Spin Experiments
Technological challenges and
selected physics highlights
Klaus Rith
University of Erlangen-Nürnberg & DESY
18th International Spin Physics Symposium, Charlottesville, VA – Oct. 6-10, 2008
Spin experiments
N. Bohr
W. Pauli
University of Lund, 31.5.1951
K. Rith
2
Spin experiments
N. Bohr
• are fascinating
W. Pauli
• enhance small signals
• provide new observables
• often generate surprises
K. Rith
2
Prerequisites - Technological achievements
strained GaAs polarized e--sources with pe  0.8
longitudinal polarization of 27.6 GeV stored e-/e+ beam
with pe  0.6 – spin rotators
internal storage cell polarized H, D, 3He gas targets
with no dilution (f=1), pTH,D  0.8
large polarized kryogenic targets (NH3, ND3, LiD)
polarized stored proton beams (presently up to 100 GeV)
with pp  0.6 – Siberian snakes
Fast and precise polarimeters
Challenges: - polarized positron sources (for ILC)
- polarized antiprotons with pB  0.2
K. Rith
J. Clarke
E. Steffens
3
Strained GaAs polarized e-sources
Early ´90s @SLAC: Overcome GaAs limit of pe= 0.5 by strained GaAs
K. Rith
4
AfLR from SLD
Highlight 1
SLD achievements:
Unique data on
Af
LR
=
g2L,f – g2R,f
g2L,f
+
g2R,f
=
2gV,f/gA,f
1 + g2V,f/g2A,f
gV,e-/gA,e- = 1 – 4 sin2 W,f
Statistically most precise single
determination of sin2W
sin2W(MZ) = 0.23098  0.00026
K. Rith
5
Strained GaAs polarized e-sources
Now ‚standard equipment‘ at all electron facilities:
(AMPS, BATES), ELSA, JLAB, MAMI, JPARC,……..
Example: SLAC E158 –Run 3
Sophisticated surface structure to
overcome charge limit and achieve
high polarisation (gradient-doped
strained superlattice)
K. Rith
Precise control of beam helicity
at source
6
E158 – PV in Møller-scattering
First PV experiment in e-L,R e-  e- eEnd of SLAC fixed target programme
Pe = 0.89  0.04, Ebeam = 45 & 48 GeV (spinrotation by extra  due to g-2)
E‘  13-24 GeV, Q2  0.026 GeV2
Challenge: Control of beam-helicity related systematic errors
E158 – PV in Møller-scattering
Highlight 2
Electroweak interference:

e-
e-L,R
+
Z0
e-L,R
e-
APV = - A(Q2,y)[1-4sin2Weff(Q)] = [-1.310.14 (stat)0.10 (sys)]10
-7
sin2Weff(Q)= 0.23970.00100.008
P.L. Anthony et al., PRL 95 (2005) 081601
Limit on SO(10) MZ’  1.0 TeV
6.2 
Limit on l.h. contact interaction
LLL ~ 7 or 16 TeV
Limit on lepton flavor
violating coupling ~ 0.01GF
K. Rith
8
PV in eN  eN
SAMPLE/BATES
A4/MAMI
HAPPEX/JLAB
G0/JLAB
Superconducting
Coils
Particle
Detectors
Electron Beam
LH2 Target
K. Rith
9
Highlight 3
Nucleon Strange Form Factors
Very small asymmetries: O(10-6)
D.S. Armstrong et al., PRL 95 (2005) 092001
K. Rith
J. Liu et al., Phys. Rev. C76 (2007) 025202
10
Highlight 4
Precise determination of GEn
Tool: double-polarization experiments
D(e,e‘n) with polarized D target
2D(e,e‘n)
with LD2 target and n recoil polarimeter
3He(e,e‘n) with polarized 3He target
E. Geis et al. (BLAST), PRL 101 (2008) 042501
K. Rith
11
Polarized muon beam – EMC, SMC, COMPASS
Myon beam: 100 - 200 GeV
    ,
-  - 
Advantage: ‘Natural‘ beam polarization:

Disadvantage: low beam currents – I  1 pA

J=0

Require high-mass polarized target, N  1024 nucleons/cm2
NH3 (f 3/17) , ND3 (f 6/20), LiD (f 4/8),……
f = fraction of polarizable nucleons
W. Meyer
K. Rith
12
Polarized target - COMPASS
World‘s largest kryogenic polarized target
180 mrad
K. Rith
13
COMPASS spectrometer
K. Rith
14
Polarized e beam in storage ring– LEP, HERA
Mechanism: Spin flip by emission of synchrotron radiation
 1 / 1010 - 1011 emissions (Sokolov-Ternov)
Degree of polarization: depends critically on machine
energy and magnet alignement
Example: LEP – mZ = 91.1874  0,0021 GeV
Beam energy dominant error of 1.7 MeV
Measurement by resonant depolarization of
transversely polarized beam
Influence of tide, TGV leaving airport station etc.
Longitudinal polarization: requires spin rotators
Polarization measurement: Compton backscattering of
circularly polarized laser light
K. Rith
15
Polarized e beam in HERA
e beam: E = 27.6 GeV, Ie < 50 mA
K. Rith
16
Polarized e beam in HERA
e beam: E = 27.6 GeV, Ie < 50 mA
2002-2007
1995-2000
1 spinrotator
K. Rith
3 spinrotators
After ´high –luminosity upgrade‘:
depolarization by beam-beam interactions
17
Total polarized CC cross section
Highlight 5
L
We-L
e-
R
W+
q
e +R
q
CCep (Pe) = (1  Pe)CCep (Pe=0)
R
No sign of r.h currents
We-R
e+
q
Chiral structure of SM confirmed
Convert to 90% CL on heavy WR:
mW,R > 208 GeV (H1)
K. Rith
18
Internal polarized storage-cell gas targets
K. Rith
19
Internal polarized storage-cell gas targets
Principle: Stern-Gerlach separation of HF-states + RF-transitions
Target polarisation: PT  0.85, Dilution factor: f=1
Ideal for storage rings: IUCF, COSY, BATES..
K. Rith
19
Highlight 6
Asymmetries in polarized pp scattering
Example: PINTEX at IUCF: p p  p p
Precise measurement of spin correlation parameters
F. Rathmann et al., Phys. Rev. C58 (1998) 658
before
K. Rith
after
Test and improvement of NN potential models
20
Polarized proton collider - RHIC
RHIC pC Polarimeters
Absolute Polarimeter (H jet)
PHOBOS
BRAHMS
Siberian Snakes
Siberian Snakes
PHENIX
STAR
Spin Rotators
(longitudinal polarization)
-
Pol. H Source
LINAC
Spin Rotators
(longitudinal polarization)
Partial Snake
BOOSTER
Helical Partial Siberian Snake
200 MeV Polarimeter
AGS
Strong AGS Snake
May 2006
Ep = 100 GeV
K. Rith
AGS pC Polarimeter
Masterpiece of
accelerator physics
T. Roser
May 2006
Ep = 100 GeV
21
Polarized proton collider - RHIC
PHOBOS
BRAHMS
PHENIX
-
Pol. H Source
LINAC
STAR
BOOSTER
AGS
K. Rith
22
Asymmetries in polarized pp collisions
K. Rith
23
Nucleon Spin
½ = ½  (Quark spins)
+ G
(Gluon spins)
+ Lq + Lg
(Orbital angular momenta)
EMC (1987):
K. Slifer
K. Rith
 = 0,12 0,090,14
24
Nucleon Spin - Tools
Polarised DIS
 or jet production in pp
l‘ =
l=
P1
x1P1
Hard Scattering
Process
x2P2
P2
 = E - E‘, Q2 = -q2 = -(l – l ‘)2
x = Q2/(2M) = fraction of nucleon‘s
longitudinal momentum carried by struck quark
g2
gq
q2
q(x) = quark number density
W -production
p
p
K. Rith
d
u
X
W+
X
Drell-Yan
( )
l
+
l
p
p
qv
qv
X
*
l+
X
l25
Asymmetries in polarized DIS
Highlight 7
beam target
A1(x) 
K. Rith

-

 + 
From W. Vogelsang
 qeq2q(x)
g1(x)


2
q eq q(x)
F1(x)
L.O.
26
g1(x), 
Highlight 7
g1(x) = ½ q zq2q(x)
HERMES
MS
q(x) = q (x) – q (x)
(exp)
(theory)
(evol.)
 = 0,330 ± 0,025 ± 0,011 ± 0,028
1
(from 1d)
q = q(x) dx
COMPASS
MS
(stat)
(evol.)
0
1

g1(x)
dx
 = 0,30 ± 0,01 ± 0,02 
(from
QCD
fit)
1 = NLO
0
   q q
K. Rith
27
Quark helicity distributions from SIDIS
Leading hadron originates with large probability from struck quark
Dqh(z):= Fragmentation function (FF)
z = Eh/
Measure
Measure hadron
hadron asymmetries
asymmetries A1h(x,z) =
z
z
2q(x)
q q
2
q q
Dqh(z)
q(x) Dqh(z)
Targets: H, D ; h = ±, K±, p
K. Rith
28
Quark helicity distributions
Highlight 8
HERMES, PRL 92 (2004) 012005, PRD 71 (2005) 012003
HERMES, PLB 666 (2008)
u > d ?
S = 0.037  0.019(stat.)  0.027(syst.)
s < 0 ?
K. Rith
x
(inclusive data and SU(3):
S = - 0.085  0.013(stat.) 0.012(syst.))
29
Highlight 8
Valence-quark helicity distributions
L.O.
L.O.
COMPASS, PLB 660 (2008) 458
u + d = 31N - ½1v + a8/12
Flavor asymmetric polarized sea (u = - d) favoured
K. Rith
30
Gluon helicity distribution
Highlight 9
Photon-gluon fusion
e‘
e
N
L.O. analyses
K. Rith
E. Rondio
31
G from pp 0 (jet) X
Highlight 9
A. Adare et al. (PHENIX); arXiv:0810.0694

De Florian et al.: hep-ph/0804.0422
g seems to be rather small !!
jets
NLO analysis (without data
from previous slide)
K. Rith
32
Spin budget
Origin of nucleon spin still unclear:
Where do the missing 65% come from?
X. Ji: ‚Dark Spin‘
Is there a substantial contribution of g and/or q
at very low x?
EIC
What is the contribution of orbital angular momenta
Lq, Lg ???
K. Rith
33
Transversity q(x)
For a complete description of momentum and spin distribution
of the nucleon at leading-twist: 3 distribution functions (DF)
Unpolarised DF
Helicity DF
Transversity DF
q(x)
q(x)
q(x)
well known
K. Rith
known
unknown before
HERMES, COMPASS
34
Azimuthal angular distributions
Amplitude has 2 components:
Transversity DF
2sin( + S)hUT ~q(x)H1q(z)
Collins Fragmentation Function
Unpolarised FF
2sin( - S)hUT~ f1Tq(x)D1q(z)
Sivers DF
K. Rith
U: unpol. e-beam
T: transv. pol. Target
z = Eh/
: conv. integral over pT and KT
(Requires non-vanishing orbital
angular momenta Lq of quarks)
N. Makins, M. Anselmino
35
Highlight 10
Collins amplitudes (proton)
2sin( + S)hUT ~ q(x)  H1q(z)
M. Diefenthaler @ DIS07, hep-ex 0707.0222
(also HERMES, P. R. L. 94 (2005) 012002)
First measurement of non-zero
Collins effect
Both Collins fragmentation
function and transversity
distribution function are sizeable
Surprisingly large - asymmetry
Possible source: large contribution
(with opposite sign) from
unfavored fragmentation,
i.e. u H1,disf  - H1,fav
K. Rith
36
Collins amplitudes (d +p )
Highlight 10
COMPASS, hep-ex/0802.2160
deuteron
K. Rith
Compatible with zero
information about d
Different from zero,
Compatible with HERMES
N. Makins, M. Anselmino
37
Sivers amplitudes (p)
Highlight 11
2sin( - S)hUT ~ f1Tq(x)  D1q(z)
M. Diefenthaler @ DIS07, hep-ex 0706.2242
(also HERMES, P. R. L. 94 (2005) 012002)
First observation of non-zero
Sivers distribution function in
DIS
Experimental evidence for
orbital angular momentum Lq
of quarks
But: Quantitative contribution of
Lq to nucleon spin still unclear
K. Rith
38
Sivers ampl. (d +p )
Highlight 11
COMPASS, hep-ex/0802.2160
deuteron
Compatible with zero
K. Rith
Still compatible with zero
39
Highlight 12
AN for identified hadrons in p p
E 704
Possible origins:
Collins FF, Sivers DF, Twist-3
Combinations of above
Important data for test of
theoretical models
K. Rith
40
q and f1T in p p Drell-Yan
Future
q is chiral-odd. Its measurement requires another
chiral-odd object!
Semi-Inclusive DIS: Collins fragmentation function.
(Requires additional input e.g. from BELLE)
p p Drell-Yan: Combination of qv and qv
E. Steffens
QCD prediction: f1T(x)SIDIS = - f1T(x)DY
q  q
qq    l+ lanti-lensing

green quark
anti-green
anti-quark
K. Rith
anti-green remnant
green quark
anti-green remnant
41
Determination of Lq - GPDs
Tool: Generalised Parton Distributions
Formfactors:
Fouriertransform of e.g. a
radial charge distribution
K. Rith
PDFs:
Number density of quarks
with longitudinal
momentum fraction x
GPDs:
Generalised description in
2+ 1 dimensions
C. Weiss
42
Determination of Lq
Ji relation:
1

Jq =1/2 + Lq = lim dx x [H(x,,t) + E(x,,t)]
t  0
1
H(x,,t), E(x,,t): Generalised Parton Distributions (GPDs)
Access: exclusive processes
Q2
Final state sensitive to different
GPDs
t
K. Rith
Vector mesons (, , ) H, E
 
Pseudoscalar mesons(,) H, E
 
DVCS () H, E, H, E
E. Voutier
43
Calorimeter and
superconducting magnet
within CLAS torus
Hard exclusive processes - Lq
HERMES/Recoil Detector, (2006 – 2007)
dedicated calorimeter (424 PbWO4
crystals) detect photons from 5o
JLab/Hall A, (2004 – 2005)
CLAS12
HRS + PbF2
H(e,e’p), D(e,e’N)
K. Rith
44
Azimuthal asymmetries
DVCS: Beam-spin asymmetry
HERMES, PRL 87 (2001) 182001
CLAS, PRL 87 (2001) 182002
JLAB/HALL A, PRL 97 (2006) 262002
d4 - d4
CLAS, PRL 97 (2006) 072002
DVCS: Longitudinal
target-spin asymmetry
K. Rith
45
Highlight 13
Hard exclusive processes - Lq
DVCS: Beam charge asymmetry
HERMES, JHEP 0806 (2008) 66
Pioneer measurements
DVCS: Transv. target spin asymmetry (SA)
DVCS: Long. target SA
0: Transv. target SA
K. Rith
46
Highlight 13
Determination of Jq
First model dependent attempt:
HERMES, JHEP 06 (2008) 066;
JLAB/HALL A, PRL 99 (2007) 242501
Lattice: Ld  -L u  0.2 !??
K. Rith
47
Conclusion
Spin physics is exciting
We are looking forward to see
plenty of new results
at this conference
and from the next generation of
high precision spin experiments
K. Rith
48
Backups
Azimuthal angular distributions
C
O
L
L
I
N
S
S
I
V
E
R
S
Transverse quark spin + spin-dependent fragmentation
Azimuthal asymmetry ~ sin( +  S)h
h
q
q
h
Left-right distribution asymmetry (due to orbital
angular momentum) + final state interaction
Azimuthal asymmetry ~ sin( -  S)h
green quark
anti-green remnant
Highlight 10
Sivers distribution
Fits of HERMES (p) and COMPASS (d) data
by Anselmino et al., arXiv: 0805.2677 and 0807.0166
Highlight 11
Collins FF and Transversity
Fits of HERMES (p), COMPASS (d) and BELLE data
by Anselmino et al., arXiv:0807.0173
Highlight 11
Collins amplitudes
Fits of HERMES (p), COMPASS (d) and BELLE data
by Anselmino et al. (from A. Prokudin @ DIS2008)
Sivers amplitudes (p)
Highlight 10
2sin( - S)hUT ~ f1Tq(x)  D1q(z)
M. Diefenthaler @ DIS07, hep-ex 0706.2242
(also HERMES, P. R. L. 94 (2005) 012002)
First observation of non-zero
Sivers distribution function in
DIS
Experimental evidence for
orbital angular momentum Lq
of quarks
But: Quantitative contribution
of Lq to nucleon spin still
unclear
Precise determination of GEn