18 October 2010 Frascati WorkshopSemi-Inclusive Deep-Inelastic Scatteringand Kaons from CLAS6K. GriffioenCollege of William & [email protected] onProbing Strangeness in Hard ProcessesLaboratori Nazionale di FrascatiFrascati, Italy18 October 2010218 October 2010 Frascati Workshopeg1-dvcs with CLAS eg1-dvcs ran for 120 days in2009 on NH3, ND3 and C targets We have an order of magnitudemore DIS data than in eg1 The Inner Calorimeter greatlyincreased π0 detection318 October 2010 Frascati WorkshopLongitudinally Polarized TargetNH3• Dynamic nuclear polarization of NH3 and ND3• Polarizations of 70-80% for p and 25-40% for d• Luminosity 1035 cm-2s-1418 October 2010 Frascati WorkshopDouble Spin AsymmetriesInclusive Asymmetry Part BInclusive Asymmetry Part C•Asymmetry•Target Polarization•Asymmetry•Target PolarizationND3NH3NH3CC Deuteron polarization (red)between 25 and 40% Inclusive asymmetries, integratedover all kinematic variables (blue),are proportional to PbPt Proton polarization (red) between70 and 90% Inclusive asymmetries, integratedover all kinematic variables (blue),are proportional to PbPt518 October 2010 Frascati WorkshopInner Calorimeter The IC measuresangles below 18-22odepending on thedistance from the target Most of eg1-dvcs hadIC coverage below 18o618 October 2010 Frascati WorkshopExperimental DetailsRunPeriodBeamEnergyfiles (2min0.4M trig.)NominalIC locationPart ANH3/C5.8874.73015490-57.5 cmPart BNH3/C5.95433506-67.5 cmPart CND3/C5.75214659-67.5 cm718 October 2010 Frascati WorkshopSIDIS Kinematic Coverage in eg1-dvcsBecause the IC cuts offelectrons below 18O andthe target above 45O, wemeasure DIS in a smallstripe in the x-Q2 planeThere is nearly fullcoverage in z and pTKaon kinematics aresimilarThe pT distributions for π+and π- are different from π0because most π0 s comefrom the IC (small angles)whereas the chargedpions are obscured by theIC.pTπ+π+π-π0zxQ2818 October 2010 Frascati WorkshopKaon Identification in eg1bNegative kaons populate theregion to the left of the straightnegative pion band Positive kaons are clearly seenbetween the straight pion bandand the curved proton bandpκπ+π-pΔt918 October 2010 Frascati WorkshopKaon Identification in eg1-dvcsThere may be somenegative kaons present tothe left of the vertical bandof pions, but the backgroundis very high Some positive kaons are seenbetween the vertical band ofpions and the corner band forprotons. However, TOF does notprovide a clean kaon samplepκπ+π-Δtp1018 October 2010 Frascati WorkshopLund MC Predictions for κ+ Lund Monte Carlofor eg1b which isat a similar beamenergy Negative kaonsare at least a factorof three worsecompared tonegative pions Since the ICobscures smallangles in eg1-dvcs,one can expectthis ratio to beworse at low x ineg1-dvcs A reasonable ratio estimate of 5% means, allapart from systematics, that the statistical errorbars will be 4-5 times bigger for positive kaonsthan for positive pions. If systematics are undercontrol, this could still produce a significantresult.1118 October 2010 Frascati WorkshopPolarized SIDIS Kinematics1218 October 2010 Frascati WorkshopDifferential Cross Section Bacchetta, et al., JHEP 2(2007)093Only includes terms important for longitudinal asymmetries AUL, ALU and ALLsince A is a polarized difference over an unpolarized sum1318 October 2010 Frascati WorkshopTMDs and Fragmentation The observables are the structure functions such as FsinφUL, not the TMDdistribution or fragmentation functions. Therefore pT and kT are only indirectlydetermined through the following convolutions. Four-fold differential data in x, z,Q2 and PT are essential to test the factorization built in below. eg1-dvcs can dothis.1418 October 2010 Frascati WorkshopAsymmetries from eg1b Existing data fromthe eg1bexperiment whichran in the year2000. Here thekaon sample wastoo small toeffectively analyze.eg1-dvcs has anorder of magnitudemore data1518 October 2010 Frascati WorkshopA1 ≈ g1/F1 for eg1-dvcs•eg1-dvcs data (25%) of total•PT dependence → µ0 ≠ µ2•Also for π+ and π0•Would like this for kaons, butthese are hard to find in eg1-dvcs1618 October 2010 Frascati WorkshopA1 ≈ g1/F1 for eg1-dvcs•eg1-dvcs data (25%) of total•PT dependence → µ0 ≠ µ2•Also for π+ and π0•Would like this for kaons, butthese are hard to find in eg1-dvcs1718 October 2010 Frascati WorkshopAUL for eg1-dvcs The target spin asymmetries as a function of φ have both sinφ and sin2φcomponents. AsinφUL (higher twist) is significant for π+ Asin2φUL (leading twist) is small suggesting, like eg1b and HERMES, that theCollins favored and unfavored fragmentation functions are nearly equal andopposite. The analysis is currently being refined1818 October 2010 Frascati WorkshopConclusions• ALL:– Eg1-dvcs measures ALL → g1/F1 better than ever before, allowing us tolook for differences in the transverse momentum distributions forpolarized vs. unpolarized quarks.– We can measure for π+, π-, π0, and possibly κ+ to explore the flavor-dependence of these quantities.– Kaon data will be hard to come by using only CLAS6 (without a RICHdetector). Careful background checks on what we call π or κ will need tobe done in our analysis.• AUL:– AsinφUL is non-zero, indicating that higher-twist is important– Asin2φUL is small suggesting that the Collins favored and unfavoredfragmentation functions are nearly equal and opposite.• Outlook:– EG1-dvcs analysis will improve over the next year– This greatly inreases the eg1b statistics. CLAS12 will do much