370 likes | 527 Views
New Results. Klaus Rith University of Erlangen-Nürnberg. HERA Symposium 2011 July 5, 2011. Main HERMES research topics:. Origin of nucleon spin. Details of nucleon structure.
E N D
New Results Klaus Rith University of Erlangen-Nürnberg HERA Symposium 2011 July 5, 2011
Main HERMES research topics: Origin of nucleonspin Details of nucleonstructure Quark Properties: fractionalcharge spin-1/2 longitudinal momentumxP intrinsictransversemomentumpT spatialpositionr orbital angular momentum L 2
Atom (non-relativistic electrons in Coulomb potential) Rutherford Bohr, Schrödinger, .. Add angular momentum n, l, ml(r,,) 3
Nucleon (Relativisticquarks in colourfield) Inclusive DIS r Add angular and transverse momentum Wigner DF W(pT,r) TMDs GPDs Number density of quarks with longitudinal momentum fraction x (pT-dependence) (r-dependence) 4
Accessing the nucleon‘s structure inclusive DIS Integrated DFs 1D exclusive reactions semi-inclusive DIS 3D 3D GPDs TMDs theory after G. Schnell 5
TMDs Accessible in inclusive DIS Quarkpolarisation T N/q U L f1 h1 U Number Density Boer-Mulders Nucleon structure described by 8 leading-twist (+ many subleading) quark distributions containing information about quark orbital motion and spin-orbit effects Nucleonpolarisation g1 h1L L Helicity Mulders-Kotzinian h1 g1T T Transversity Sivers Worm-gear h1T Pretzelosity 6
Longitudinal double-spin asymmetry: g1, old P.R. D 75 (2007) 012007 g1(x) = ½eq2q(x) q = 0,330 ± 0,025 ± 0,011 ± 0,028 (from 1d) q(x) = q (x) – q (x) MS (exp) (theory) (evol.) q = q(x) dx = 0,12 ± 0,09 ± 0,14 (from 1p) EMC, P.L. B 206 (1988) 364 1 = g1(x) dx q q new g/g = 0,045 ± 0,034 ± 0,126 (high-pT hadrons) HERMES, JHEP 08 (2010) 130 Furthermore: 8
Unpolarised DIS cross section: F2 From global fit: HERMES relative normalisation ~2% for p and d and ~0.5% for the ratio Exploring perturbative to non-perturbative regime in an unmeasured x-Q2 region 0.006 < x < 0.9 0.1 GeV2 < Q2 < 20 GeV2 new JHEP 05 (2011) 126 Ratio d/p (F2d/F2p) Proton Deuteron New region covered by HERMES Good agreement with world data in the overlap region 9
Semi-inclusive Deep-Inelastic Scattering z = Eh/ eqeq FactorisationeNehX = DFNq FFqh DF(x,Q2): Parton Distribution Function – q(x,Q2) f1q(x,Q2), q(x,Q2) g1q(x,Q2), q(x,Q2) ) h1q(x,Q2), … FF(z,Q2): FragmentationFunction – D1qh(z,Q2), H1qh(z,Q2), … 11
Charged-hadron multiplicities I Proton-deuteron asymmetry UU f1q D1qh LO interpretation: Reflects different flavor content Correlated systematics cancel new new Disagreement Disagreement 12
Charged-hadron multiplicities II UU f1q D1qh Disentanglement of z and Ph dependences Access to intrinsic quark pT and fragmentation kT <Ph2> = z2<pT2> + <kT2> new new Ph Ph 13
Double-spin asymmetry A1h LL g1q D1qh Refined studies extending the work in Phys. Rev. D 71 (2005) 012003 With charge conjugation symmetry in fragmentation D1,qh+ = D1,qh- uv + dv new A1d h+-h- = (x) uv + dv new x x x x Integral over sum of valence distributions compatible with Sea contribution to nucleon spin small x x 14
Leading-twist TMDs Nucleon structure described in leading-twist by 8transverse-momentum dependent quark distributions (TMDs) HERMES has access to all of them through specific azimuthal modulations (, s) of the cross section thanks to the polarised beam and target d6 dx dy dz d ds dP2h cos2 sin2 sin(+s) sin(3-s) sin(-s) cos(-s) Chiral-oddDFs, needchiral-oddFF: H1,qh 15
Leading-twist TMDs Pioneering measurements by HERMES Quarkpolarisation T N/q U L Indication to be non-zero! Preliminary result h1 f1 U Number Density Boer-Mulders Consistent with zero PLB 562 (2003) 182 PRL 84 (2000) 4047 Nucleonpolarisation g1 h1L L Different from zero PRL 94 (2005) 012002 PLB 693 (2010) 11 Helicity Mulders-Kotzinian new h1 g1T T Transversity Sivers Worm-gear Consistent with zero Preliminary result h1T Pretzelosity Different from zero PRL 94 (2005) 012002 PRL 103 (2009) 152002 Small Preliminary result 16
sT pT pT sT Boer-Mulders DF h1,q UU cos2 h1,q H1,qh transversely polarised quarkswith pTin unpolarised nucleon h1 is chiral-odd and naive T-odd (like f1T) requires FSI/ISI new Opposite sign for + and -, larger magnitude for - h1,u and h1,d have same sign Large signal with same sign for K sea fragmentation important 17
Worm-gear DF g1T,q longitudinally polarised quarksin transversely polarised nucleon LT cos(-s) g1T,q D1qh new Related to parton orbital motion: requires interference between wave functions with OAM difference by 1 unit Slightly non-zero g1T,q = - h1L,q (supported by many models) 1 dy g1T,q x g1q(y) (Wandzura-Wilczek type approximation) y x 18
Generalised parton distributions Generalisation of Form Factors (moments of GPDs) and PDFs (forward limit) Correlated information about longitudinal momentum xp and transverse spatial position r Ji relation: Jq=1/2 + Lq= lim dx x [H(x,,t) + E(x,,t)] t0 Final state sensitive to different GPDs Spin-½ target: 4 chiral-even leading-twist quark GPDs H,H (E,E) conserve (flip) nucleon helicity Vector mesons (, , ) H, E Pseudoscalar mesons(,) H, E DVCS () H, E, H, E Access: exclusive processes ~ ~ ~ ~ ~ ~ 20
Hard exclusive 0-meson production I Photon SDMEs Meson SDMEs EPJC 62 (2009) 659 Helicity amplitudes FV= TV + UV EPJ C 71 (2011) 1609 new Helicity amplitudes are the fundamental quantities to be compared with theory They form a basis for the SDMEs Re-derived SDMEs consistent with published ones Enhanced sensitivity for polarised SDMEs 21
Hard exclusive 0-meson production II Hierarchy predicted by theory, confirmed by HERMES LL TT TL LT Large (as for H1) EPJ C 71 (2011) 1609 new expected small by GPD models tan(11) = Im(t11)/Re(t11) 1/Q dependence expected from pQCD Sizeable UPE 22
Deeply Virtual Compton Scattering & GPDs Theoretically cleanest way to access GPDs Interference between DVCS and Bethe-Heitler amplitude TDVCS << TBH@ HERMES AXY Access to GPD combinations through azimuthal asymmetries beam target polarisation HERMES: Complete set of asymmetries Bothbeam charges Both beam helicities Unpolarised H,D and nuclear targets Longitudinally polarised H and D targets Transversely polarised H target 23
DVCS asymmetries measured @ HERMES Beam chargeasymmetry GPD H H: PRL 87 (2001) 182001 PR D 75 (2007) 011103 JHEP 11 (2009) 083 D: Nucl. Phys. B 829 (2010) 1 Beam helicityasymmetry GPD H Transverse target-spinasymmetry GPD E H: JHEP 06 (2008) 066 Transverse double-spinasymmetry GPD E new H: arXiv:1106.2990 Longitudinal targetspinasymmetry GPD H ~ H: JHEP 06 (2010) 019 D: Nucl. Phys. B 842 (2011) 265 new Longitudinal double spinasymmetry GPD H ~ 24
DVCS: transverse target asymmetry AUT Sensitive to GPD E old JHEP 06 (2008) 066 Model: VGG with variation of Ju, while Jd=0 25
DVCS transverse double-spin asymmetry ALT Beam polarisation Target polarisation Beam charge arXiv:1106.2990 new Sensitive to both GPDs entering the Ji sum rule Consistentwithzero, cancellationsbetween E and H Sensitivityto Jusuppressedbykinematicfactors 26
DVCS with Recoil Detector Recoil Detector to tag exclusivity ep ep 1 T SC Solenoid Photon Detector Scintillating Fiber Tracker Silicon-Strip Detector Unpolarised H and D targets 27
Pure elastic DVCS new Indicationthatleadingamplitudefor pure elasticprocessisslightly larger thanforunresolvedsignal (elastic + associated) 29
DVCS with RD Helicityamplitudes F2 Worm-gear DF x(uV +dV) Boer-Mulders DF HadronMultiplicities 30
Pure elastic DVCS new Indicationthatleadingamplitudefor pure elasticprocessisslightly larger thanforunresolvedsignal (elastic + associated)
Double-spin asymmetry A1h LL g1q D1qh Refined studies extending the work in Phys. Rev. D 71 (2005) 012003 A1h(x,Ph) 2D - dependencies new Sensitive to differences in transverse momentum dependence of g1 and f1 No significant Ph dependence observed Ph 14
Transversity, Collins Amplitudes TransversityDF 2sin( + S)hUT h1q(x) H1q(z) CollinsFF arXiv:1006.4221 proton Both Collinsfragmentation function and transversity distribution function are sizeable Surprisingly large - asymmetry Possible source: large contribution (with opposite sign) from unfavored fragmentation, H1 ,disf - H1 ,fav
Extraction of Transversity Fit to HERMES (ep ->ehX), COMPASS (d ->hX), BELLE (e+e- ->h+h-X) data M.Anselmino et al., Nucl. Phys. Proc. Suppl. 191 (2009) 98 xu(x) xu(x) xd(x) xd(x)
Sivers Amplitudes for Pions SiversDF 2sin( - S)hUT f1T,q(x) D1q(z) PRL 103 (2009) 152002 First observationof non-zeroSiversDFinDIS proton Rise at low Ph, plateau at high Ph Clear rise with z Non-zero at low x Experimental evidence for orbital angular momentum Lqofquarks But: Quantitative contributionofLqtonucleonspinstillunclear
Sivers distribution Fit to HERMES (ep -> ehX) and COMPASS (d -> hX) data M.Anselmino et al., Phys. Rev. D79 (2009) 054010 Lattice Orbital angular momenta ofupand down quarkshaveoppositesign -xf 1T(x) Ld -L u 0.2 Ld + d/2 0 !?? x