1. The Large Hadron Collider beauty Experiment & Physics
Sheldon Stone
Syracuse Univ.
LHCb
The Large Hadron Collider beauty Experiment & Physics

2. General Physics Justification
Expect New Physics will be seen at LHC
Standard Model is violated by the Baryon Asymmetry of Universe & by Dark Matter
Hierarchy problem (why MHiggs<<MPlanck)
However, it will be difficult to characterize this physics
How the new particles interfere virtually in the decays of b’s (& c’s) with W’s & Z’s can tell us a great deal about their nature, especially their phases
Aspen Winter Conference, February, 2006

3. Example MSSM from Hinchcliff & Kersting (hep-ph/0003090) BsJ/yh
Contributions to Bs mixing
BsJ/yh
CP asymmetry  0.1sinfmcosfAsin(Dmst), ~10 x SM
Contributions to direct CP violating decay
Asym=(MW/msquark)2sin(fm), ~0 in SM
B-fK-
Aspen Winter Conference, February, 2006

4. Limits on New Physics From b’s
Is there NP in Bo-Bo mixing?
Assume NP in tree decays is negligible
Use Vub, ADK, SyK, Srr, Dmd, ASL
Fit to h, r, h, s h s
cl>5%
cl>32%
cl>90%
From Perez
For New Physics via Bdo mixing, h is limited to ~<0.3 of SM except when sBd is ~0o or ~180o of SM decays
New physics via Bs mixing, or bs transitions is unconstrained
Aspen Winter Conference, February, 2006

5. Most Currently Desirable Modes
BS mixing using BSDS+p-
High Statistics Measurement of forward-backward asymmetry in B K*m+m-
Precision measurements of CP ’s
CP violating phase in BS mixing using BSJ/yf
g (or f3) Using B- DoK- tree level decays
g using BSDS+K- time dependent analysis
a especially measurement of Bo roro
b at high accuracy to pin down other physics
CPV in various rare decay modes
B(S) m+m-
Important: Other modes, not currently in vogue
Aspen Winter Conference, February, 2006

6. Detector Requirements - General
Every modern heavy quark experiment needs:
Vertexing: to measure decay points and reduce backgrounds, especially at hadron colliders
Particle Identification: to eliminate insidious backgrounds from one mode to another where kinematical separation is not sufficient
Muon & electron identification because of the importance of semileptonic & leptonic final states including J/y decay
g, po & h detection
Triggering, especially at hadronic colliders
High speed DAQ coupled to large computing for data processing
An accelerator capable of producing a large rate of b & anti-b hadrons in the detector solid angle
Aspen Winter Conference, February, 2006

7. Basics For Sensitivities
# of b’s into detector acceptance
Triggering
Flavor tagging
Background reduction
Good mass resolution
Good decay time resolution
Particle Identification
Aspen Winter Conference, February, 2006

8. The Forward Direction at LHC
100 mb
230 mb
Pythia production cross section
In the forward region at LHC the bb production s is large
The hadrons containing the b & b quarks are both likely to be in the acceptance
LHCb uses the forward direction, 4.9 > h >1.9, where the B’s are moving with considerable momentum ~100 GeV, thus minimizing multiple scattering
At L=2x1032/cm2-s, we get 1012 B hadrons in 107 sec pT h
Production
 Of B vs B
q B (rad)
q B (rad)
Aspen Winter Conference, February, 2006

9. The LHCb Detector Muon Detector Tracking stations Trigger Tracking
proton
beam
interaction
region
Trigger
Tracking
Aspen Winter Conference, February, 2006

10. The VELO Geometry Sensor Half Vacuum Tank  R f sensors
3 cm separation f
sensors R
R sensor: 38 mm pitch inside to
103 mm outside
f sensor: 39 mm pitch inside to
98 mm outside 
Interaction point
Geometry
Sensor
Half
Vacuum
Tank
Aspen Winter Conference, February, 2006

11. Triggering
Necessary because b fraction is only ~1% of inelastic cross-section
At peak luminosity interaction rate is ~10 MHz, need to reduce to a few kHz. The B hadron rate into the acceptance is 50 kHz
General Strategy
Multilevel scheme: 1st level Hardware trigger on “moderate” pT m, di-muons, e, g & hadrons, e.g. pT m >1.3 GeV/c; veto on multiple interactions in a crossing except for muon triggers.
Uses custom electronics boards with 4 ms latency, all detectors read out at 1 MHz
Second level and Higher Level software triggers
Aspen Winter Conference, February, 2006

12. Software Triggers
Second Level: All detector information available. Basic strategy is to use VELO information to find tracks from b decays that miss the main production vertex; also events with two good muons are accepted & single muon with pT > 2.1 GeV/c. Strategies are constantly being improved.
Higher Level Triggers: Here more sophisticated algorithms are applied. Both inclusive selections and exclusive selections tuned to specific final states done after full event reconstruction has finished. Output rate is ~2 kHz
Aspen Winter Conference, February, 2006

13. Trigger Output
Output rate
Trigger Type
Physics Use
200 Hz
Exclusive B candidates
Specific final states
600 Hz
High Mass di-muons
J/, bJ/X
300 Hz
D* Candidates
Charm, calibrations
900 Hz
Inclusive b (e.g. bm)
B data mining
Rough guess at present (split between streams still to be determined)
Large inclusive streams to be used to control calibration and systematics (trigger, tracking, PID, tagging)
Aspen Winter Conference, February, 2006

14. Trigger Monitoring
Trigger lines need constant monitoring to adjust prescales, especially at beginning of experiment.
General approach: for a particular trigger
Define TOSTrigger On Signal
Define TIS Trigger Independent of Signal
Efficiency =(TISTOS )/TIS
Aspen Winter Conference, February, 2006

15. Trigger Monitoring Example
Comparison of L0 trigger efficiency on muon tracks that miss the IP as a function of Pt for both “traditional” Monte Carlo method & (TISTOS )/TIS
Can be done quickly with real data
TIS & TOS
Method
Traditional MC
Aspen Winter Conference, February, 2006

16. Flavor Tagging Method (For BS) m± e± K± same K± opp 1.5 0.7 3.1 2.5
“opposite side”
For Mixing & CP measurements
it is crucial to know the b-flavor
at t=0. This can be done by
detecting the flavor of the other B
hadron (opposite side) or by using
K± (for BS) p± (for Bd) (same side)
Efficacy characterized by eD2, where
e is the efficiency and D the dilution = (1-2w)
Several ways to do this
“same side”
Method
(For BS) m± e±
K± same
K± opp
Jet charge
eD2(%)
1.5
0.7
3.1
2.5
0.8
eD2 (%)
Not exactly
same cuts as table
Expect eD2 ~ 7.5% for BS & 4.3% for Bd
Aspen Winter Conference, February, 2006

17. Background Reduction Using st
Excellent time resolution ~40 fs for most modes based on VELO simulation
Example
BS mixing
Bs→Ds-π+
100 mm
Bs→Ds-π+ (tagged as Bs)
10 mm
LHCb can measure DmS up
68 ps-1 in 2 fb-1
Aspen Winter Conference, February, 2006

18. Background Reduction from Particle ID
LHCb has two RICH detectors. Most tracks in range 100>P>2 GeV/c. Tagging kaons at lower momentum < 20 GeV/c; Bh+h- up to 200 GeV/c, but most below 100 GeV/c
Good Efficiencies with small fake rates
CDF data
Bd signal
Excellent
mass resolution
s=14 MeV
Aspen Winter Conference, February, 2006

19. The RICH Detectors HPD Photon Detectors RICH I Design 80 mm
Aspen Winter Conference, February, 2006

20. RICH II RICH2 –installed in the pit
Aspen Winter Conference, February, 2006

21. CP Asymmetry in BSJ/ f
Just as BoJ/ KS measures CPV phase b BSJ/ f measures CPV BS mixing phase fS
Since this is a Vector-Vector
final state, must do an angular
(transversity) analysis
The width difference DGS/GS
also enters in the fit
LHCb will get 120,000 such
events in 2fb-1. Projected errors are ±0.06 in fS & 0.02 in DGS/GS (for DmS = 20 ps-1)
Including BSJ/ h, will increase sensitivity (only 7K events)
Aspen Winter Conference, February, 2006

22. Neutral Reconstruction
Mass resolution is a useful s=~6 MeV
Efficiency within solid angle is OK using both merged and resolved po’s
Example: time dependent Dalitz Plot analysis ala’
Snyder & Quinn for Borp p+p-po
14K signal events in 107 s with S/B 1/3, yielding s(a)=10o
Resolved 0
Merged 0
Aspen Winter Conference, February, 2006

23. Other Physics Sensitivities
Afb
Zero to ±0.04 GeV2
Only a subset of modes
For ~2 years of running
Aspen Winter Conference, February, 2006

24. Status Magnet installed & mapped ECAL, HCAL, RICH II
& Muon Filter Installed
Construction on all
other items proceeding
Software is progressing
New MC-data challenges using Grid
Aspen Winter Conference, February, 2006

25. Overview Overall in very good shape for startup in 2007
Aspen Winter Conference, February, 2006

26. View of Pit Muon system Calorimeter -E-cal, H-cal modules RICH2 Magnet
-iron shielding
-electronics tower
Calorimeter -E-cal, H-cal modules
RICH2
Magnet
RICH1
-HPD shielding box
OT: straw module production completed Muon: more than half of chambers produced
Aspen Winter Conference, February, 2006

27. Possible Improvements
Run at higher luminosity.
Increase to 5x1032 /cm2-s
Gains in event yields,
especially dimuon modes
Bop+p-
BSfg
BSJ/yf
BSDSK-
Aspen Winter Conference, February, 2006

28. Possible Upgrades
VELO needs to be replaced after ~6-8 fb-1 due to radiation damage
Are considering hybrid Silicon pixels as a replacement
Since they are much more rad hard than current VELO, we could move closer to the beam getting better vertex s
These could possibly allow some vertexing in first trigger level with minor modifications
EM calorimeter upgrades such as having a central PbWO4 region
Major modifications to readout including long digital pipelines that would enable extensive 1st level vertex triggering and allow higher luminosity running (very expensive)
Aspen Winter Conference, February, 2006

29. Conclusions
LHCb will study CP Violation and Rare Decays in the BS, B-, & Bd systems at an unprecedented level of accuracy
These studies are crucial for specifying any new physics found directly
at the Tevatron or LHC
LHCb is on schedule
LHCb is starting to think
about upgrades
From Hewett & Hitlin
Aspen Winter Conference, February, 2006

30. The End