1. News on ZEUS Leading Baryon analyses Roberto Sacchi Università di Torino and INFN DIS2004 Workshop Slovakia, April 14-18, 2004 Introduction Study of the pion trajectory in  p interactions with LN Study of DIS events containing a LP Conclusions Legenda: LP = Leading Proton LN = Leading Neutron LB = Leading Baryon OPE = One Pion Exchange

2. p,n Introduction Events with LB are a large fraction of the HERA cross-section Production mechanism is still unclear Models: p,n , IR, IP Standard fragmentation LB from hadronization of p remenant MC models Virtual particle exchange LP : neutral iso-scalar iso-vector ( , IR, IP) LN : charged iso-vector (  +,  +,...) Vertex factorization LP, LN : also from p fragmentation in diffractive processes Lepton variables: Q 2, W, x, y LB variables: x L = E’ LB /E p t = (p-p’) 2

3. p beam window of acceptance  n <0.8 mrad  =0 o ZEUS forward detectors FNC: 10 I Pb-sci. sandwich  /E = 65%/  E e-scale accuracy  2% FNT: scintillator hodoscope at 1 I installed in 1998  X,Y = 0.23 cm    = 22  rad LPS: 6 stations with  strip detectors only stations S4-S6 analyzed sofar hit position resolution  30  m  X L < 1%  P T ~ few MeV momentum accuracy <1% For both, p T resolution is dominated by p T spread of p-beam (50-100 MeV) Z = 106m

4. ZEUS forward detectors: acceptance Limited by apertures and detector size (LPS) Integrating over the azimuth: LPS (s4-s6): p T range varies with x L ; for x L >0.6, p T 2 <0.5 GeV 2 acceptance  15 % FNC: restricted to  n <0.8 mrad p T range increases with x L ; for x L >0.6,  n < 0.8 mrad acceptance  25 % FNC coverage  n <0.8 mrad 10 -3 10 -2 10 -1

5. In the following: Study of the pion trajectory in  p interactions with LN (DESY-04-037, submitted to Phys. Lett. B, hep-ex/0404002) Study of DIS events containing a LP in 1997 data (EPS03 #544) Main Issues in LB analyses Measure x L, Pt spectra of LB; Compare to hadronization and particle exchange models Test validity of vertex factorization hypotesis with different reactions In LN production, test the validity of OPE  , F 2 

6. 2000 data, L = 9 pb -1 special LUMI + FNC trigger Kinematic range: Q 2 < 0.02 GeV 2 = 220 GeV 0.6 < x L < 0.925 0.05<|t|<0.425 GeV 2 (1-x L ) distributions are consistent with a power-law dependence dN/dxL  (1- x L ) a(t) x L distribution vs t in  p  nX reaction Aim: test the consistency of the OPE model Lumi FNC

7. Interpretation: the reggeized OPE model Pion flux: Total  p cross section: where   0.1 (IP) and   0.5 (IR) dominant term even at largest x L (s’ min  60 GeV) Neglecting the (s’) -  term, the OPE model predicts the power law dependence dN/dx L  (1-X L ) a(t) where: Pomeron intercept Pion slope ignored

8. Powers a(t) as a function of t Powers a(t) nicely fit to a line. Assuming a(t)=  IP (0) - 2  ’  ·t yields Consistent with: soft pomeron intercept  IP (0)  1.1 pion trajectory   (t) = t-m  2 Note: contribution of a  trajectory   (t) = 0.5 + t would lead to a(t) = 2.08 - 2·t Data further support OPE as the dominant process in  p  nX reaction

9. DIS events containing a leading proton 1997 data, L = 12.8 pb -1 4 X larger statistics ! Standard DIS selection + LPS track   pipe >0.4 mm   pot >0.2 mm  E+P Z >1655 GeV DA reconstruction used Kinematic range: Q 2 > 3 GeV 2 45 < W < 225 GeV x L > 0.56 p T 2 < 0.5 GeV 2 CAL LPS track containment (rejects beam halo)

10. DIS events containing a leading proton MC used: Ariadne+SCI (diff. and nondiff.DIS) (K + and  + contamination) Pythia (PHP) Reweight Ariadne to reproduce LP data distributions  flat x L (nondiff. part)  exponential p T 2 with b=7.0 GeV -2  diff/total in bins of x L good description Remaining background:  PHP and low Q 2 (11%)  residual beam halo (7% at x L >0.98)  K + and  + (max 8% at x L =0.56) Note: K + and  + contamination cross-checked with LPS+FNC coincidences

11. Cross section vs x L diffractive peak Constant bin widths chosen according to resolution (  x L =0.03) Syst. uncertainties added.  vary Vz, Ee’, E-p Z cuts  vary  pipe and  pot cuts  vary MC p T 2 slope  vary K + and  + fraction Effect normally within statistical error data precise ! Flat distribution up to the diffractive peak (expected from pp interactions) Note: p-dissociative diffraction included in low x L spectrum

12. Cross section vs p T 2 Variable bin widths chosen according to p T 2 resolution Line shown is just to guide the eye Decrease by an order of magnitude in the p T 2 range considered; data are not well described by exponential function with single slope steepness decreases at high p T 2

13. Cross section vs p T 2 and x L p T 2 range in each xL bin determined by LPS acceptance range; few bins with low or unstable acceptance excluded; fit to exponential function with single slope performed in each bin;

14. b-slopes vs x L Result consistent with 95 measurement; smaller slopes observed around x L =0.75 and x L =0.95; partially explained as the effect of the different p T 2 range in different x L bins. No clear evidence of x L dependence of the exponential fall-off with p T 2

15. Summary Lot of leading baryon pre-upgrade data are available; analyses still alive and making progresses LN (1-x L ) distributions in  p interactions measured at fixed t satisfy power law dN/dx L  (1-x L ) a(t) a(t) consistent with pion trajectory being exchanged supports OPE model LP DIS cross section measured with high statistics vs x L and p T 2 flat xL distribution below diffractive peak exponential fit with single slope inadequate falling p T 2 distribution almost independent of x L

16. Outlook Increase knowledge on xL, pT spectra, vertex factorisation by completing the analyses of the available data Extend the x L range (LPS) by adding the S1-S3 stations Theoretical input needed....