Exclusive Central Production in Proton-Proton Collisions:

1
Exclusive Central Production in Proton-Proton
Collisions from the ISR (Glueballs)
to the Tevatron (leptons, photons, charmonium)
to the LHC (Higgs, WW,
Z?, SUSY?)
Mike Albrow Fermilab
2
ISR Tevatron -- LHC 63 1960
14,000 GeV
p
p
p
p
G?
H?
p
p
p
p

• Introduction
• Diffractive Excitation of High Masses (Jets,W,Z)
  CDF/D0
• Central Exclusive Production
• LHC Study of Higgs through pHp, WW and ZZ,
  Exclusive Z?
• FP420 RD project proposing extensions to
  ATLAS CMS.

ISR
Tevatron
3
ISR Intersecting Storage Rings, started
1971 First colliding proton beams.
p
p (31 GeV/c)
p (31 GeV/c)
p (28 GeV/c)
Centre of Mass Energy 63 GeV
Centre of Mass Energy 7.4 GeV
Equivalent to beam of 2110 GeV fixed p target
Into the realm of cosmic rays!
First collisions ... no detectors installed! ...
put in 4 counters!
L
R
COLLISIONS!
t(L) t(R)
Experiment 101
Emulsions on a toy train set!
4
2009 LHC Large Hadron Collider
7 TeV 7000 GeV
7 TeV 7000 GeV
55?
ATLAS 2500 physicists! One of four
experiments ALICE, ATLAS, CMS, LHCb
5
Meanwhile, back at the ISR in 1972 ...
British USA Muon Detector
Intersection I-2
W?
p
p
Medium Angle Spectrometer
Wide Angle Spectrometer British-Scandinavian
Small Angle Spectrometer CERN-Holland- Lancaster-M
anchester
Nobody knew what to do with complete
multi-particle ( 10) final states. Study
inclusive particle production pp ?
anything.
Muon Detector Looking for W(3-4 GeV!) ...
missed Wide Angle Spectrometer co-discovered
high (quark scattering) Small Angle
Spectrometer discovered high mass (14 GeV)
diffraction
p x 0.95
M 14 GeV
6
Small Angle Spectrometer Forward proton spectra
Feynman scaling
Discovery of high-x, scaling peak
x gt 0.95
M
p
M up to about 1.6 GeV at AGS/PS
14 GeV at ISR 440
GeV at Tevatron 3100 GeV at
LHC
0.95
7
Forward Proton Spectrometers ISR ? LHC
ISR (1971)
Small Angle Spectrometer
Gas Cerenkovs and tracking
Magnets and Spark Chambers
p (scattered)
20cm x 20cm
p (beam)
2m
SEPTUM
30 GeV/c
15m, dp/p 0.5
LHC (2009 FP420 2011)
20 mm x 6 mm
120m of 8T dipoles
FP420
QUADS, SEPs
3-24 mm
ATLAS or CMS
8 m
7000 GeV/c
420m, dp/p 0.01
High Precision (5-10 µm) tracking
High Precision (10 ps) timing Cerenkovs
8
Central Diffractive Excitation
Theoretically (Regge theory) if
happens, so should
x gt 0.95
x gt 0.95
both protons coherently scattered
M up to about 3 GeV at ISR 63 GeV
100 GeV at Tevatron 1960
GeV 700 GeV at LHC
14,000 GeV
H,WW
p
p
Exchanged 4-momentum must have no electromagnetic
charge or strong charge (colour), spin
gt1 Central state Quantum Numbers restricted.
W/Z exchange allowed, but p would break up.
p
p
IP (pomeron)
Vacuum Excitation
9
Central Exclusive Production
pp ? p X p where X is a simple system
completely measured
At CERN ISR Glueball Search
p
p
At Tevatron LHC
through q-loops (box) colour bleaching (g)
W
W
10
ISR Axial Field Spectrometer (R807) First
sophisticated high-pT spectrometer in pp.
Forerunner of p-pbar collider experiments.
lt 1
Uranium-scintillator full-azimuth
calorimeter 37/sqrt(E) hadron showers
Axial Field Magnet (Helmholtz coils)
ET lt 11.5/63 GeV
ET
ET 35-40/63 GeV
Circularity (2D-sphericity)
Jets in hadron-hadron co-discovered with UA2,UA1
(1982 Paris)
Phys.Lett B128 (1983) 354
11
Low Mass Central Exclusive Production
pp ? p X p X fully measured
Search for Glueballs
ISR 63 GeV
p
p
nothing else
Axial Field Spectrometer (R807) Added very
forward drift chambers
(Cambridge, RAL, QMC)
U-Cal
Central drift chamber half
U-Cal
12
Central Exclusive Production (AFS)
3500 events/25 MeV
G(1710)??
Also
Structures not well understood beyond f(980). Not
studied at higher
Pity would be great for G-spectroscopy
elastic scattering on-line dip!
In CDF?
13
CDF Detector at Fermilab Tevatron
MiniPlugs
980 GeV pbar
980 GeV p
14
Central tracking Silicon strips Drift Chamber
720,000 strips, 25 µm with 50 µm readout
Drift chamber 96 layers ? 30,240 sense wires 40
µm gold-plated tungsten ADC and TDC each
end Resolution 150 µm/wire
Surrounded by lead/iron scintillator sandwich
calorimeter for energy measurement
15
Installed very forward Beam Shower Counters (BSC)
for rapidity gaps and scintillating fiber
trackers in Roman pots for pbar detection
Not at all to scale! Roman pot detectors 20mm x
20mm 55 m downstream. Not used for most of this
(acceptance). Beam Shower Counters BSC tight
around pipe. Full coverage 7.4 lt lt 7.4
16
Diffractive W and Z Production
p
quasi-elastic
W
W
D0
W produced but p stays intact CDF
0.87 - 0.05
D0 also sees diffractive W and Z all consistent
with 1 diffractive
Z
So making an H with both p staying intact still
surprising, but maybe less surprising!
17
Central Exclusive Production in CDF
pp ? p X p where X is a simple system
completely measured
X
CDF ee-, ??, µµ-, J/?, ?(2S), ?_c, ? . Z?
LHC Z, H, WW-, ..
JJ
18
Rapidity (Tevatron)
y -7.5
y 7.5
t1
t-channel exchanges
t2
3 possibilities
(?) O ggg
We have now seen all 3 in h-h!
19
Also in ee- and ep Also in ep Only
in hadron-hadron
Odderon can replace photon in ppbar, not in ep
We cannot detect p and pbar require all CDF in
noise, to eta 7.4
20
Exclusive Electron-Positron Production
Phys.Rev.Lett 98,112001(2007)

16 events
QED process collisions in pp
Monte Carlos LPAIR, GRAPE, STARLIGHT
21
Exclusive 2-Photon Production
Khoze, Martin and Ryskin, hep-ph/0111078,
Eur.Phys.J. C23 311 (2002) KMRStirling
hep-ph/0409037
Tevatron
H
36 fb
Claim factor 3 uncertainty Correlated to pHp
22
Exclusive Production
Phys.Rev.Lett. 99,242002 (2007)
3 candidates observed
Prediction V.A.Khoze et al. Eur. Phys. J C38, 475
(2005) (our cuts) (36 72 24) fb 0.8
1.6 0.5 events. Cannot yet claim discovery as
b/g study a posteriori, 2 events corresponds to
90 fb, agreeing with Khoze et al.
It means exclusive H must happen (if H exists)
and probably 5 fb within factor 3.
is higher in MSSM
If really ?
Have more data, being analysed
23
Central Exclusive Production
Why
interesting? Among other things Two-photon
production continuum (QED
Form Factor). Cross section very well known lt
0.5 (QED) so can calibrate LHC luminosity (?).
Can also come through photo-production of
Forward proton momenta are precisely known
calibrate momentum scale and resolution
of forward spectrometers for p p ? p H p at
LHC .
24
New Results Submitted to PRL
Region rich in physics. First observations in
(elastic) hadron-hadron
Calibrate forward spectrometers for p p ? p H
p at LHC. Calibrate theory (cross-section)
of p H p
1)
2)
3)
t
c
H
25
Total luminosity 1.48/fb (/- 6) Trigger 2
muons Forward gaps Number of events on tape
1.6 million
Fiducial box
)
Reject cosmic ray events (ToF, colinearity)
100 efficient Exclusivity Require all detectors
lt noise cuts except in and around muons.
500
Example Beam shower counters BSC-1 (8 PMTs)
26
402 events, final sample
27
Submitted to PRL
Fit 2 Gaussians QED continuum. Masses 3.09,
3.68 GeV PDG Widths 15.8,16.7
MeVresolution. QED generator x acceptance 3
amplitudes floating
28
x QED spectrum
Only normalization A floating
STARLIGHT LPAIR MCs Good description very low
pT
29
Photoproduction (or possible
odderon exchange) Kinematics well described
by STARLIGHT MC also Much broader
than
Non-inclusive b/g? Odderon component? Some chi_c?
30
Now allow photons EmEt spectrum with J/psi mass
cut

Empirical functional form

• 65 events above 80 MeV cut.
• 3 events below (estimated from fit)
• 4 background under J/psi
• 65 /- 8

MC also estimates only few of
under the cut (But CDFSIMulation
not reliable for such low ET)
31
Summary of Results
M 3-4 GeV/c2
Assumed
Suppressed by J_z0 rule
32
Some predictions for J/psi photoproduction
e.g. Schafer and Szczurek arXiv0705.2887
hep-ph
Machado,Goncalves 3.0 nb Motyka and Watt 3.4
- 0.4 nb Schafer Szczurek 2.8 nb Nystrand
2.70.6-0.2 nb Our result 3.92 - 0.62 nb
Take 3.0 - 0.3
We are consistent, so we can put a limit on
odderon exchange. If theory gets more precise,
our limit can change
33
Odderon Limits and ratios
In QCD but not yet observed.
Our limits on O-exchange are close to, and
constrain, theoretical predictions
34
Dimuons Upsilon Region
CDF Run II Preliminary
Trigger µµ- ?lt0.6 , pT(µ) gt 4 GeV/c
Inclusive
Search for/measurement of photo-production of Y,
Y (not before seen in hadron-hadron)
CDF Run II Preliminary
Y(1S)
Invariant Mass 0 associated tracks pT(µµ) lt 1.5
GeV/c
Y(2S)
Status analysis in progress. QED continuum
check Y cf HERA (we resolve states) Can we see
?
Y(3S)
35
A. Szczurek arXiv0811.2488
30 pb
36
Exclusive Upsilon(1S) candidate
Run/Event 204413/8549136
M 9.4 GeV
R-z, Muon hits
Plugs, Miniplugs, CLC, BSC empty
37
Search for Exclusive Z, and observation of high
mass lepton pairs.
Have 12 exclusive candidate events ee- and
M 40 91 GeV2 11 have dphi lt 0.02 rad
and good QED candidates. Cross section
right. 91 GeV2 M(Z) has larger dphi pT may
be non-exclusive b/g (?)
E not ET!
CLC and BSC empty in 8/12 events, others
p?p dissociation. Paper in draft.
CLC
BSC
38

• Exclusive Z production is possible in the SM
  but ?0.3 fb, a signal would be BSM.
• The signature is very similar to the pp?p ll p
  process relaxing the pT and Mll cuts we open up
  acceptance to this process.
• First select ?? and ee pairs, then require no
  extra tracks and the calorimeters and BSC to be
  empty.
• We observe 8 candidate pp?p ll p events, with 0
  passing the Z selection in 2fb-1

?(pp?pllp) 0.240.13-0.10 pb Mll gt 40 GeV,
?l lt 4
cf/ LPAIR 0.256 pb
?(Zexcl) lt 0.96 pb_at_ 95 C.L
39
Double Diffractive Di-Jets in CDF
JET
JET
GAP
JET
(p not seen)
JET
Almost exclusive di-jet, Two jets and nothing
else
Jet ltETgt spectra same in SD and DPE
40
Exclusive Dijets (2 central jets nothing)
CDF
J
p
p
inferred
detected
J
Apparent b-jet suppression as they become
exclusive ? (Theoretically ? 0 as Rjj ? 1, Jz0
rule) Greatly reduces QCD background
ExHuME MC with exclusive di-jets.
Cross section comparison not yet done
41
Exclusive DiJet cross section
Cross section agrees with ExHuME / 3 (inside
uncertainty)
42
FP420 Forward Protons 420m downstream of CMS
ATLAS
CMS Inner Vacuum Tank insertion
CMS


420 240m
240 420m
ATLAS
43
Central Exclusive Production of Higgs
Higgs has vacuum quantum numbers, vacuum has
Higgs field. So pp ? pHp is possible. Allowed
states Process is gg ? H through t-loop as
usual with another g-exchange to cancel color and
even leave ps in ground state. If measure ps
J gt 2 strongly suppressed at small p angle (t)
t
H
4-vectors
!
MGARostovtsev hep-ph/0009336
http//www.fp420.com
44
What is exclusive H cross section?
Calculation involves gg ? H (perturbative,
standard, NLO) Unintegrated gluon
densities Prob.(no other parton interaction)
(Gap survival) Proton form factor Prob.(no
gluon radiation ? no hadrons) Sudakov Suppression
H
3 fb (M(H)125 GeV) factor 3
uncertainty 30 fb-1 ? 100 Ae events (Ae
acceptance, efficiency) Other estimates
differ by large amounts! But exclusive chi_c
etc is a check.
Exclusive
Durham Gp Khoze, Martin, Ryskin,
Stirling hep-ph/0505240
45
Cross section for pp ? p SMH p at LHC, x
branching fractions
Small ( fb) but SB can be high. ExHuMe
verified by 2-photon, JJ lt 140
GeV bbar, gt 140 GeV WW()
FP420 Acceptance fn. Mass (a) 420420 (b)
420240
(a)
(b)
46
Simulations of SMH ? b-bbar signals
background Cox, Loebinger and Pilkington
arXiv0709.3035 (JHEP)

1. 300/fb 3 years at 1034, 420420, L1 trigger on
   jets, muons, 25 kHz
2. Same with no pile-up background very high
   resolution p-timing

(a)
(b)
SMH significance, 120 GeV SMH, vs L(E33) 3 years
with no pile-up b/g. JET mu trigs
... and if 420420 in L1 trigger
future upgrade in latency?
47
Very Forward Proton Detectors ( Momentum
Measurement)
FP420
f
fractional momentum loss
48
3D Si Tracking, Cerenkov Fast Timing
Resolution Rad hardness Edgelessness Speed,
S/N Availability Enthusiasts!
QUARTIC
6mm(y) x 24mm (x) covers distribution
8 layers 10um x-y pixels
MCP
GASTOF
QUARTIC
BPM
BPM
p
3 mm
BEAM
8 m
49
Measure distance of track from beam (5-10 um)
and slope (5-10 um over 10 m) ? fractional
momentum loss
Beam
Note detector 6mm(y) x 24mm (x)
covers distribution.
50
Fast Timing Counters GASTOFs and QUARTICs
Pile-Up background ps, JJ or WW from different
collisions
Counters with 10 ps timing resolution behind
tracking 10
ps 3 mm 1) Check both ps from same collision
(reduce background) 2) Get z(vertex) to match
with central track vertex 3) Tell what part of
bunches interacting protons were
(F-M-B) Solution Cerenkov light in gas or
quartz (fused silica) bars ? MCP-PMT
(Micro-Channel Plate PMT) (or SiPM? APD? ) GHz
streak camera ? (Futuristic idea under study)
51
Oct 07 test beam at CERN Hamburg pipe, 3D Si
tracking, GASTOF QUARTIC timing
52
Non-SM cases no Higgs? MSSM Higgses?

• No SMH? Can we exclude? Suppose measure 100
  exclusive in CMS.
• ( 0.1 fb-1 effective S.I.Lum) ? predict
  pSMHp to 20
• Suppose expect (say) 100 pHp events in 30
  fb-1, see lt 40. Conclusion?
• 2) No SMH or MSSM-Hs? WW physics becomes very
  interesting!

fsi
W
Preview of ILC/CLIC physics
3) In case of SUSY, Forward p-tagging can be
crucial! Cross section can be much higher than
SMH. Decays to enhanced. A(CP ve) highly
suppressed.
Kaidalov Khoze Martin Ryskin
hep-ph/0307064
53
J.Ellis, J.S.Lee and A.Pilaftsis, PRD71075007,
hep-ph/0502251
MSSM
Can have h, A, H close together in mass (few
GeV) Hard to resolve by inclusive
production. Exclusive advantages higher
production than SM, A highly suppressed Excellent
mass resolution could separate h and H
(unique) Excellent mass resolution could even
measure H widths (if few GeV)
H
A
h
Durham Group (KMRS)
54
What is SignalBackground? H(135-200)? WW()
WW
l
Durham Gp Khoze, Martin, Ryskin, Stirling
hep-ph/0505240
p
p
J
J
Can use 50 of WW (all but JJJJ)
!!
Unfortunately very few events (SM)
In WW/ZZ case, central trigger effective (420420
OK)
55
Determining Quantum Numbers of Central State (H?)
Is it J 0, CP ?

• In gg ? X only CP is allowed.
• (a CP ve A (MSSM) is highly suppressed)
• gg ? vector (J 1) forbidden, Landau-Yang
  theorem.
• J 0, 2 can be distinguished by angular
  distributions
• partial wave analysis. Can even see states hidden
  in overall M distribution!
• Of course this needs many events.

W
g
g
IP
IP
W-
Moments H(LM) of the cos( ) distributions ?
M(J0), M(J2).
e.g. ISR/R807 glueball search in
NPB264 (1986) 154
56
Summary
Any states with vacuum quantum numbers and strong
or electromagnetic couplings can be produced at
LHC by Central Exclusive Production This
includes Higgs boson(s), W-pairs, lepton and
photon pairs. Cross section pp ? pSMHp known to
factor 3 ( 1-10 fb) If protons well measured,
can get mass of central state to 2 GeV per
event, Quantum numbers (J, CP) and couplings to
gg. need both 220m and 420m detectors. RD on SB
can be good excellent in BSM scenarios. For
good acceptance/resolution FP420 ( 240) tiny but
v.high precision tracking, timing, BPM Best
particle spectrometer ever, using part of LHC We
have proposed this as extensions to CMS and
ATLAS for installation in 2011-2012 shutdown
(maybe!)
JJ
CDF
57
BACK-UPs
58
Central Exclusive Production in Different Machines
In collisions (through LEP energies ? I
L C) ? hadrons,
and at high (ILC) energy WW ? WW, WW ? Z,H WZ
? W In ep collisions (HERA) gamma-IP ? vector
mesons ( too but
buried) In (ISR ? Tevatron
and LHC) IP IP ? hadrons (can be single
hadron), q-loop ?Higgs, -IP ? vector mesons
(.. ,Y, Z(allowed but tiny)?) ?
ll- ( too but buried) In AA (RHIC, LHC)
mainly (E-fields) -IP
and IPIP
New in CDF
59
photoproduction (or possible
odderon exchange) Kinematics well described
by STARLIGHT MC
Again much broader
than
60
Events with EM shower
Good fits to kinematics with only
, if EM shower
Confirms assignment
61
Note H1, ZEUS could not resolve states, but BR
are different
Will pre-select regions QED-low 8 - 9.2 Y(1)
9.2 9.7 Y(2) 9.7 - 10.2 Y(3) 10.2
10.5 QED-high 10.5 12.0 QED-veryhigh gt 12.0
62
What is H SignalBackground? (not pile-up)
J
3
4
J
Trigger is issue Probably need asymmetric 220m
420m and Eventual trigger upgrade??
63
MSSM SUSY cross section x BR ? b-bbar larger
than SM Heinemeyer et al., arXiv0708.3052
Ratio (MSSM/SM) h ? WW() vs M(A) and tan(beta).
M(h) 120 GeV.
Excluded by LEP Z ? Zh/H
64
What is SignalBackground? H(135-200) (not
pile-up)
Examples
NO OTHER TRACKS ON VERTEX! (But only 4.6
of WW)
e
Prob. ZERO BACKGROUND in ZZ! 4 events ?
DISCOVERY!
65
Exclusiveness brings many rewards. H ? Jet Jet
case
(fractional momentum loss)
JETS
420m just too far for L1 trigger. 420 220
Jet info.
66
BSM The White Pomeron
Alan White (ANL)
BFKL Pomeron 2 reggeized gluons / ladder White
Pomeron 1 reggeized gluon sea wee gs
Asymptotic freedom ? 16 color triplet qs Only 6
known (duscbt) But (!) 1 color sextet Q counts 5
times, so udcstb UD
works!
composite higgs

• Anomalous (quasi-diffractive) production of WW,
  ZZ
• (not WZ) production at LHC ( M(DPE_at_LHC) lt 700
  GeV).

Dramatic effects at LHC, especially in pp ? p
WW/ZZ p
pp ? p Z p