Marco Bozzo

1
TOTEMTotal cross section measurement at LHC

• Marco Bozzo
• INFN Genova and Universita di Genova (Italy)
• On behalf of the Collaboration
• Brunel, CERN, Cracow, Dresden, Genoa, Grenoble,
  Helsinki, Prague.
• http//totem.web.cern.ch/Totem/

2

• stot

• Dispersion relation fit (logs)g , g2.2?0.3
• Current models predictions 100-130mb
• Aim of TOTEM 1 accuracy
• Absolute calibration of Luminosity

3

• Luminosity Independent Method

4
Large-t Elastic scattering
Impact picture and Regge models
5
Inelastic detectors (T1 and T2) in CMS
Elastic detectors (RPs) in the straight section
6
Elastic scattering and High ß optics

• Cross section is large ? can be measured with
  relatively Low Luminosity 1028 and short
  runs.
• Low divergence beams require a special High- ß
  optics.
• For ß1100m beam angular spread at IP is 0.67
  µrad and beam size is 0.74mm (rms).
• Crossing angle of 150 µrad is not sufficient to
  avoid parasitic collisions ? need zero crossing
  angle
• 36 bunches 1 out of 100 (time separation 2.5
  µsec or 740m)
• to measure precisely the angle of scattered
  particles very close to the beam detectors
  placed at p/2 phase advance.
• This results in a parallel to point focus
  optics where displacement correlates directly to
  the scattering angle
• 2 such points exist in warm section
• measure possible in Horizontal or Vertical plane.

7
b 1100 m
x Leff q vx0
Leff(m)
v
RP1 147m
RP3215m
efficient as close as possible to the
beam Acceptance pots at same position
and different inefficient edge of
detectors Acceptance track seen in either pot
compared with single pot acceptance
8
15/GeV2
9
Roman Pot
Window thickness 0.1 mm First tests in 2002
10
Roman Pot detector 50 µ pitch Si strip detector
CMS Hybrid
Thermal analysis
11
Detectors for elastic scattering

• Much effort to operate
• At the largest beta,
• closest to the beam (10 s-beam 0.5 mm),
• with reduced n. of bunches and emittance
• Delicate part is the design of the detector
  region near the beam and the choice of the
  detector
• The window, the mechanical tolerances, the
  inefficient region should be minimum
• Standard silicon 0.5 mm guard ring (typ.)
• Silicon strip at LN2 temperature. Can work with
  efficiency up to the edge when edge is cut (and
  properly treated).
• idea suggested by RD39 and NA60, tested on beam
  by TOTEM in 2002.
• 3D detectors edgeless and warm (S. Parker,
  C.Kenney), test of 1cm x 1cm detector with beam
  in summer 2003.

12
Cold Silicon can be edgeless - tests 2002
Cut diode gives Pair of edgeless detectors (RD39
- Z.Li / BNL)
Edge of cut Silicon Strip Detector (50mm pitch)
Microstrip Silicon Sensor (NA60 - Z.Li / BNL)
13
Cold Silicon Tests Results
Edgeless Silicon Diodes (RD39)
Silicon Microstrip Detectors
100 V bias 500 000 Events
with tracks 471 ? 50 ?m from Survey 435 ?
35 ?m
measured
Sensitive edge coincides with survey within one
strip 0 30 mm
14
Edgeless and warm Si 3D Detectors
S. Parker, C. Kenney 1995
New 3D TECHNOLOGY E-field line contained
by edge (p) electrode SENSITIVE Within 10 mm of
EDGE
Side view

• EDGE SENSITIVITY lt10 mm
• COLLECTION PATHS 50 mm
• SPATIAL RESOLUTION 10-15 mm
• DEPLETION VOLTAGE lt 10 V
• SPEED AT RT 3.5 ns
• SENSITIVE AREA 3X3 cm2
• SIGNAL AMPLITUDE 24 000 e before Irradiation
• SIGNAL AMPLITUDE 15 000 e- at 1015n/cm2

50 mm pitch
Top view
Pictures of processed structures Brunel, Hawaii,
Stanford 2003
15
3D response to particles
Brunel, CERN, Hawaii, Stanford
Pulse height

• Small bias 5-8 V
• Speed 1.5, 3.5 ns
• Radiation hardness

MIP
Ileak 0.45 nA (average) 200 mm Ileak 0.26 nA
(average) 100 mm C 0.2 pF per electrode
MIP
1x1015 p/cm2 20oC
16
Miniaturization of Roman pots

• To approach beams in the horizontal direction,
  where space between the pipes is small, the
  detector movement might become smaller if put in
  secondary vacuum together with detectors. We
  are studying how to approcah beam horizontally.

P beam
17
Roman Pot Trigger

• Silicon detector pad for trigger 500 pads of
  1mm2
• Trigger chip under study
• Fast-or of 128 channels for fast trigger
  extraction
• Multiplicity processor to reject multi-hit events
• Timing precision about 1 nsec
• useful to identify by TOF the longitudinal
  position of interaction

18
Elastic detectors

• 100-150 small size (4x4cm) silicon detectors
  necessary for entire experiment
• No extreme requirements on radiation hardness .
• Read out with CMS APV hybrid allows also easy
  integration in CMS DAQ for common runs.
• Choice on final technology will be taken during
  2004.

19
The measurement of the inelastic rate

• Needs fully inclusive minimum bias trigger with
  known efficiency
• Identification of beam-beam events from
  background by reconstruction of interaction
  vertex
• Two main event topologies
• NSD 85 of inelastic cross section
• SD 15
• Trigger combinations
• Double arm for NSD (clean, low background)
• Single arm for SD (beam gas int. looks like SD!)
• requiring signal in roman pot on opposite side
  can help in removing background
• Simple detector telescopes in the forward region

20
30 of s tot
EVENT TOPOLOGIES
21
TOTEM inelastic rate measurement
Measurement of the inelastic rate with an overall
error of the order of 2 detector spans 4 units
of pseudorapidity (on both sides)
The value at coordinates ?min, ?max is the
event loss of the TOTEM inelastic telescopes
85 of events
15 of events
1-2 loss
10-15 loss
3-4 global loss
22
T1 TELESCOPE (?3.1 to 4.7 on each side)

• Trigger by RPC
• Two double gap chambers
• Pads with projective geometry
• Time resolution 1ns
• Vertex reconstruction by CSC
• 5 planes with 3 coordinates point
• 2 cathode planes anode wire plane

-CANTILEVERED STRUCTURE FIXED ONLY ON OUTERMOST
CMS YOKE RING -TWO PLATES (Light blue) ARE THE
ONLY INTERFACE BETWEEN CMS AND TOTEM T1 -RAILS
INSIDE PROFILE TO OBTAIN MAXIMUM ACCEPTANCE
23
T1 telescope
HALF DETECTOR SUB-ASSEMBLY
DETECTOR IN TWO HALVES, TO ALLOW INSTALLATION
WHEN THE VACUUM CHAMBER IS ALREADY IN
PLACE ALUMINIUM FRAME HOLDS EACH CSCs, RPCs PLANE
AND PROVIDES SUPPORT AND COOLING FOR ELECTRONICS,
SERVICES AND LINK TO THE RAILS DETECTORS OVERLAP,
PLANES ARE ROTATED IN f WITH RESPECT TO EACH
OTHER.
24
T1 Telescope

• T1 is installed in the CMS end cap
• ?3.1 to 4.7 on both sides
• Detector on platform (cables and service racks)
  tested before installation.
• Cables and services arrive via HF cable ducts
• Installation time of the order of 1-2 days. Last
  operation in CMS closing schedule

T1 installation platform
25

• In 2002 we have also built and tested full scale
  prototypes of CSC and RPC.
• CSC
• Built using similar technology as CMS
• (invaluable advice and help from CMS FNAL Muon
  group)
• 1.0m x 0.8m prototype. R/O with AD16 from CMS
  (anodes) and Gassiplex from ALICE for cathode
  strips.
• Plan to use adapted CMS CSC R/O chain
• RPC
• 1.0m x 1.0m quadrant and 30cm x 30cm 3 m apart
• Square pads 4x4 cm.
• Test original pad structure and measure
  resolution time for trigger

26
CSC prototype tests 2002
Spatial resolution rms0.6 mm
Residuals (mm)
27
RPC prototype tests 2002
Time resolution 0.66 ns Rate capability up to
at least 1 kHz/cm2
28
T2 telescope
HF (CMS)
Geometrical Considerations Distance from
IP 13570 mm T2 inner radius 25 mm 8 mm
33mm   Vacuum chamber inner radius 25 mm Outer
radius 135mm Wall thickness, clearances and
tolerances 8 mm  Length 400 mm h range
5.32 lt h lt 6.71
29
T2 telescope
Space for services
Lightweight Structure
? 470 mm
Bellow
Thermal Insulation
Weight estimate 30 Kg
400 mm
30
T2 telescope
Cooling pipe
one plane 10 double sided detectors mounted on a
lightweight disk.
31
Luminosity

• High ß runs provide stot and Luminosity
• study various trigger combinations from T1 and
  T2
• The T1 (LeftRight) will have low background and
  high efficiency, sees a large fraction of the
  cross section.
• define an effective monitor cross section and
  obtain absolute calibration of rate
• Calibrate CMS monitors (tracker, HF, ) in common
  run
• Luminosity will then be monitored as

32
Conclusion

• TOTEM will measure total and elastic cross
  section in dedicated runs in the initial phase of
  the LHC.
• It will provide absolute luminosity measurement
  and calibration to CMS monitors.
• The experiment will be ready to run at machine
  startup.