Upgrade of the Polarized 3He Target System

Patricia Solvignon

Hall C Users meetingJanuary 24-25, 2013

Polarized 3He System: the basics

Spin-exchange optical pumping (SEOP) is a two-step process:

1) Rb vapor is polarized by optical pumping with circularly polarized light

2) Rb e- polarization is transferred to 3He nucleus by spin-exchange interaction

from J. Singh’s Ph.D thesis

Boosting the spin exchange rates

5.5. NUMERICAL SIMULATIONS 436

300 350 400 450 500 550 600

-310

-210

-110

1

300 350 400 450 500 550 600

Temperature (K)

ηEf

ficie

ncy

of S

pin

Exch

ange

Cs

Rb

K

Na

360 380 400 420 440 460 480

-210

-110

360 380 400 420 440 460 480

Temperature (K)

ηEf

ficie

ncy

of S

pin

Exch

ange

Rb

K

Figure 5.10: Spin Exchange Efficiency for Na, K, Rb, & Cs. The spin ex-change efficiency is defined as the fraction of angular momentum that success-fully transferred from the alkali atom to the 3He nucleus: 1/⌘A = 1 + (ksd/kse) +(k0sd/kse)([A]/[3He]) + (k00sd/kse)([N2]/[3He]). At low temperatures and alkali den-sities, it is approximately kse/(kse + ksd). At high temperatures, it approacheskse[3He]/(k0sd[A]). The blue lines are calculated from the values in Tab. (??). Thered data points are from [69]. In the left figure, the solid blue lines representan alkali density range of (1014 to 1015)/cm3. In the right figure, the red line isa parameterization of the data given by ⌘K(T) = 0.756 � 0.00109T and ⌘Rb(T) =0.337� 0.00102T(1� 0.0007T).

parameter Na K Rb Cs units

�D1 (air) 589.6 769.9 794.8 894.3 nmspin orbit splitting 5.2E+02 1.7E+03 7.1E+03 1.7E+04 GHznaive XD2

↵ (2.0 nm) 1.1E+01 3.7E�02 1.6E�03 2.6E�04naive XD2

↵ (0.2 nm) 9.0E�02 5.7E�03 3.6E�04 7.8E�05k�1

se /(1015/cm3) 5.86 5.39 4.12 2.51 hrs[A]3 for ��1

se = 3 hrs 1.95 1.80 1.37 0.84 1015/cm3

T3 for [A]3 634.3 537.6 483.5 453.2 K�A

sd/�Ase 5.5E�05 8.1E�05 5.7E�04 1.1E�02

kHe/kse at T3 0.31 5.2 33 155Xani 0.16 0.057 0.046 0.026

Table 5.8: Basic Alkali Parameters for SEOP.

SPIN 2008 - UVa - Charlottesville, Va - October 6, 2008

Rb K

Rb K 3He

K 3He

Rb 3He

Rb 3He

Hybrid Spin ExchangeA-A spin exchange is very likely and efficient!

A-He spin exchange is very unlikely and inefficient...

Rb-K hybrid cell: addition of potassium in the spin exchange process

optimal value of [K]/[Rb]: 4-10

Narrowed laser vs. broad laser

5.5. NUMERICAL SIMULATIONS 433

40 60 80 100 120

0.4

0.5

0.6

0.7

0.8

0.9

1.0

40 60 80 100 120

Power (W)

vo

lum

e a

vera

ge

d a

lka

li p

ola

rizati

on

9

277

12

6

6

5

4

4

3

= 3 hrsse

-1! = 2.0 nm , "#

= 3 hrsse

-1! = 0.2 nm , "#

Figure 5.7: Optimal Alkali-Hybrid Ratio vs. Laser Power.

0 5 10 15 20

0

20

40

60

80

100

0 5 10 15 20

depth into cell (cm)

Rb

po

lari

za

tio

n (

%)

223 cm

75 W ; laser intensity = 3 hrs

1 = se!

= 2.0 nm"#D = 0

$gradual roll off

= 0.2 nm"#D = 0

sharp drop% = 2.0 nm"#

D = 6$gradual roll off

= 0.2 nm"#D = 6

sharp drop%

Figure 5.8: Alkali Polarization vs. Depth into Pumping Chamber.

from J. Singh’s Ph.D thesis

The combination of hybrid cells and narrowband pumping significantly reduce our sensitivity to small imperfections, regardless of their origin.

The 6 GeV era Hall A polarized 3He System

Results: ‣ Spin-up time shorten from about 24 hrs to 5 hrs‣ In-beam polarization increases from about 40% to 60-65%

Approved experiments in Halls A & CExp eriment Density Length Pol. Current Lumi Polarimetry A1n-A: 23 days, A-, BigBite, thin window/collimation, BB field shield/compensation proposed 10 amg 60 cm 65% 30 uA 3x1036 3%

should 10 amg 40 cm 55% 30 uA 2x1036 3% accept able 10 amg 40 cm 55% 15 uA 1x1036 3%

GENII-A: 50 days, A-, BigBite/SBS, thin window/coll., BB/SBS field shield/comp. proposed 10 amg 60 cm 65% 60 uA 6x1036 3% acceptable 5/8 FOM

SIDIS-A: 64 days, A-, BigBite/SBS, vertical polarization and fast spin flip (2 min) proposed 10 amg 60 cm 65% 40 uA 4x1036 3% acceptable 5/8 FOM

d2n-C: 29 days, A-, HMS/SHMS proposed 10 amg 60 cm 55% 30 uA 3x1036 3% accept able 10 amg 40 cm 55% 15 uA 1x1036 3%

A1n-C: 36 days, A, HMS/SHMS proposed 10 amg 60 cm 60% 60 uA 6x 1036 3% accept able 10 amg 60 cm 60% 40 uA 4x1036 3%

Note: Another two of the Hall A experiments, which need the SoLID spectrometer and only require the target performance of already achieved from 6 GeV performance, not considered here

First upgrade

1) Using cells with convection flow

2) Using a single pumping chamber with 3.5” diameter sphere

3) Shielding pumping chambers from radiation damage

4) Using pulsed NMR, calibrated with EPR and water NMR

5) Necessary to measure EPR calibration constant K0 to higher temperature range covering the hybrid cell operation temperature (user responsibility)

6) Metal end-windows are desirable (optional for 30 uA, must for higher current, will be an user R&D project)

7) Keep existing Helmholtz coils configuration

X. Zheng, August 2010, Jefferson Lab PAC36 10/21

Improvements on the Polarized 3He Target (2010)

Air

Vacuum

Metal end-caps,the rest of the cell can remain in glass

First step: upgrade to a 40cm target with 30uA beam current and reaching ~60% polarization

Second step: upgrade to meet the need for A1n in Hall C and GENII in Hall A depends on funds available.

Second upgrade (if funds available)

In the QCD DSE approach, it is the diquark that causes

such a different behavior for the u and d quarks.

All target components are mounted in an aluminum frame that can be

translated up, down and sideways.

Design concept for the Hall A A1n target

Wednesday, January 4, 2012

In the QCD DSE approach, it is the diquark that causes

such a different behavior for the u and d quarks.

All target components are mounted in an aluminum frame that can be

translated up, down and sideways.

Design concept for the Hall A A1n target

Wednesday, January 4, 2012

1) Metallic target chamber necessary to handle 60 μA

2) Two-pumping-chamber cell allows more laser power

3) More laser power is needed for the increase in gas volume: from 2-3 STP liters to 6-7 STP liters

5) New oven, new support structure, new optics

6) 60 cm long target chamber and change in the Helmholtz coils configuration

Convection Cell New convection style cell (single pumping chamber) • “Protovec-I”  tested  at  UVa, is at Jlab now • 3D measurement of the cell, transferred into CAD model • Made customized mount and oven bottom piece • Start to do test on this convection cell

12/11/2012 Polarized 𝐻𝑒 Target status

Hall A Collaboration Meeting Jie Liu <jie@jlab.org> 12

12 GeV System Progress

Polarimetry Adiabatic fast passage (AFP) - NMR • AFP-NMR will be only used for calibration • If use hybrid glass/metal cells • Work for both He and water Electron Paramagnetic Resonance (EPR) • EPR will still work Pulse NMR • Send a pulse tuned at Larmor Frequency • Spin presses tipping from holding field 𝜃 = 𝛾𝐻 𝑡 • Spin components orthogonal to holding field, have free-induction-decay, Amplitude∝ 𝑀 sin(𝜃 )

• If we use hybrid glass/metal cells, RF is heavily attenuated, AFP-NMR will not be suitable for measurement on target chamber. • Pulse NMR can work on transfer tube.

𝜃 𝑀      𝐻

𝑀

12/11/2012 Polarized 𝐻𝑒 Target status

Hall A Collaboration Meeting Jie Liu <jie@jlab.org> 9

12 GeV System Progress

Theory: S∝ 𝑀 sin(𝜃 )𝑒 ⁄ sin(wt)

Lasers New QPC laser • The Comet laser (25W, 0.2nm width) production was discontinued • Possible laser procurement with QPC. • Already got one QPC laser (Hall C, 25W, 0.3nm width) and doing test now

12/11/2012 Polarized 𝐻𝑒 Target status

Hall A Collaboration Meeting Jie Liu <jie@jlab.org> 13

12 GeV System Progress

Hall A polarized 3He system on Hall C pivot

Hall A polarized 3He system on Hall C pivot

UNIVERSITY of NEW HAMPSHIRE

LLCmedXe

June 17,2010

An ex situ high pressure target !  Continuous SEOP within a large volume vessel !  Compress polarized 3He by 20:1 pressure ratio and deliver to titanium

target cell at 1 scfm !  Requires compression ratio ~20, immersion in magnetic field, rubidium-

free gas leaving polarizer, <3% polarization loss !  Throttle polarized gas back into the polarizer, de Laval nozzle

15 Bar 238 Bar

Recirculating at 1.0 scfm 1 cm x 40 cm titanium target cell

Requires two ports, entrance and exit 4

Next generation polarized 3He target

Summary• Improvements in the polarized target should be able to reach about 2-3 times higher FOM than the “Transversity” target:

➡ Hall C is planning to use the same target system as Hall A A1n.➡ Every steps are being coordinated between Hall A and Hall C.

• More substantial system/cell upgrades will be needed to gain another factor of 2 or 3 in FOM

• Next generation polarized 3He target may be a Xemed-type target which may allow another order of magnitude improvement in luminosity:

➡ Access to exclusive experiments and low cross-sections kinematical regions.➡ Positive response from PAC39: “The PAC encourages the group to pursue all these technical efforts to provide later a new generation of high luminosity polarized 3He target from which several other experiments can benefit.”

First running of polarized 3He experiment is expected to be in 2016 in Hall A.

Extra slides

Projected precision for A1n

Figure 3: Projected data (red solid circles) for measurements of asymmetries An1 in the

large-x region using a 11 GeV beam and HMS and SHMS in Hall C. Both DIS andresonance data are shown. The error bars show the expected statistical error and theerror bands around the horizontal axis illustrate the expected systematic uncertainties.The horizontal axis shows the SU(6) prediction that An

1 = 0. The curves illustrate(from top to bottom in the region x > 0.6): 1) the LSS(BBS) parametrization at Q2 =4 (GeV/c)2 (light blue curve) [5]; 2) the BBS parameterization at Q2 = 4 (GeV/c)2 (darkblue curve) [6]; 3) the chiral soliton prediction by Weigel et al. (magenta curve abovethe yellow shaded band) [20, 21, 22]; 4) the RCQM (yellow shaded band) [7]; 5) theLSS2001 parameterization (black curve) [10, 23]; and 6) another chiral soliton predictionby Wakamatsu (magenta curve below horizontal axis) [24, 25]. Data shown are fromSLAC E142 [26] and E154 [27, 28], HERMES [29], and JLab 6 GeV E99-117 [1].

x

A 1n

SLAC E142SLAC E154HERMESJLab Hall A E99117

HMS+SHMS, DIS, 684 hoursHMS+SHMS, RES, 48 hours

0

0.5

1

0 0.25 0.5 0.75 1

16

Table 5: Projected statistical and systematic uncertainties for resonance data at di!erentx and Q2. Resonance data will be taken at a scattering angle of 12.5! (same as the lowQ2 DIS data). The DIS fit for A1 was used in the systematic uncertainty study.

x "An1 (stat.) "An

1 (syst.) "An1 (total)

0.55 0.0072 0.0145 0.01620.60 0.0061 0.0169 0.01800.65 0.0074 0.0197 0.02100.71 0.0095 0.0242 0.02600.77 0.0138 0.0323 0.03520.83 0.0302 0.0530 0.06100.89 0.0593 0.1003 0.1165

Figure 4: Statistical and systematic uncertainties for the proposed An1 measurement.

Only DIS data are shown here. Systematic uncertainties shown here are mostly dueto nuclear corrections in the data analysis. Uncertainties due to instrumentation andbackgrounds (such as the detector’s PID performance which determines the uncertaintiesfrom pion and pair production background) are not shown because they are expected tobe negligible compared to the statistical error for the proposed measurements.

xΔA 1

n

-0.2

0

0.2

0.4

0 0.25 0.5 0.75 1

4.2 Expected Results for Neutron hg1(x)

Figure 5 shows the expected uncertainty on An1 at di!erent Q2 settings. This Q2 leverage

will allow a study of the Q2-dependence of An1 , and further allow extraction of the higher-

17

Projected precision for d2n

JLab PAC36 3August, 2010

Lines of integration for d2

n at

Q2 = 3, 4, 5, 6 GeV

2

Motivation & Updated KinematicsMotivation & Updated Kinematics

Updated kinematics:• 4 settings/arm• 125 hours/setting

• Directly measure the Q2 dependence of the neutron d2

n

(Q2)

at Q2 ≈ 3, 4, 5, 6 GeV

2 with the new polarized 3He target.

· The SHMS is ideally suited to this task!

• Doubles number of precision data points for g2

n(x, Q

2) in DIS region.

· Q2 evolution of g

2

n over (0.23 < x < 0.85)

• d2 is a clean probe of quark-gluon

correlations / higher twist effects

• Connected to the color Lorentz

force acting on the struck quark

(Burkardt)

· same underlying physics as in

SIDIS k⊥ studies

• Investigate the present

discrepancy between data and

theories.

· Theory calcs consistent but

have wrong sign, wrong value.

JLab PAC36 4August, 2010

Projected results for E12-06-121Projected results for E12-06-121

• Q2 evolution of d2

n in a

region where models are

thought to be accurate.

• Direct overlap with 6 GeV

Hall A measurement.

Projected g2

n points are

vertically offset from zero

along lines that reflect

different (roughly) constant

Q2 values from 2.5—7 GeV2.