1

Pierre MarageUniv. Libre de Bruxelles

On behalf of the H1 and ZEUS collaborations

Diffraction at HERA

CIPANP 2006Puerto-Rico29/5-4/6/2006

2

Content

1. Introduction : diffraction and QCD at HERA

2. Inclusive diffraction measurements

3. Diffractive parton distribution functions (dpdf)

4. Semi-inclusive diffraction and comparison with dpdf jets charm

5. Exclusive production vector mesons: rho photoproduction - BFKL DVCS -GPD

3

Hadronic interactions

elastic scattering diffractive dissociation

(differential absorption

of hadronic components)

high energy soft pomeron - unitarity !

Diffraction

(0) 1( ) (0) 1.08IIPIPh h s

HERAsome 10% of cross section in electroproduction !

4

Diffraction at HERA and QCD

Hard scales : Q2, large pt (jets), large quark mass (charm), large t

QCD analyses

challenge : quark and gluon understanding of diffraction, unitarity mechanisms, etc.

pomeron structure

diffractive parton densities

factorisation theorem

colour dipole

dipole – proton cross section

lowest order = 2 gluon exchange

small dipoles in presence of hard scale

5

xxDiffractive variables

W

2 2

2 2

2 2

2

2 2

2

(1 )

( ')

XIP L

IPX

Y

Q q

Q Mx x

Q W

Qx x

Q M

t p p

M

5 2 2(5) 2

2 42

(1 ) ( , , , , )2

DDem

IP YrIP Y

d yy x Q M t

dx d dQ dM dt Q

Inclusive cross section

6

Inclusive diffraction : 3 measurement methods

1. Proton spectrometers -> t measurement

2. Large rapidity gap NC and CC diffraction

3. Non-exponential fall off

of ln MX2 distribution :

2

2ln

lnX

X

B MdND C e

d M

diffr. non-diffr.

e p X p

7

MX

Q2

ZEUS

H1

W

IPx

Cross sections measurements

8

Factorisation and pdf’s

Collinear factorisation cf. total ep cross section

2 2( ) ( , , , ) ˆ ( , ) ( , )

for fixed , ( )

Di i IP ei

IP Y

d ep eXY f x Q x t d x Q i q g

x t M

Vertex factorisation

2 2/( , , , ) ( , ) ( , ) ( / )D IP

i IP IP p IP i IPf x Q x t f x t f Q x x

usefull hypothesis (Regge inspired)

9

Scaling violations

large gluon content !

10

Parton distribution function parameterisations

• Fit A

H1

11

Vertex factorisation :

(4) 22 ( , , , )D

IPF x Q t

12

~1/Q²at large

=> HigherTwist behaviour

J. Bartels at al., Eur. Phys. J. C7, 443 (1999).

2-gluon exchange model

Colour dipole

q q g (L or T)q q

13

14

hard scales (Q2, pT ) pdf evolution

(perturbative QCD)

sensitivity to gluon distr. fct.

Dijets in diffractive DIS : factorisation ?

( var.)IPZ

15

Joint (inclusive + jets) pdf fits

Low sensitivity of fits to inclusive cross section to gluon pdf, especially at large zIP

use jets in combined fits

H1 Fit A H1 Fit B

Combined fit close to Fit B; dijets provide significant constraints on high gluons !

16

xDiffractive charm production (photoprod.)

ZEUS photoproduction

hard scale

pdf evolution (gluons)

17

xDiffractive charm production (DIS)

H1 DIS

Agreement with NLO calc. within the (large) errors.

18

Factorisation

Factorisation tested with jets, charm

But still large errors.

NB low sensitivity of inclusive pdf’s to gluons at high joint fits important

Extrapolation from HERA

CDF data

Factorisation breaking

by underlying interactions

in hadron-hadron (Tevatron, LHC)

and for resolved photons

gap survival probability

19

Vector meson production

photoproduction

soft diffraction (ex. photoprod.)

- low cross section increase with energy

(soft pomeron)

- fast fall of t distribution

hard diffraction (J/, large Q2)

- stronger energy dependence

large gluon density at small x

- flatter t distribution

small dipole size

B tde

dt

hard

soft

20

Large |t| photoproduction and BFKL

t

helicity flips best described by BFKL-like approach

21

DVCS

Generalised Parton Distributions

rather ``hard´´ energy dependence

rather ``soft´´ t dependence

22

Measurement precision of inclusive cross section

Extraction of diffractive pdf’s

Factorisation tests / joint fits with jets, charm

(especially for gluon at large )

Exclusive processes : vector mesons, DVCS, etc.

QCD understanding of diffraction

Conclusions

(3) (4)2 22 2( , , ) ( , , , )D D

IP IPF x Q F x Q t

23

Extra slides

24

25

comparison with LO MCZEUS ZEUS

26

Diffractive jet prod.: sensitivity to pdf’s

H1 Fit A H1 Fit B

Low sensitivity of fits to inclusive cross section to gluon pdf, especially at large zIP

use jets in combined fits

H1

27

comparison with LO MC

28

Combined fit of F2D

and dijet data :

2/ndf (total) = 196 / 217

2/ndf (dijets) = 27 / 36

2/ndf (F2D) = 169 / 190

Combined fit ~ close to Fit B; dijets

provideimportant

constraints on high z

gluon !

Both datasets well described by the combined fit.

Joint (inclusive + jets) pdf fits

29

30

Vector meson production

31

B values large forDVCS (compatible with

a soft hypothesis)!

hard W dependence=>

Many ideas discussed

32

Leading neutron production

33