Measurements of the Neutron Skin of 208Pb and 48Ca Caryn Palatchi University of Virginia JLab Hall A Collaboration Meeting January 18, 2017

CREX PREX Kent Paschke * UVa Krishna Kumar Stony Brook University Seamus Riordan* Stony Brook University Robert Michaels Jefferson Lab Kent Paschke UVa Paul Souder Syracuse Univeristy Dustin McNulty Idaho State University Juliette Mammei Manitoba University Silviu Covrig Kent Paschke * UVa Krishna Kumar Stony Brook University Robert Michaels Jefferson Lab Paul Souder Syracuse Univeristy Guido Urcioli INFN Rome * contact persons

Introduction to PREX/CREX Parity violation experiments map the weak charge distribution : unpolarized target σ ∝ |Mγ + Mweak|2 ~ |Mγ|2 + 2Mγ(Mweak)* +… γ proton neutron Electric charge 1 Weak charge ~0.08 for spin-0 nucleus

Introduction to PREX/CREX Parity violation experiments map the weak charge distribution : unpolarized target σ ∝ |Mγ + Mweak|2 ~ |Mγ|2 + 2Mγ(Mweak)* +… γ proton neutron Electric charge 1 Weak charge ~0.08 for spin-0 nucleus weak FF almost entirely couples to neutron distribution & e-charge FF entirely couples to proton distribution

Introduction to PREX/CREX Parity violation experiments map the weak charge distribution : unpolarized target σ ∝ |Mγ + Mweak|2 ~ |Mγ|2 + 2Mγ(Mweak)* +… The weak FF (and neutron radius) are not as well understood as the proton radius (and e-charge FF) γ proton neutron Electric charge 1 Weak charge ~0.08 for spin-0 nucleus weak FF almost entirely couples to neutron distribution & e-charge FF entirely couples to proton distribution

Introduction to PREX/CREX Parity violation experiments map the weak charge distribution : unpolarized target σ ∝ |Mγ + Mweak|2 ~ |Mγ|2 + 2Mγ(Mweak)* +… The weak FF (and neutron radius) are not as well understood as the proton radius (and e-charge FF) γ Ratio of weak to E&M FF is directly related to neutron skin thickness on heavy nuclei as predicted by nuclear theory proton neutron Electric charge 1 Weak charge ~0.08 for spin-0 nucleus weak FF almost entirely couples to neutron distribution & e-charge FF entirely couples to proton distribution

Introduction to PREX/CREX Parity violation experiments map the weak charge distribution : unpolarized target σ ∝ |Mγ + Mweak|2 ~ |Mγ|2 + 2Mγ(Mweak)* +… The weak FF (and neutron radius) are not as well understood as the proton radius (and e-charge FF) γ Ratio of weak to E&M FF is directly related to neutron skin thickness on heavy nuclei as predicted by nuclear theory proton neutron Electric charge 1 Weak charge ~0.08 for spin-0 nucleus weak FF almost entirely couples to neutron distribution & e-charge FF entirely couples to proton distribution Neutron skin thickness is highly sensitive to the pressure of pure neutron matter & EOS: the greater the pressure, the thicker the skin as neutrons are pushed out against surface tension. Apv provides clean measure of neutron skin thickness

Neutron skin measured by APV X. Roca-Maza, M. Centelles, X. Vi˜nas, and M. Warda, Phys. Rev. Lett. 106 252501 (2011) Models Weak charge density Predicted Asymmetry Solve Dirac Eq. Predicted Neutron Skin Models that predict a weak charge density and an electric charge density demonstrate neutron skin and Apv are not independent parameters Apv in PVES provides a clean probe of the neutron distribution One is a direct measure of the other Robust correlation between 208Pb APV and the neutron skin over existing nuclear structure models PREX: APV to 3% from 208Pb -> rn to 0.06 fm CREX: APV to 2.5% from 48Ca -> rn to 0.02 fm

Neutron skin and Symmetry Energy Mean-Field predictions show a clear correlation between neutron skin of a heavy nucleus and the density slope of the symmetry energy. Energy penalty for breaking n=Z symmetry Slope of symmetry energy at saturation density rn calibrates the Equation of State of neutron rich matter directly, constrains and guides models needed for heavy nuclei via L So far the probes for stable heavy nuclei have been strongly interacting, having a somewhat more complicated interpretation

APV = 0.657 ± 0.060(stat) ± 0.014(sym) ppm PREX-I Result APV = 0.657 ± 0.060(stat) ± 0.014(sym) ppm First observation that weak charge density more extended than (E+M) charge density → showed there is indeed a weak skin around a heavy nucleus. Interestingly, current PREX central value not consistent with measured neutron star properties and existing models. Rn - Rp= 0.33 +0.16-0.18 fm Phys Rev Let. 108, 112502 (2012), Phys. Rev. C 85, 032501(R) (2012) (200+ references on Inspire) PREX-2: achieve 3x improvement on Rn-Rp uncertainty

48Ca : CREX 48Ca: nuclear distribution 208Pb: bridge between ab initio models and effective theory (DFT) 48Ca: nuclear distribution influenced by finite size effects. Within reach of ab initio, microscopic calculations Chiral effective field theory, based on NN and NNN potentials coupled-cluster model of 48Ca: G. Hagen et al, Nature Physics 12, 186–190 (2016) 208Pb: lies further up in A in realm of uniform nuclear matter & DFT serves as terrestrial laboratory to test n-star structure PREX and CREX together span the nuclear landscape of highly dense matter

PREX / CREX in Hall A PREX-I Hall A HRS with septum Very clean separation of elastic events by HRS optics Analog integration of everything that hits the detector PREX-I Q2 ~ 0.009 GeV2 APV ~ 0.6 ppm Rate ~1 GHz 5o scattering angle PREX-II expects to achieve similar conditions

PREX/CREX Experiments PREX-2: 3% stat, 0.06 fm CREX: 2.4% stat, 0.02fm PREX-II E=1.1 GeV, 5o A=0.6 ppm 70 μA, 25+10 days CREX E=1.9 GeV, 5o A = 2.3 ppm 150 μA, 35 + 10 days PREX-I E=1.1 GeV, 5o A=0.6 ppm Charge Normalization 0.2% Beam Asymmetries 1.1% Detector Non-linearity 1.2% Transverse Asym Polarization 1.3% Target Backing 0.4% Inelastic Contribution <0.1% Effective Q2 0.5% Total Systematic 2.1% Total Statistical 9% Charge Normalization 0.1% Beam Asymmetries* 1.1% Detector Non-linearity* 1.0% Transverse Asym 0.2% Polarization* Target Backing 0.4% Inelastic Contribution <0.1% Effective Q2 Total Systematic 2% Total Statistical 3% Charge Normalization 0.1% Beam Asymmetries 0.3% Detector Non-linearity Transverse Asym Polarization 0.8% Target Contamination 0.2% Inelastic Contribution Effective Q2 Total Systematic 1.2% Total Statistical 2.4% *Experience suggests that leading systematic errors can be improved beyond proposal Achieved CREX more sensitive to Q2 uncertainty than PREX Rate, absolute precision is similar to HAPPEX-II, achievable statistics limited result, systematics well under control Suppress statistical errors by 3x

PREXI demonstrated technologies for PREXII What Worked: New Septum We now know how to tune it to optimize FOM Fast Helicity Flipping We know how to control false asymmetries and monitor performance HRS Tune We have a tune and good first-guess optics matrix for a tune optimized for the small detectors Injector Spin Manipulation Second Wein and solenoid are calibrated and used for helicity reversal. Important cancellation for systematic beam asymmetries from the polarized source. Polarimetry at low energy High-field Moller at 1.3%, Integrating Compton at 1.2% Beam Modulation System Fast beam kicks cancel low frequency noise and improve precision of beam position corrections New Detectors Suitable energy resolution achieved for 1 GeV electrons. <5% precision loss. AT false asymmetry AT is small (<1 ppm Pb, <10 ppm C) and Afalse will be small if PT is minimized Lead Target Established Min Survival lifetime >25 C exposure Needed to resolve: Target Vacuum system Radiation in Hall Ready to meet PREX-II/CREX systematic and statistical goals

Progress Still on target for being ready for Summer/Fall 2018 *Actual designer worked on this: Wayne Sachleben with Silviu Covrig Completed “physicist” design of target region Completed radiation shielding and collimator design new scattering chamber* to combine PREX/CREX installations Design document completed Detailed design work started Septum Modeled (TOSCA) Calcium target studies (surface contamination) Beam studies demonstrated accelerator readiness High-field Moller polarimeter commissioned with beam new Compton polarimeter commissioned with beam ERR in May 2016 Radiation sufficiently controlled Second ERR in Spring 2017

New PREX / CREX Scattering Chamber One cryo-cooled production target ladder and one optics ladder for single target location Following recommendations after initial design/engineering discussions Solves vacuum and mechanical assembly considerations cost effective Improves PREX/CREX installation compatibility Allows more efficient shielding Septum Collimator Box Silviu Covrig, consulting with target group Scattering Chamber

Power in collimator (W/μA) Radiation PREX-I distributed significant power throughout the hall, damaging electronics 1 MeV neutron-equivalent integrated dose Solution: Localize power in hall at collimator, and shield it HRS PS platform PREX-2 (1 MeV-n / cm2 ) PREX-2 /PREX PREX-2 /HAPPEX-2 electrons 1.4E+10 11% 70% neutrons 7.6E+09 7% 94% total 2.1E+10 9% 83% PREX-I PREX-II CREX Power in collimator (W/μA) 9.7 28.8 6.8 Power in hall (W/μA) 18.0 3.0 ~1.5 1 MeV-neq/cm2 is below expected electronics damage threshold (1011-1013) PREX-2 is an order of magnitude below PREX-I, equivalent to HAPPEX-2 PREX+CREX~2.5x HAPPEX2 high-energy n rate (SEU) is also down by 5-10x beam collimator lead poly septum Site Boundary Dose tungsten JLab annual administrative limit: 10 mrem Extrapolation from PREX-I measurements is well within limits (~ 1.7mrem PREX+2.7mrem CREX = 4.4mrem Total < 50% limit). Std RadCon simulations are higher

PREX-II collimator Collimator front face 85cm from target, intercepts electrons >0.78o Power deposited: 2.1kW @ 70 μA Water Circuit Inner cylinder 30% Cu-70% W Water-cooled with Cu brazed sleeve, similar to Qweak collimator Outer box: Tungsten Tungsten PREX-I collimator: >1.27o, ~500W much larger opening angle <1/3 power deposited Outer tungsten cover traps E&M power, self-shields produced neutrons

g2p Configuration (3 coils and shims) for CREX requirements Septum Septum required for 5o PREX/CREX(same target position for both) Use existing power supplies and water-cooling systems g2p Configuration (3 coils and shims) for CREX requirements PREX: No change to performance CREX: gap-shimmed, 3-coil septum Septum used in the three-coil configuration as in g2p measurement 5 degree configuration for CREX brings the current density into a same regime as g2p Use the same shims as g2p Same water-cooling system as was adequate for g2p Current density, water flow & water pressure are manageable for present target position CREX Central Field = 13.2 kG Data with shims well reproduced by TOSCA/ANSYS (Juliette Mammei, Manitoba) Data w/o shims

Lead / Diamond Target Never degraded! 0.5mm lead, 0.25mm diamond, 1 sqin Lead has low melting point, and low thermal conductivity Diamond foils have excellent thermal conductivity, He cooled 12C is isoscaler, spin-0 (and well-measured) so benign background! Raster Scan to measure density loss - PREX I test Use synchronized 4x4mm raster to handle non-uniform lead thickness thin diamond Survived >1week production Rate drop - 92% Never degraded! Last 4 days at 70 uA thick diamond Melting - CFD thermal analysis performed 3 targets for PREX1 were used for 1/2 #C on target of PREXII, 6 enough 10 Production ‘thick diamond’ targets for PREX-2: factor of ~2 (or more) margin based on PREX-I performance

CREX target Run “tilted" at 45 to compensate for thinner target - 1.1g/cm2 Windowless target at 45 provides best, simplest option for running Scraping of oxidized surface sufficient for decontamination Contamination tolerances are loose and shouldn't present a problem Relevant contaminant nuclei asymmetries are known, within 10% of 48Ca, and have rates suppressed by Z2.  Overall corrections are negligible (<0.16% )with very conservative residual contamination estimates (10%) Narrow raster minimizes potential geometric-dependent systematics Transfer under oil should prevent oxidation under normal conditions Scattering chamber isolated with gate valves & gas purge in case of vacuum leaks Collaboration will have help of target group to finalize design

Integrating Detectors PREX-I: RMS/Mean ~ 50% (12% penalty) Thin quartz integrating e-detectors with phototubes to detect Cherenkov radiation from particles traversing the quartz Design Finalized for PREXII Tested at Mainz Simulation and data benchmarked quartz properties GEM tracking chambers on track Tested at Idaho and Stony Brook Beam tests at Mainz Mainz: RMS/Mean ~ 19% (2% penalty)

Beam AQ: 100-300 ppm RMS. Δx: 5-25 um RMS. (20 μA, 30 Hz, 2.2 GeV) Tests during recent parasitic beam studies to check beam quality, monitors BCM resolution exceeds PREX requirements “double-difference” width PREX-II statistical width ~ 120ppm @ 30Hz octet/quartet BCM resolution of 40ppm would be 5% loss of resolution Resolution ~10-30 ppm 1 MHz BCM electronics: ~25 ppm resolution @ 30 Hz, 20uA Confirmed by excess noise with small-angle monitors Similar to width measured in PREX-I Charge and position jitter looks similar to 6 GeV era AQ: 100-300 ppm RMS. Δx: 5-25 um RMS. (20 μA, 30 Hz, 2.2 GeV) 1 pass beam 4 pass beam X, Y

Moller Polarimetry Spectrometer Iron Foil Target 2015/16 operation On track for <1% in PREX-2 / CREX Iron Foil Target Spectrometer High field @90° New magnet New target frame Simulated and Tuned for PREX Energy 2015/16 operation Shown 0.4% statistical error Target system to Test Lab for studies in 2017 - reinstall for PREX/CREX Rigid target rotator parts assembled, tested, en route Kerr monitor testing (MOLLER pre-R&D), in situ monitor of MFe Commissioned at high E, needs low E commissioning Study possible systematic effects, in data and simulation (Levchuck etc.) Four targets, in frame designed for rotation and Kerr monitoring

Compton Polarimetry S/B: 38/1 Compton polarimetry techniques on track for PREX2/CREX Laser S/B: 38/1 Green laser system successful for PREX-I Required replacement of seed laser and fiber amplifier (spares used) Recent operation - 2kW locking, PPLN crystal temperature decrease Improved laser polarization measurement (0.2%) will improve error bar for CREX & PREX https://logbooks.jlab.org/entry/3434872 Photon Detector Electron Detector (CMU) Integrating photon analysis successful for PREX-I Normalization precision ~0.5% GSO calorimeter, low energy photons used for calibration need PMT for GSO operational at 150uA at 2GeV and 70uA at 1GeV highly linear base, linearity diagnostics are critical - LED Synchrotron radiation non-issue for 1-2GeV beam electron arm difficult to use at 1 GeV possible at 2 GeV Nice to have, but not strictly required for CREX

Summary Neutron skin studies at JLab will provide a crucial benchmark for the understanding of nuclear structure 208Pb: Uniform nuclear matter, DFT The density dependence of symmetry energy Tightly linked to neutron stars (radioactive nuclei, atomic PV, heavy ions) 48Ca: Extend studies over long lever arm of size and atomic number Bridge microscopic models to DFT Collaboration is active with beam tests and reviving experimental systems; aims to be ready to run whenever scheduled Still on target for readiness in Summer/Fall 2018 Note: PREX requires 1 GeV requires both HRS, so must run before SBS is on the floor Experimental Review (May 2016) confirmed radiation shielding concept & beam readiness. Septum conditions addressed by the design document. Second ERR in Spring 2017 Design Document completed and provides explicit requirements and rough design, for start of detailed design

backup

Gravity Waves and EOS Ben Lackey, Syracuse U PREX informs neutron star size vs. mass, which is critical to matter effects in GW Kent Paschke Photonuclear GRC August 11, 2016

Beyond JLab: MREX at MESA Ebeam = 155 MeV, ~25o Q2 = 0.0074 GeV2 APV to ~1.3% (9 ppb) ΔR = 0.03 fm Use P2 Apparatus - Solenoid provides high resolution with full azimuthal acceptance Kent Paschke Photonuclear GRC August 11, 2016

Nuclear Weak Form Factor Our picture of nuclei is the electric charge distribution: Neglecting relativistic recoil, the form-factor F(q) is the Fourier Transform of charge density Differential cross-section 208Pb q (fm)-1 1 2 3 As Q increases, nuclear size becomes important correction One can map the weak charge using parity-violation: unpolarized target σ ∝ |Mγ + Mweak|2 ~ |Mγ|2 + 2Mγ(Mweak)* +…

Density Dependence of the Symmetry Energy Energy penalty for breaking N=Z symmetry Slope of symmetry energy at saturation density Isovector properties are not well measured. Models informed mostly by measurements of properties sensitive to p+n. Neutron properties in stable medium and heavy nuclei have been mainly measured by using strongly interacting probes.

New information in a poorly measured sector Within a specific model, correlation of prediction vs. changes to one empirical input Good Isovector Indicators Poor Isovector Indicators (collective excitations, binding energies, etc.) Relatively well measured Jorge Piekarewicz

Neutron Star Mass vs. Radius Current PREX central value inconsistent with measured neutron star properties and existing models. Potential phase changes(!) would disrupt this argument

48Ca bridge between ab initio models and effective theory (DFT) 48Ca: finite size effects. Within reach of microscopic calculations coupled-cluster model of 48Ca: G. Hagen et al, Nature Physics 12, 186–190 (2016) 208Pb: uniform nuclear matter terrestrial laboratory to test n-star structure

MITP NSkins workshop A. Bauswein A.Steiner J. Piekarewicz

Activation Design aims to reduce de-installation dose Simulation to guide de-installation planning G4 Sim output is input to FLUKA to estimate activation in the pivot region (Work in progress, Lorenzo Zana) Design aims to reduce de-installation dose Shielded housing for collimator: hot bore is covered quick removal

Calcium Target Target vessel: Previous experiments have used a “bare” target in the vacuum No plan to contain the target in a separate vessel. Existing 48Ca target: Thinner than proposal. Tilt the foil to recover thickness Oxidized surface Contaminants are not a dangerous background - generous uncertainties ok. Scraping surface should sufficiently eliminate contaminants

How Neutrons Behave Reflection Transmission Iron high reflection, not efficient at stopping 25 cm concrete 50 cm concrete 100 cm concrete Concrete 30% reflection, 0.5m to block HD Polyethylene low reflection, 30cm blocks well

Neutron Stars Strong analogy to nuclei: Symmetry pressure pushes against gravity All neutron star radii between 10.4 and 12.9 km Suggests Rn(208Pb) < 0.2 fm 8 accreting neutron stars in globular clusters Steiner, Lattimer, Brown (2013) curves parametrized by density

Transverse Asymmetry

Weak Charge Distribution and Symmetry Energy ( R.J. Furnstahl ) The single measurement of Fn translates to a measurement of rn (via mean-field nuclear models) rn in 208Pb provides input to models to pin density dependence of symmetry energy

Neutron Skin at JLab APV ~ 0.6 ppm Q2 ~ 0.01 GeV2 Rate ~1.5 GHz 5o scattering angle PREX-I (2012): APV = 0.657 ± 0.060(stat) ± 0.014(sys) ppm rn - rp= 0.33+0.16-0.18 fm 0.5 mm 208Pb foil, 70 μA 5o scattering Pb~ 90% +/- 1% Analog integration of everything that hits the detector Very clean separation of elastic events by HRS optics no PID needed; detector sees only elastic events Summer 2017: PREX (3% APV, rn to 0.06 fm), CREX (2.5% APV, rn to 0.02 fm)

rn and Nuclear Structure 208Pb: uniform nuclear matter terrestrial laboratory to test n-star structure Important calibration point for FRIB studies ρ(n) corrections to atomic PV 208Pb rn crucial information for neutron star E.o.S. Mass/radius ratio, compare to observation cooling mechanisms (URCA or not) 48Ca: finite size effects. Within reach of microscopic calculations 208Pb 48Ca Chiral effective field theory, based on NN and NNN potentials DFT: effective theory for uniform nuclear matter Fundamental test of nuclear structure models

PVES has become a precision tool Interplay between probing hadron structure and electroweak physics Beyond Standard Model Searches Strange quark form factors Neutron skin of a heavy nucleus QCD structure of the nucleon photocathodes, polarimetry, high power cryotargets, nanometer beam stability, precision beam diagnostics, low noise electronics, radiation hard detectors sub-part per billion statistical reach and systematic control 0.5% normalization control For future program:

PREX-I: Sources of Radiation photon neutron Neutrons: Photoproduction in the hall dominates elastics from the thick, high-Z target dominantly GDR excitation in the collimator or beampipe soft neutron spectrum (<10 MeV) target area dump tunnel PREX-I: Even with W plug (~1.27o), significant spray into beam pipe and hall plots: origin of radiation striking “detector”, G4 MC

Relative Silicon Damage vs. Neutron Energy 100 keV 10 MeV low silicon damage high boron capture insignificant rate

Radiation PREX-I distributed significant power into the hall, damaged electronics 1 MeV neutron-equivalent integrated dose HRS PS platform PREX-2 (1 MeV-n / cm2 ) PREX-2 /PREX PREX-2 /HAPPEX-2 electrons 1.4E+10 11% 70% neutrons 7.6E+09 7% 94% total 2.1E+10 9% 83% Solution: Localize power in hall at collimator, and shield PREX-I PREX-II CREX Power in collimator (W/μA) 9.7 28.8 9.0 Power in hall (W/μA) 18.0 3.0 1.5 1 MeV-neq/cm2 is less than expected electronics damage threshold PREX-2 is an order of magnitude below PREX-I, equivalent to HAPPEX-2 high-energy n rate (SEU) is also down by 5-10x Site Boundary Dose JLab annual administrative limit: 10 mrem Extrapolation from PREX-I measurements is well within limits (~75% of annual dose) Further improvements are possible JLab admin limit: 10 mrem/year (mrem) (mrem) (mrem) RSAD simulation 3.8 9.2/2 8.4 Extrapolated from PREX-I data 1.7 5.4/2 4.4

208Pb & 48Ca: Complementary sensitivity Jorge Piekarewicz