6
Already the first stage of an EIC gives access to sea quarks and gluons
Need polarization and good acceptance to detect spectators & fragments
An EIC aims to study the sea quark and gluon-dominated matter.
Stage I+II
Stage I
EIC staging
JLab12 GeV
7
Gluon dynamics plays a large role in proton spinGluon dynamics plays a large role in proton spin
3D structure of nucleons and nuclei is not trivial3D structure of nucleons and nuclei is not trivial
Why do quarks contribute so little (~30%) to proton spin?Why do quarks contribute so little (~30%) to proton spin?
How do gluons and quarks bind into 3D hadrons?How do gluons and quarks bind into 3D hadrons?
Gluons in nucleiGluons in nuclei (light and heavy) (light and heavy)
Does the gluon density saturate at small x?Does the gluon density saturate at small x?
Physics highlights from the EIC program
EIC stage II measurement?
EIC stage I measurement
Saturation-scale dependence on centrality of collision (from fragnent detection)?
8
Elastic form factors
Transverse spatial distributions
(Naively Fourier transform of Q2 or t)
Parton Distribution Functions
Longitudinal momentum distributions
Generalized Parton Distributions
A unified descriptions of partons
(quarks and gluons) in momentum
and impact parameter space
3D structure (also important part of JLab 12 GeV)
9
Imaging in coordinate and momentum space
k yk
x
-0.5
0.5
0.0
-0.5 0.0 0.5
bx [fm] Anselmino et al., 2009QCDSF/UKQCD
Coll., 2006
2+1 D picture in momentum space2+1 D picture in impact-parameter space
TMDsGPDs
• Accessed through Semi-InclusiveSemi-Inclusive DIS
• OAM through spin-orbit correlations?
• Accessed through exclusive processes
• Ji sum rule for nucleon spin
F1T
┴(x
) [
Siv
ers
fun
ctio
n]
quark density
Lattice QCD
10
Imaging in coordinate and momentum space
2+1 D picture in momentum space2+1 D picture in impact-parameter space
TMDsGPDs
Transverse gluon distribution from J/ψ production
J/ψγProjections from EIC white paper
11
Measure G - gluon polarization
Measure TMD and GPDs - orbital motion
Two complementary approaches to resolve proton spin puzzleTwo complementary approaches to resolve proton spin puzzle
The number ½ reflects both intrinsic parton properties and their interactions
The spin of the proton
½polarization orbit polarization orbit
quarks gluons
~ 0.35 small ?? ?
12
(DVCS)(DVMP)
• DVCS on a transversely polarized target is sensitive to the GPD EGPD E
– GPD H GPD H can be measured through the beam spin asymmetry
– Opportunity to study spin-orbit correlations
x = 10-3
F. Sabatie
Ei(x, ξ, t) = κi(t) Hi(x, ξ, t)
Error bars shown only for κsea = +1.5
DVCS on proton
Model:
GPDs and angular momentum
1
1
),,(),,(2
1txEtxHxdxJ qqq
13
current data
w/ EIC data
Longitudinal spin – ΔG (gluon polarization)
JLab 12 GeV
• EIC stage I will greatly improve our understanding of ΔG– Stage II will further reduce the uncertainty
Q2 = 10 GeV2
M. Stratmann
Green:Green: RHIC spin, etc
Red:Red: EIC Stage I (MEIC or eRHIC)
Yellow:Yellow: EIC Stage II
14
pT
q
h
*e
e
tproduction>RA
pT2 vs. Q2
Quark propagation in matter (hadronization)
tproduction< RA
pT
q*
e
e
h
• Broadening of pT distribution
Accardi, Dupre
large υ smaller υlarge υ
• Heavy flavors: B, D mesons, J/Ψ
• Hadron jets at s > 1000 GeV2
• Impact parameter dependence?Impact parameter dependence?
– Fragments and “wounded nucleons” can help understanding the path length
R. Dupre, POETIC, Valparaiso, Chile
15
Forward processes
Recoils in deep-exclusive (diffractive) processesRecoils in deep-exclusive (diffractive) processes
Partonic fragmentation in SIDISPartonic fragmentation in SIDIS
Nuclear fragmentsNuclear fragments
• Large t (pT) range desirable
– Small t: needs Roman pots to go close to the beam
– Large t: needs large acceptance magnets
• Spectator tagging with light ions
• Acceptance for complete final state (all fragments) in heavy-ion reactions?
– Could measure mass-, multiplicity-, and angular distributions and correlations
• Coherent nuclear processes
– Recoiling heavy ions are very difficult to detect
– Good acceptance at small pT extends the mass range
• Also decays of strange and charmed baryons
16
Opportunities with polarized light ions
Partonic structure of the neutron (including spin)Partonic structure of the neutron (including spin)
The bound nucleon in QCDThe bound nucleon in QCD
Collective quark/gluon fieldsCollective quark/gluon fields
• Spin/flavor decomposition of parton densities
• Modifications of nucleon's quark/gluon structure due to nuclear binding
– Off–shellness controlled by kinematics
• Coherent scattering probes the quark/gluon field of the entire nucleus
• Tensor-polarized structure function of deuterium identifies the QCD double-scattering contribution - insight into onset of gluon saturation
Experiment: requires excellent forward detection and spectator tagging
Theory: combines high-momentum-transfer processes and low-energy nuclear structure
FY13 LDRD funds approved at JLab for joint project (theory / experiment)
17
Z. Kang
„If one could tag neutron, it typically leads to larger asymmetries“
Sivers asymmetryEIC kinematics
Z. Kang
D p
nn X
CLASCLAS + BoNuS
EIC
Spectator tagging with vector-polarized deuterium
Deuteron momentum distribution and comparison with fixed target
18
New Hall
Add arc
Enhanced capabilitiesin existing Halls
Add 5 cryomodules
Add 5 cryomodules
20 cryomodules
20 cryomodules
Upgrade arc magnets and supplies
CHL upgrade
The completion of the 12 GeV Upgrade of CEBAF was ranked the highest priority in the 2007 NSAC Long Range Plan.
Scope of the project includes: • Doubling the accelerator beam energy• New experimental Hall and beamline• Upgrades to existing Experimental Halls
New C100 cryomodules in
linac tunnel
JLab 12 GeV upgrade – probing the valence quarks
19
The EIC at Jefferson Lab
• Circumference of MEIC and CEBAF are similar (1.4 km)
• 12 GeV CEBAF is a full-energy e-injector
– Continuous injection as at SLAC (~50 nA)
– Parallel running with fixed target Halls A-D
• MEIC can store 20-100 GeV protons, or heavy ions up to 40 GeV/A.
• The stage II EIC will increase the energy to 250 GeV for protons and 20 GeV for electrons.
• Two high-luminosity, full-acceptance detectors
– IP2 could also host IP2 could also host sPHENIXsPHENIX.
Pre-booster
Ion linacMEIC
High-Energy Arc (Stage II)
e injection
IP2
IP1
Hall D
Halls A-C
C E
B A
F
20
The MEIC magnets and tunnel (in CAD)
• The MEIC magnetic lattice design is complete
The background is an artist's impression
• Detector locations minimize synchrotron- and hadronic backgrounds
• Close to arc where ions exit• Far from arc where electrons exit
21
MEIC – specific design goals
Spin control for all light ionsSpin control for all light ions
Full-acceptance detectorFull-acceptance detector
Minimized technical riskMinimized technical risk
• Figure-8 layoutFigure-8 layout
• Vector- and tensor polarized deuteriumVector- and tensor polarized deuterium
• Ring designed around detector requirementsRing designed around detector requirements
• Detection of all fragments – nuclear and partonicDetection of all fragments – nuclear and partonic
MEIC
EIC
MEIC
(arXiv:1209.0757)
Stable concept – detailed design report released August 2012
22
Dechirper Rechirper
Cooler Test Facility @ JLab FEL ERL
Regular electron cooling using a 55 MeV ERL
• Regular electron cooling is a well- established technique (e.g. at Fermilab)
• Recirculator tests are planned at the JLab Free-Electron Laser (FEL)
• A single-pass Energy-Recovery Linac (ERL) allows reaching higher electron energies and currents
• A recirculator ring is not needed for the ready-to-build MEIC cooling scheme, but would reduce the current requirements
23
Science program demands highly polarized (>70%) light ions (p, D, 3He, Li, ...)
Figure-8 shape used for all ion booster and collider rings Spin precession in one arc is canceled by the other arc
No preferred periodic spin direction
Energy-independent spin tune
Simplified polarization control and preservation for all ion species
Needs only small magnetic fields (instead of Siberian Snakes) to control polarization at Ips
The electron ring has a figure-8 shape because it shares a tunnel with the ion ring
• Figure-8 ring is the only practical way to accelerate polarized deuteronspolarized deuterons
Pre-Booster
Large booster
Ion Collider Ring
MEIC – ion polarization in figure-8 ring
24
far forwardfar forwardhadron detectionhadron detectionlow-Q2
electron detectionelectron detection large-apertureelectron quads
small-diameterelectron quads
central detectorcentral detector with endcaps
ion quads
50 mrad beam(crab) crossing angle
n,
ep
p
small anglesmall anglehadron detectionhadron detection
~60 mrad bend
(from GEANT4)
2 Tm 2 Tm dipoledipole
EndcapEndcap Ion quadrupolesIon quadrupoles
Electron quadrupolesElectron quadrupoles
1 m1 m11 m m
The MEIC full-acceptance detector concept
Forward hadron detection in three stages:Forward hadron detection in three stages:
1. Endcap with 50 mrad crossing angle1. Endcap with 50 mrad crossing angle
2. Small dipole covering angles up to a 2. Small dipole covering angles up to a few degreesfew degrees
3. Far-forward, up to one degree, for 3. Far-forward, up to one degree, for particles passing the accelerator quadsparticles passing the accelerator quads
IP FP
Roman potsRoman potsThin exit Thin exit windowswindows
Fixed Fixed trackers in trackers in vacuum?vacuum?
Trackers and “donut” calorimeterTrackers and “donut” calorimeter
RICH+
TORCH?
dual-solenoid in common cryostat4 m coil
barrel DIRC + TOF
EM
cal
ori
met
er
EM calorimeter
Tracking
EM
cal
ori
met
er
e/π
th
resh
old
Ch
eren
kov
25
dual-solenoid in common cryostat4 m coil
EM
cal
ori
met
er
e/π
th
resh
old
Ch
eren
kov RICH
+TORCH?
EM calorimeter
EM
cal
ori
met
er
barrel DIRC + TOF
Central detector options
(top view)
• Goal: two sufficiently different, complementary central detectors
– No need to for beam sharing at a ring-ring collider!
• First TOSCA model of 3T dual solenoid.
– Inspired by ILC 4th concept detector
• IP2 can be instrumented using an old magnet (CLEO or BaBar)
– Focus on hadronic calorimetry + small TPC central tracker
• IP1 (shown above): new 3T dual solenoid and large tracker
– Si-pixel disks and micropattern (GEM/micromega) forward trackers
– Low-mass cluster-counting He-filled DC (or micropattern) central tracker
2 m deep1 m deep
3 m
Co
il w
all
5 m3 m
Si-pixel vertex + disks
central tracker
forward tracker
forward tracker
Co
il w
all
26
low-Q2 tagger
final focusing elements
e-
ions
The low-Q2 tagger – small angle electron detection
e-
ions
Electron beam aligned with solenoid axis
x e-
(top view)
• High-resolution detection of low-Q2 electrons using dipole chicane.
27
89 T/m, 9.0 T, 1.2 m89 T/m, 9.0 T, 1.2 m 51 T/m, 9.0 T, 2.4 m51 T/m, 9.0 T, 2.4 m36 T/m, 7.0 T, 1.2 m36 T/m, 7.0 T, 1.2 m
Permanent magnetsPermanent magnets
34 T/m34 T/m 46 T/m46 T/m 38 T/m38 T/m2 x 15 T/m2 x 15 T/m e
5 T, 4 m dipole5 T, 4 m dipole
Ion quadrupoles: Ion quadrupoles: gradient, peak field, lengthgradient, peak field, length
2 T 2 T dipoledipole
Endcap detectorsEndcap detectors
Electron quadrupolesElectron quadrupoles
TrackingTracking CalorimetryCalorimetry
1 m1 m1 m1 m
• Large crossing angle (50 mrad)– Moves spot of poor resolution along solenoid axis into the periphery
– Minimizes shadow from electron FFQs
Crossing angle
• Dipole before quadrupoles further improves resolution in the few-degree range
Hadron detection between endcap and ion quads
• Low-gradient quadrupoles allow large apertures for detection of all ion fragments
– Peak field = quad gradient x aperture radius
7 m from IP to first ion quad7 m from IP to first ion quad
28
ions
Hadron detection after the ion quads
e
p
(n, γ)20 Tm dipole (in)
2 Tm dipole (out)
solenoid
Roman pots at Roman pots at focal pointfocal point
Thin exit Thin exit windowswindows
Aperture-free drift spaceAperture-free drift spaceZDCZDC
S-shaped dipole configuration S-shaped dipole configuration optimizes acceptance for neutralsoptimizes acceptance for neutrals
50 mrad crossing angle
• Transport optimizes acceptance for both neutrals and charged particles.
• Lots of space for Zero-Degree calorimeter (ZDC) on the outside of the ring
– Hcal and EMcal
• Hcal: RD52 „DREAM“ or particle flow type.
29
1. Good acceptance for ion fragments1. Good acceptance for ion fragments (rigidity different from (rigidity different from beam)beam)
2. Good acceptance for low-p2. Good acceptance for low-pTT recoil baryons recoil baryons (rigidity similar to (rigidity similar to
beam)beam)
3. Good momentum- and angular resolution3. Good momentum- and angular resolution
• Large downstream magnet apertures• Small downstream magnet gradients (realistic peak fields)
• Roman pots not needed
• Small beam size at second focus (to get close to the beam)• Large dispersion (to separate scattered particles from the beam)
• Roman pots important
• Large dispersion (but with D = D' = 0 at IP)• Long, instrumented, magnet-free drift space
4. Sufficient separation between beam lines (~1 m)4. Sufficient separation between beam lines (~1 m)
Ultra-forward hadron detection – requirements
D n
Xp
DVCS on the proton
30
Asymmetric ion optics
• Beam angular spread is also proportional to √(ε/β*) → good ion beam cooling essential
Ions x
IP FP
βx* = 10-20 cm
βy* = 2 cm
D* = D'* = 0
βFP < 1 m
DFP ~ 1 m
• β* x/y asymmetry allows a high luminosity with relatively small βmax
• Only dispersion component (D) generated after the IP aids detection, but dispersion slope (D') at IP adds to the beam angular spread (Dp/p)
• 7 m from IP to first downstream ion quad
31
Δp/p = -0.5 Δp/p = 0.0 Δp/p = 0.5
Far-forward detection of charged fragments
(protons rich fragments)
(exclusive / diffractive recoil protons)
(tritons from N=Z nuclei)(spectator protons from deuterium)
(neutron rich fragments)
• For light ions focus the mass/charge ratio of fragments is usually very different from the beam
• For heavy ions one needs both good acceptance and momentum resolution
32
6 T max
9 T max
horizontal plane vertical plane50 mr crossing angle in ion beam
Forward acceptance vs.magnetic rigidity
RedRed: Detection before ion quadrupoles
BlueBlue: Detection after ion quadrupoles
magnetic rigidity relative to beam
acc
eta
nce
(d
eg
ree
s)
• The angle is the original scattering angle at the IP
Fragment acceptance vs quadrupole peak field
• Q3P can be weaker– “9 T” is actually 9, 9, and 7 Tesla
33
e
p
n, γ
Far-forward hadron detection summary
20 Tm dipole2 Tm dipole
solenoid
• Excellent acceptance for all ion fragmentsall ion fragments
• NeutralsNeutrals detected in a 25 mrad (total) cone down to zero degreesdown to zero degrees
– Space for large (> 1 m diameter) Hcal + Emcal
• Recoil baryonRecoil baryon acceptance:
– up to 99.5%99.5% of beam energy for all anglesall angles
– down to at least 2-3 mrad2-3 mrad for all momentaall momenta
– full acceptance for x > 0.005
• 15 MeV resolution for a 50 GeV/A tagged deuteron beam
• Resolution limited only by beam
– Longitudinal (dp/p): 3x103x10-4-4
– Angular (all φ): 0.2 mrad0.2 mrad
34
Floor space in the new test lab at JLab – DVCS solenoid shown.
CLAS DVCS solenoid with 9 inch boreCLAS FROST solenoid with 5 inch bore
Generic detector R&D for an EIC – example• R&D program not site specific
– Coordinated by Tom Ludlam (BNL)
• As part of the program, a new, permanent EIC facility for sensor tests in high magnetic fields is being set up at Jefferson Lab
– Two 5T magnets provided by JLab
• Tests will include MCP-PMTs with small pore size (2-6 μm), SiPMs and LAPPDs
Non-magnetic dark box with pulsed LED for the DVCS solenoid – note the GlueX SiPM
(Hamamatsu S11064-050P(X))
35
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025
12 GeV Upgrade
FRIB
EIC Physics Case
NSAC LRP
EIC CD0
EIC Machine Design/R&D
EIC CD1Downselelect
EIC CD2/CD3
EIC Construction
Assumes endorsement for an EIC at the next NSAC Long Range PlanAssumes relevant accelerator R&D for down-select process done around 2016
EIC timeline
36
Summary
• JLab or BNL implementations possible– Agreement on global parameters
– Collaboration on detector R&D
The EIC is the next-generation US QCD facilityThe EIC is the next-generation US QCD facility
The EIC at JLab offers some unique capabilitiesThe EIC at JLab offers some unique capabilities
Complementarity with the LHC/LHeCComplementarity with the LHC/LHeC
• The MEIC will cover all kinematics between JLab 12 GeV and the LHeC
• Vector- and tensor polarized deuterium
• Excellent detection of recoil baryons, spectators, and target fragments
– Full acceptance, high resolution
37
Backup
38
Electron source for electron cooling
• MEIC high-energy electron cooling requires a source of ~2 nC few-cm long electron bunches with a repetition rate of ~15 MHz and an average current of ~30 mA.
• Slava suggested using a short-pulse high-bunch-charge high-repetition-rate magnetized DC gun with subsequent bunch compression.
• Such a scheme, with the exception of beam magnetization, seem to exist at BINP and has been used as an electron source for NovoFEL since 2003.
• A DC gun with a thermionic metal-oxide gridded cathode has the following parameters:
• All gun (and injector) parameters seem consistent with the electron cooler requirements.
• A thermionic RF gun is being developed at BINP to replace the DC gun. The reasons for upgrade are not clear but, perhaps, involve improved cathode lifetime and higher current.
• Beam tests of the RF gun started recently with encouraging results.
Electron energy 300 kVBunch charge ~1.5-2 nCMaximum peak current ~1.8 AMaximum average current ~30-45 mAMaximum bunch repetition rate 22.5 MHzBunch length ~1.3 nsNormalized emittance 10 mmmrad
39
Electron polarization – continuous injection
• Polarization at t+Δt
0))(1( PN
Nt
t
PP
N
NP t
ttt
10 )1(
injdk
ringrevequ I
ITPP
• Equilibrium Polarization
Lost or Extracted P0
(>Pt)Pt
• Note that:– Polarization lifetime at 5 GeV is 1 or 3 hours depending on helicity
(Sokolov-Ternov)– 50-100 nA beam injected from CEBAF can maintain polarization close
to its initial value of 80% indefinitely for any elevtron beam energy.
Equilibrium Polarization vs. Average Injected Current
40
Deuteron polarization in figure-8 ring
• Beam injected longitudinally polarized, accelerated and then desired spin orientation adjusted
(B||L)1,2 (Tm) vs. p (GeV/c)
longitudinal polarization
radial polarization
y
yz
sin
)sin(1
yz
sin
sin2
Gy
is the spin rotation angle between the solenoids
is the orbit rotation angle between the solenoids is the angle between the polarization and velocity directions
21, zz are the spin rotation angles in the solenoids
B
LBG i
zi||1
41
Proton polarization in figure-8 ring
Last two arc dipoles
(BL)i (Tm) vs. p (GeV/c)
longitudinal polarization radial polarization(BxL)1
(BxL)2
(BxL)3
(BxL)4
42
Ion acceptance and resolution at the focal point
±10σ beam size at 60 GeV
• Large deflections allow precise tracking over long distances with cheaper detectors
– Particles with deflections > 1 m at the FP will be detected closer to the dipole
• Detection past the focal point is also possible, but with acceptance restrictions
43