EIC Overview Rolf Ent Nuclear Chromo-Dynamic Studies with a Future EIC Workshop at Argonne National Laboratory April 07-09, 2010 (mainly accelerator and detector parameters) Lots of credit to Yuhong Zhang, Alex Bogacz, Slava Derbenev, Geoff Krafft, Tanja Horn, Charles Hyde, Pawel Nadel-Turonski for multiple accelerator and detector/interaction region ideas

EIC Project - Status NSAC 2007 Long-Range Plan: “An Electron-Ion Collider (EIC) with polarized beams has been embraced by the U.S. nuclear science community as embodying the vision for reaching the next QCD frontier. EIC would provide unique capabilities for the study of QCD well beyond those available at existing facilities worldwide and complementary to those planned for the next generation of accelerators in Europe and Asia.”

Current Ideas for a Collider Energiessluminosity to 11 x Close to Future Up to 11(22?) x (22000?)Close to Staged Up to 4 x Close to to 20 x Few x to 3 x 15180Few x to 70 x Design Goals for Colliders Under Consideration World-wide Present focus of interest (in the US) are the (M)EIC and Staged MeRHIC versions, with s up to 2650 and 4000, resp.

4 GeV e x 250 GeV p – 100 GeV/u Au MeRHIC 2 x 60 m SRF linac 3 passes, 1.3 GeV/pass STAR PHENIX V.N. Litvinenko, RHIC S&T Review, July 23, pass 4 GeV ERL Polarized e-gun Beam dump MeRHIC detector MeRHIC Medium Energy IP2 of RHIC (IP12?) Up to 4 GeV e - x 250 GeV p L ~ cm -2 sec -1 s =

A High-Luminosity EIC at JLab - Concept Legend: MEIC 1 low-energy IR (s ~ 200) 2 medium-energy IRs (s < 2600) ELIC =high-energy (s = 11000) (E p ~ 250 limited by JLab site) Use CEBAF “as-is” after 12-GeV Upgrade Note: s HERMES = 51unpol. L = s COMPASS = 340unpol. L = s COMPASS++ = 340L =

assumptions: (x,Q 2 ) phase space directly correlated with s (=4E e E p ) Q 2 = 1 lowest x scales like s Q 2 = 10 lowest x scales as 10s -1 General science assumptions: (“Medium-Energy”) option driven by: access to sea quarks (x > 0.01 or so) deep exclusive scattering at Q 2 > 10 (?) any QCD machine needs range in Q 2  s = few seems right ballpark  s = few 1000 allows access to gluons, shadowing Requirements for deep exclusive and high-Q 2 semi-inclusive reactions also drives request for (lower &) more symmetric beam energies. Requirements for very-forward angle detection folded in IR design x = Q 2 /ys

 Energy Wide CM energy range between 10 GeV and 100 GeV Low energy: 3 to 11 GeV e on 3 to 12 GeV p (and ion) Medium energy: up to 11 GeV e on 60 GeV p or 24 GeV/n ion and for future upgrade High energy: up to 11 (22?) GeV e on 250 GeV p or 100 GeV/n ion  Luminosity up to cm -2 s -1 per collision point Multiple interaction points  Ion Species Polarized H, D, 3 He, possibly Li Up to heavy ion A = 208, all stripped  Polarization Longitudinal at the IP for both beams, transverse for ions Spin-flip of both beams All polarizations >70% desirable  Positron Beam desirable MEIC/ELIC Design Goal

MEIC Design Choices Achieving high luminosity  Very high bunch repetition frequency (1.5 GHz)  Very small β* to reach very small spot sizes at collision points  Short bunch length (σ z ~ β*) to avoid luminosity loss due to hour-glass effect (unless other mitigation schemes used)  Relatively small bunch charge for making short bunch possible  High bunch repetition restores high average current and luminosity MEIC High repetition rate Small bunch charge Short bunch length Small β* MeRHIC Low repetition rate Very high bunch charge Long bunch length Large β* VS. This luminosity concept has been tested at two B-factories very successfully, reaching luminosity above cm -2 /s -1

MEIC Design Choices (cont.) Ring-ring vs. Linac(ERL)-ring  Linac-ring option requires very high average current polarized electron sources (few decades times state-of-the-art)  Linac-ring option does not help ELIC which already employs high repetition electron and ion beams  ring-ring collider Ensuring high polarization for both electron beam and light ion beams  Figure-8 shape rings  Simple solution to preserve full ion polarization by avoiding spin resonances during acceleration  Energy independence of spin tune  g-2 is small for deuterons; a figure-8 ring is the only practical way to arrange for longitudinal spin polarization at the IP

MEIC Luminosity: Energy Dependence MEIC luminosity is limited by  Electron beam current  synchrotron radiation ~ N e γ 4 /ρ (radiation power, emittance degradation)  Proton/ion beam current  space charge effect ~ N i / γ 2 (emittance growth, tune-shift/instabilities)  (Vertical) beam-beam tune-shift (bad collisions) ~ 1/ γ Main design limits (based on experience)  Electron beam SR power density: ≤ 20 kW/m  Ion beam space charge tune-shift: ≤ 0.1  (Vertical) beam-beam tune-shift: ≤ (proton), ≤ 0.1 (electron) Given an energy range, MEIC collider ring optimized with  Synchrotron radiation  prefers a large ring (for more arc bends)  Space charge effect of i-beam  prefers a small ring circumference  Multi IPs and other components require long straight sections

MEIC Luminosity with a Compact Ring 660 m ring circumference was determined to be the minimum arc length required to accommodate 60 GeV protons in a Figure-8 ring Illustration is based on back-of-envelope estimations, using three main design limits, omitting more complicated beam dynamics issues Given the ring size, the synchrotron limit determines the optimal luminosity/electron energy. Beyond this electron energy a fast drop. Just an illustration

Luminosity of 60 GeV protons in a 1 km ring, based on a back-of-the- envelope calculation, is lower than in 0.6 km ring, since space charge effects are more severe in a larger ring Luminosity of 100 GeV protons is on the other hand much better in a 1 km ring, since space charge effects are reduced with higher energy Consider what the minimum luminosity is at both lower and higher ion energies! Just an illustration MEIC Luminosity with a Compact Ring

Near-Term MEIC Design Parameters ElectronProton Collision energyGeV3 – Ion booster 3–12 GeV, ring accepts 12 GeV injection Max dipole fieldT6Not too aggressive after LHC Max SR powerkW/ m 20Factor two beyond best achieved? Max currentA21~ max B-factory current, HOM in component HERA 0.15 A (?) RHIC 0.3 A RF frequencyGHz1.5 Use combination of gap (crossing angle) and RF shift to accommodate lower ion energies Bunch lengthmm556 mm demonstrated in B-factory, 10 cm in RHIC (?) IP to front face of 1 st quad (l) m+/- 3 to 4+/- 7 Vertical β*cm22Keep β max below 2 km, with  max = l 2 /  * Crossing anglemrad10050 to 150 desired for detector advantages Luminosity expected to reach up to 1 x e-nucleons/s/cm 2 around 60x5 GeV 2

Detector/IR in simple formulas  max ~ 2 km = l 2 /  * (l = distance IP to 1 st quad) IP divergence angle ~ 1/sqrt(  *) Luminosity  /  * Example: l = 7 m,  * = 20 mm   max = 2.5 km Example: l = 7 m,  * = 20 mm  angle ~ 0.3 mr Example:12  beam-stay-clear area  12 x 0.3 mr = 3.6 mr ~ 0.2 o Making  * too small complicates small-angle (0.5 o ?) detection before ion Final Focusing Quads, AND would require too much focusing strength of these quads, preventing large apertures (up to 0.5 o ?)

Figure-8 Ion Ring – Optics Arc length C = 120 m + 20 m (for spin manipulation) m (increased due to assumption of 60 GeV & 6T max dipole fields) Straight length L = 194 m Total Ring Circumference would then be: 2 (L + C) = 908 m Uncompensated dispersion from arcs

Interaction Region Optics (ions) l * = 7m IP FF triplet : Q3 Q2 Q1 f Q1 G[kG/cm] = -9.7 Q2 G[kG/cm] = +6.7 Q3 G[kG/cm] = -6.3 Natural Chromaticity:  x = -44  y = -137 Q3 aperture 10 cm m)  7T peak field  Particles < 0.5 degrees through FF quads

total ring circumference: 908 m 64 degrees arc/straight crossing angle Figure-8 Collider Rings Present thinking: ion beam has 100 mr horizontal crossing angle Advantages for dispersion, crab crossing, very-forward particles 200 mr bend would need 40 Tm ~20 m from IP (Reminder: MEIC/ELIC scheme uses 100 mr crab crossing)

Ion Ring – Beam envelopes  x,y = 0.2 mm IP ~ 20 meters FF about 10 m,  x,y = 2-3 mm Beam-stay-clear area near IP:   5 10 m = 5 mr Beam-stay-clear area away from IP: 8-10   2 20 m = 0.1 mr Very forward tagging?

Solenoid yoke integrated with a hadronic calorimeter and a muon detector EM calorimeter Overview of central detector layout EM calorimeter RICH (DIRC?) Tracking RICH Hadronic calorimeter Muon Detector? HTCC EM calorimeter ions electrons (not to scale) Time-of-flight detectors shown in green IP is shown at the center, but can be shifted left – Determined by desired bore angle and forward tracking resolution – Flexibility of shifting IP also helps accelerator design at lower energies (gap/path length difference induced by change in crossing angle) DIRC would have thin bars arranged in a cylinder with readout after the EM calorimeter on the left

RICH based on ALICE design might push the limit from 4 to 7 GeV – Requires a more detailed study Detector/IR - Kinematics With 12 GeV CEBAF, has the option of using higher electron energies – DIRC no longer sufficient for π/K separation RICH would extend the minimum diameter of solenoid from approximately 3 to 4 m – Main constraint since bore angle is not an issue in JLab kinematics 4 on 30 GeV s = 480 GeV 2 5 on 50 GeV s = 1000 GeV 2 (10 on 50 GeV) s = 2000 GeV 2 Q 2 > 10 GeV 2 – Vertical lines at 30° (possibly up to 40°) indicate transition from central barrel to endcaps – Horizontal line indicates maximum meson momentum for π/K separation with a DIRC

Detector/IR – Particle Momenta 4 on 6011 on 60 1 H(e,e’π + )n SIDIS  Need Particle ID for p > 4 GeV in central region  DIRC won’t work, need more space for RICH Need Particle ID for well above 4 GeV in forward region (< 30 o ?)  determines bore of solenoid In general: region of interest up to ~10 GeV/c mesons (not including very-forward angle detection, see later) { {

Pion momentum = 5 GeV/c, 4T ideal solenoid field Detector/IR – Magnetic Fields Add 2Tm transverse field component to get dp/p roughly constant vs. angle Resolution dp/p (for pions) better than 1% for p < 10 GeV/c obtain effective 1Tm field by having 100 mr crossing angle 200 mr ~ 12 o gives effective 2Tm field  need to add 1-2Tm dipole field for small- angle pions (1 o -6 o ) only

Downstream dipole on ion beam line has several advantages – No synchrotron radiation – Electron quads can be placed close to IP – Dipole field not determined by electron energy – Positive particles are bent away from the electron beam – Long recoil baryon flight path gives access to low -t – Dipole does not interfere with RICH and forward calorimeters Excellent acceptance (hermeticity) solenoid electron FFQs 100 mrad 0 mrad ion dipole w/ detectors (approximately to scale) ions electrons IP detectors ion FFQs 2+3 m 2 m Detector/IR cartoon exclusive mesons ° recoil baryons 4 on 30 GeV Q 2 > 10 GeV 2 Make use of a 100 mr crossing angle for ions!

Detector/IR – Forward Angles Example: map t between t min and 1 (2?) GeV  0.2 to 4.9 (9.8) 12 GeV  0.2 to 3.0 (5.9) 20 GeV  0.2 to 2.0 (3.9) 30 GeV  Cover between about 0.5 and 6 degrees? Example I:separation between 0.5 o and 0.2 o (BSC) ~ 2.5 cm at 5 meter distance May be enough for ~ 30 GeV protons and neutrons from an O(1A) beam (also need good angle (t) resolution!) Example II:6 degrees ~ 0.5 meter radius cone at 5 meter t ~ E p 2  2  Angle recoil baryons = t ½ /E p

Detector/IR – Forward Angles  Must cover between 1 and 5 degrees  Should cover between 0.5 and 5 degrees  Like to cover between 0.2 and 7 degrees  = 5  = 1.3 E p = 12 GeVE p = 30 GeVE p = 60 GeV t ~ E p 2  2  Angle recoil baryons = t ½ /E p

Detector/IR – Very Forward - Ion Final Focusing Quads (FFQs) at 7 meter, allowing ion detection down to 0.5 o before the FFQs - Use large-aperture (10 cm radius) FFQs to detect particles between 0.3 and 0.5 o (or so) in few meters after ion FFQ triplet  12 meters from IP = 2 mm 12  beam-stay-clear  2.5 cm 0.3 o (0.5 o ) after 12 meter is 6 (10) cm  enough space for Roman Pots & “Zero”-Degree Calorimeters - Large dipole 20 meter from IP (to correct the 100 mr ion horizontal crossing angle) allows for very-small angle detection (< 0.3 o )  20 meters from IP = 0.2 mm 10  beam-stay-clear  2 mm 2 mm at 20 meter is only 0.1 mr…  (bend) of 29.9 and 30 GeV spectators is 1.3 mr = 5 4 m Situation for zero-angle n detection very similar as at RHIC!

Forward Neutron Detection Thoughts – A Zero Degree Calorimeter The RHIC Zero Degree ColorimetersarXiv:nucl-ex/ v1 Context: The RHIC ZDC’s are hadron calorimeters aimed to measure evaporation neutrons which diverge by less than 2 mr from the beam axis. <2 mr at 18 meters from IP  neutron cone ~ 4 cm ZDC = 10 cm (horizontal) x 13 cm (vertical) (& 40 cm thick) Have good efficiency and only 1 cm “dead- edge” (albeit not very good  E resolution). Implication: do not make earlier ion bend dipole strong < 2 TM!

EIC Overview - Summary Near-term design concentrates on parameters that are within state-of-the-art (exception: small bunch length & small vertical  * for proton/ioan beams) Detector/IR design has concentrated on maximizing acceptance for deep exclusive processes and processes associated with very-forward going particles Exact energy/luminosity profile still a work in progress Summer 2010: MEIC design review followed by internal cost review (and finalizing input from user workshops) Many parameters related to the detector/IR design seem to be well matched now (crossing angles, magnet apertures/gradients/peak fields, field requirements), such we may end up with not too many “blind spots”.

Appendix

Proton Tagging 100 mr horizontal crossing angle for ion beam would require large 40Tm magnet at 20 meter from the IP. If so, can need this for spectator proton tagging  (bend) 30 GeV vs GeV = 1.3 mr If roman pots after 4 m  5 mm bend 1% (300 MeV/c) would become 15 mm Roman pots (photos at CDF (top) and LHC (bottom), …) ~ 1 mm from beam achieve proton detection with < 100  resolution  Proton tagging concept looks doable, even if the horizontal crossing angle was reduced by a factor of two or three.

Neutron Tagging The RHIC Zero Degree ColorimetersarXiv:nucl-ex/ v1 case: 40 Tm bend magnet at 20 meters from IP  very comparable to above RHIC case! 40 Tm bends 60 GeV protons with 2 times 100 mr  a distance of about 4 meters = 80 cm (protons)  no problem to insert Zero Degree Calorimeter in this design Zero Degree Calorimeter properties: Example: for 30 GeV neutrons get about 25% energy resolution (large constant term due to unequal response to electrons and photons relative to hadrons)  Should be studied more whether this is sufficient Timing resolution ~ 200 ps Very radiation hard (as measured at reactor)

Where we are - Design provides excellent luminosity for s ~ Good luminosity (10 33 or more???) down to s = 240 and up to s = 2640 (can access gluons down to x = or so) x = Q 2 /ys HERA experience: Can map between y = 0.8 and y = s = 1200 & x = 0.8: 12 (W 2 = 4) < Q 2 < s = 2640 & Q 2 = 1: down to x = 4.7x10 -4 Note: for deuteron Z/A = 0.5 for heavy ion Z/A = 0.4

CTEQ Example at Scale Q 2 = 10 GeV 2 “dip” in u,d pdf’s at x ~ 0.01 Q 2 =10 GeV 2 ) Gluon splitting dominates

“EMC Effect”F 2 A /F 2 D x The Venerable (Nuclear) EMC Effect Space-Time Structure of Photon x < (5 times ) for saturation in shadowing to start? Need about decade in Q 2 to verify LT vs. HT of effects  want to push down to x ~ Q 2 = 1) w. MEIC. E cm = 10 – 45 (s = 100 – 2000) is in the right ballpark for nucleon/nuclear structure studies

What E cm and Luminosity are needed for Deep Exclusive Processes? New Roads:   and  Production give access to gluon GPD’s at small x (<0.2)  Deeply Virtual Meson Q 2 > 10 GeV 2 Well suited processes for the EIC  transverse spatial distribution of gluons in the nucleon Can we do such measurements at fixed x in the valence quark region? Important to disentangle flavor and spin… fixed x:  ~ s/Q 2 (Mott) x 1/Q 4 (hard gluon exchange) 2 sLQ 2 reach DVCS Q 2 reach (e,e’  ) 12-GeV = 7 EIC example10003 x ~100~17

‘Relaxed’ IR Optics (electrons) l * = 3.5m IP FF doublets f Natural Chromaticity:  x = -96  y = -444

MEIC: Layout