Argonne National Laboratory PR12-11-112 Precision measurement of the isospin dependence in the 2N and 3N short range correlation region Spokespersons: J. Arrington, D. B. Day, D. Higinbotham, P. Solvignon John Arrington Argonne National Laboratory PAC38 August 25, 2011

Short range correlations (SRC) Short-range NN interaction generates high momenta Universal mechanism for all nuclei Similar shape for k>kFermi Cioffi Degli Atti et al, PRC53, 1689 (1996) 2

Short range correlations (SRC) Short-range NN interaction generates high momenta Universal mechanism for all nuclei Similar shape for k>kFermi Mean-field region: collective behavior, strongly A-dependent Cioffi Degli Atti et al, PRC53, 1689 (1996) 3 3

Short range correlations (SRC) Short-range NN interaction generates high momenta Universal mechanism for all nuclei Similar shape for k>kFermi High-momentum region: short-range physics, largely A-independent. Dominated by 2N-SRC tail A(e’,e) at x>1: For momentum sufficiently above kFermi, mean field contributions small; dominated by 2N-SRCs Cioffi Degli Atti et al, PRC53, 1689 (1996) 4 4

Short range correlations (SRC) Pmin vs. x, Q2 k>kFermi A(e’,e) at x>1: For momentum sufficiently above kFermi, mean field contributions small; dominated by 2N-SRCs Cioffi Degli Atti et al, PRC53, 1689 (1996) 5 5

SRC evidence at JLab Simple SRC Model: K. Egiyan et al, PRL96, 082501 (2006) Simple SRC Model: 1N, 2N, 3N contributions dominate at x≤1,2,3 2N, 3N configurations “at rest” (total ppair = 0) Isospin independent N. Fomin, et al, arXiv:1107.358 (2011) Experimental observations: Clear evidence for 2N-SRC at x>1.5 Suggestion of 3N-SRC plateau Isospin dependence ? 6 6

Tensor force dominance R. Subedi et al, Science 320, 1476(2008) Tensor force dominance Simple SRC Model: 1N, 2N, 3N contributions dominate at x≤1,2,3 2N, 3N configurations “at rest” (total ppair = 0) Isospin independent R. Schiavilla, R. Wiringa, S. Pieper and J. Carlson, PRL98, 132501 (2007) pp np Scaled deuteron momentum distribution From A(p,p’pn) and 12C(e,e’pN): 90% of observed pN pairs are pn; tensor force  isosinglet dominance R(pp/pn) = 0.0560.018 R(T=1/T=0) = 208% 7

Main physics goals: E12-11-112 Isospin-dependence Triple coincidence: Demonstrated isoscalar dominance Good precision: R(T=1/T=0) = 208% Large FSI: Dominates the high-Pm 12C(e,e’p) events used to look for SRCs Inclusive 3H/3He: Improved precision: Extract R(T=1/T=0) to 3.8% FSI much smaller (inclusive) and expected to cancel in ratio No Pm dependence 3N SRCs structure (momentum-sharing and isospin) Q2 dependence can test models of the momentum sharing 3He/3H ratio very sensitive to isospin-momentum correlations Improved A-dependence in light and heavy nuclei Average of 3H, 3He  A=3 “isoscalar” nucleus Determine isospin dependence  improved correction for N>Z nuclei, extrapolation to nuclear matter Absolute cross sections (and ratios) for 2H, 3H, 3He Test calculations of FSI for simple, well-understood nuclei 8 8

Isospin structure of 2N-SRCs 3He/3H is simple/straightforward case: Simple estimates for 2N-SRC Isospin independent Full n-p dominance (no T=1) 40% difference between full isosinglet dominance and isospin independent Few body calculations [M. Sargisan, Wiringa/Peiper (GFMC)] predict n-p dominance, but with sizeable contribution from T=1 pairs Goal is to measure 3He/3H ratio in 2N-SRC region with 1.5% precision  Extract R(T=1/T=0) with uncertainty of 3.8% Extract R(T=1/T=0) with factor of two improvement over previous triple-coincidence, smaller FSI 9

3N-SRC configurations p3 = p1 + p2 p1 = p2 = p3 Different configurations possible for 3N-SRCs, for example: “Linear configuration” p3 = p1 + p2 extremely large momentum “Star configuration” p1 = p2 = p3 Inclusive measurement can help to differentiate between these configurations and test isospin structure 10 10

Light-cone fraction: a2n Relativistic: x  a [light cone momentum fraction] a = A(Ei-pi,z)/(EA-pA,z) = A(Eilab-pi,zlab)/MA Cannot reconstruct in inclusive scattering, but for scattering from 2N-SRC, a ≈ a2N For 3N-SRC, equivalent a3N depends on the details of the momentum sharing in the SRC Ratios vs x Combine with E08-014 data on 3He Quality of scaling in a3N tests picture of 3N-SRC momentum sharing Ratios vs a2N 11 11

R≠1.4 implies isospin dependence AND non-symmetric momentum sharing Momentum-isospin correlations “Linear configuration” p3 = p1 + p2 extremely large momentum “Star configuration” p1 = p2 = p3 (a) yields R(3He/3H) ≈ 3.0 if nucleon #3 is always the doubly-occurring nucleon (a) yields R(3He/3H) ≈ 0.3 if nucleon #3 is always the singly-occurring nucleon (a) yields R(3He/3H) ≈ 1.4 if configuration is isospin-independent, as does (b) R≠1.4 implies isospin dependence AND non-symmetric momentum sharing 12 12

(3He+3H)/2H yields `isoscalar’ A=3 result A-dependence of 2N-SRCs N. Fomin, et al, arXiv:1107.358 (2011) Red points: assume isosinglet dominance Cyan points: assume isospin independence Reality is somewhere in between (closer to isosinglet dominance) Large difference for 3He and 9Be, heavy nuclei  modifies A-dependence for light nuclei and extrapolation to nuclear matter (3He+3H)/2H yields `isoscalar’ A=3 result 3He/3H yields improved extraction of (T=1)/(T=0)  better correction for all non-isoscalar nuclei 13 13

Extra impact of improved A-dependence 2N-SRC ratio EMC effect 3He 4He 9Be C Cu,Au Extra impact of improved A-dependence Correlation between EMC effect and SRCs suggests that they may both be driven by short-distance physics (density) or high-momentum physics (virtuality) L. Weinstein, et al., PRL 106:052301,2011 Improved measurements of A-dependence of SRCs can test different interpretations; Requires better quantification of isosinglet “dominance”

Quasielastic data Worlds 3H QE data: Q2 ≤ 0.9 GeV2 Propose: 0.6-1.0 GeV2 1.4,1.7 GeV2 2.2-3.0 GeV2 In PWIA, 3He/3H with 1.5% uncertainty corresponds to 3% on GMn * Limited to Q2≤1 GeV2, where QE peak has minimal inelastic contribution * This is the region with ~8% discrepancy between the Anklin, Kubon data and the CLAS ratio and Hall A polarized 3He extractions Nuclear effects expected to be small, largely cancel in ratio For CLAS D(e,e’n)/D(e,e’p) extraction of GMn, nuclear corrections on the ratio were <0.1% over the full Q2 range

PR12-11-112: Experimental setup Standard Hall A HRS configuration Gas Cerenkov + Calorimeter PID 1H, 2H, 3H, 3He room temperature cells, 20 atmospheres (10 for 3H) Empty cell for window subtraction (check software cut on windows, subtract residual contribution) Carbon foils for optics 16

Tritium target: updated design Four identical cells: 1H, 2H, 3He at 20 atmospheres, 3H at 10 atm. Operate at room temperature Length: 30cm, Diameter: 1.25cm 18 mil windows and walls Technical review last year Prototype requested and funded Goal is to be ready by 2014

Kinematic coverage QE 2N 3N Beam current: 25 mA, unpolarized Raster interlock Beam energy: 4.4, (2.2) GeV 17.5 Days 4.4 GeV [main production] 1.5 days 2.2 GeV [checkout+QE] QE 2N 3N 18

Kinematic coverage QE 2N 3N Left HRS running (380 hours) Beam current: 25 mA, unpolarized Raster interlock Beam energy: 4.4, (2.2) GeV 17.5 Days 4.4 GeV [main production] 1.5 days 2.2 GeV [checkout+QE] QE 2N 3N Left HRS running (380 hours) 19

Kinematic coverage QE 2N 3N Right HRS running (parasitic) Beam current: 25 mA, unpolarized Raster interlock Beam energy: 4.4, (2.2) GeV 17.5 Days 4.4 GeV [main production] 1.5 days 2.2 GeV [checkout+QE] QE 2N 3N Right HRS running (parasitic) Worlds 3H QE data: Q2 ≤ 0.9 GeV2 Left HRS running (380 hours) 20 20

Summary Study of isospin dependence of 2N-SRC from 3H/3He from inclusive scattering: will complement the results of 2N knockout experiments Greater precision Smaller final-state interactions First look at isospin-momentum structure in 3N-SRC region Quasi-elastic data on 3H and 3He for Q2-values of 0.6-3.0 (GeV/c)2 Beam time requested: 19 days including data taking, calibrations, background studies and configuration changes Hall A spectrometers in standard configuration, same 3H target system needed for the approved MARATHON experiment (E12-10-103) 21

Backup Slides 22

Relationship to E08-014 (“x>2” on light nuclei) E08-014: Finished data taking in April/May Extract Q2 dependence for 3He/2H, 4He/3He ratios Precisely determine scaling regions for 2N, 3N-SRCs in light nuclei Initial Q2 dependence of ratios vs. different 3N scaling variables (a3N) Isospin (40Ca/48Ca) Intrinsic sensitivity is roughly half of 3H/3He Total uncertainty in R(T=1/T=0) about 12% (40Ca target problems) PR12-11-112 Difference of 3H, 3He to study (T=1)/(T=0) Project 3.8% uncertainty compared to 8% from (e,e’pN), 12% from 40/48Ca Yield improved isoscalar correction for light and heavy nuclei Look for asymmetry in momentum-isospin distribution for 3N SRCs Average of 3H, 3He yields “isoscalar” A=3 nucleus Improve extraction of A-dependence of SRCs in light nuclei Comparison to state-of-the-art calculations without isospin structure uncertainties 23 23

Final-state interactions No indication of Q2-dependent FSI * A/D ratios in plateau region No indication of FSI within 2N-SRC * nD(k) from y-scaling vs. calculations * high-k data consistent with calculations * Scatter of NN potentials limits constraints SLAC [Fe] CLAS [Fe] Hall C [Fe] – E89-008 Hall C [Cu] – E02-019 24 24

Dummy and 3He targets (E08-014, kin 6.5)

SRC evidence at SLAC, JLab Hall B, Hall C Frankfurt, Strikman, Day, Sargsian, PRC48, 2451 (1993) Naïve SRC Model: 1N, 2N, 3N contributions dominate at x≤1,2,3 2N, 3N configurations “at rest” Isospin independent Experimental observations: Evidence of 2N-SRC at x>1.5 Suggestion of 3N-SRC plateau Isospin dependence ? 26 26

QE peak at low Q^2

Onset of 2N SRC region is not A-independent 2H 197Au

Acceptance correction Systematics δσ/σ δR/R (normalization) (pt-to-pt) Acceptance correction 2.0% 0.3% 1.0% Radiative correction 3.0% 0.4% Tracking efficiency - 0.2% Trigger efficiency 0.5% 0.1% PID efficiency 1.5% Target thickness 1.0%* Charge measurement Energy measurement 0.05% COMBINED UNCERTAINTY 4.6% 1.1% 1.2% * Expect to know relative target thickness to <2%, and be able to normalize to DIS at the 1% level (based on MARATHON proposal) 29

EMC slope vs SRC • EMC slopes are fit from 0.325<x<0.7; preliminary results from SRC, EMC (heavy nuclei) • Some kind of break down for heavy nuclei, SRC factors are more or less same but EMC slope keep on increasing with A Without deuteron “constraint” point, fit does not have expected behavior for deuteron Rescaling assuming n-p only in SRC, all N-N pairs in EMC fixes linearity, deuteron limit. Suggests short-distance over high-momentum?