b physics prospects at lhcb
DESCRIPTION
B Physics prospects at LHCb. Emanuele Santovetti on behalf of the LHCb Collaboration Università di Roma Tor Vergata e INFN. Physics motivations Detector requirements Physics program. Physics motivation. - PowerPoint PPT PresentationTRANSCRIPT
E. Santovetti DIS2007 Munich 16-20 April 1
B Physics prospects B Physics prospects at LHCbat LHCb
Emanuele Santovetti on behalf of the LHCb CollaborationUniversità di Roma Tor Vergata e INFN
Physics motivations Detector requirements Physics program
DIS2007 Munich 16-20 April 2E. Santovetti
Physics motivationPhysics motivation LHCb is a dedicated B physics precision experiment at LHC to study
CP violation and rare decays
Standard Model describes CP violation by a single complex phase in the unitary CKM matrix
We will over constrain the unitarity triangles and search for new physics
LHCb precision enough to distinguish unitarity triangles (equal up to the 3 order)
New Physics can manifest through the exchange of a new intermediate particle in box and penguin diagram
compare measurements where NP effects are expected with tree level ones (no loop or penguins)
DIS2007 Munich 16-20 April 3E. Santovetti
Detector design: B acceptanceDetector design: B acceptance Designed to maximize B acceptance
(within cost and space constraints)
Forward spectrometer, 1.9 < < 4.9 more b hadrons produced at low angles
single arm OK since b-b pairs produced correlated in space
Luminosity tuned (2×1032 cm-2s-1) to maximize the probability of single interaction per x-ing
pT trigger can be lowered up to 2 GeV/c, efficient also for purely hadronic B
1 year of running = ~2 fb-1 and 1012 b-b events @nominal luminosity
230 μb
100 μb
Pythia production cross section
bb correlation
DIS2007 Munich 16-20 April 4E. Santovetti
Detector requirementsDetector requirements Time dependent measurements
Reconstruction in the harsh LHC environment
B events ~ few % of the total cross section Need of a selective trigger
Mass and pointing constraint to reduce background
Bs
Btag
K-
K+
+
+,K+
p pPV
Good K/ separation
B flavor tagging
~1 cm
SV
Good vertex spatial resolutionto determine proper time
DIS2007 Munich 16-20 April 5E. Santovetti
LHCb detectorLHCb detector
beam
VELO
Tracking stations
RICH system ECAL and HCAL
Muon stations
DIS2007 Munich 16-20 April 6E. Santovetti
Detector performancesDetector performances
PV
B vertex
IPdp/p = 0.5%
Momentum resolution
Particle Identification Eff.
(0) =4.7 MeV/c2
+ 12C → 0 + X → 2 + X0 invariant mass spectrum
DIS2007 Munich 16-20 April 7E. Santovetti
Flavor taggingFlavor tagging Opposite side
Charge of the kaon in the b→ c→ s chain Charge of the lepton in semi-leptonic
decays Charge of accompanying b jet
Same side Charge of the K accompanying Bs
Charge of the from B** → B*±
Tag (%) Bd Bs
Muon 1.1 1.5
Electron 0.4 0.7
Kaon opp. side 2.1 2.3
Vertex charge 1.0 1.0
Same side /k 0.7 () 3.5 (K)
Combined (neu.net) ~ 5.1 ~ 9.5
Tagging power in ε(1-2ω)2
DIS2007 Munich 16-20 April 8E. Santovetti
LHCb Physics programLHCb Physics program Bs mixing parameters: s, ms, s
with Bd→0+-
with Bd → J/Ks
with different methods Rare decays
Bs → to the level of the SM prediction Radiative penguin Bd → K*, Bs → , Bd → ω Electroweak penguin Bd → K*
and much more, e.g. Bc, charm physics, D0 mixing and CP violation)
not exhaustive list !
DIS2007 Munich 16-20 April 9E. Santovetti
sin2sin2with Bwith B00J/J/KKSS
One of the first CP measurements golden mode, very well measured
by b-factories will be an important check of CP
analyses and of tagging performance
can search for direct CP violating term cosmdt
Expect 240k reconstructed* B0 J/ KS events/2fb-1
Precision stat(sin 2 ~ 0.02 in 2fb-1 of collected data [currently (sin 2 ~ 0.03]
ACP(t) - background subtracted
*after trigger and reconstruction
DIS2007 Munich 16-20 April 10E. Santovetti
Measure Measure
is the least well measured CKM angle and LHCb has several ways to measure it (independent)
Most promising method is the ADS+GLW applied to B → DK B+→D(K)K+, D(K3)K+, D(KK)K+
B+→D*(K)K+
B0→D(K)K*0, D(KK)K*0, D()K*0
BS→DSK
Dalitz analysis for the neutral and charged B → DK decays
B0 → and Bs → K+K-, sensitive to new physics
DIS2007 Munich 16-20 April 11E. Santovetti
The BThe B → DK → DK decays (GLW and ADS method) decays (GLW and ADS method)Consider the following decays (tree level)
210
3
2002
001
2 AAKDBAA
eKDBAKDBAA
KDBAKDBAAi
11 AA
2A 2A3A
3A
2
Measuring the three different decay rates (relative) and the c.c. will allow to extract the gamma angle in a clean way, but… because the color suppression, the two amplitudes, A1 and A2, very different → large error
need the D0 tag
Decays of D0, D0 to same final state allows the two tree diagrams (theoretically clean!) to interfere. Consider the decay D0 → K-+
KDBKDBKDB 000
KDKD 00
Cabibbo favored Cabibbo suppressed
Color favored Color suppressed
For these decays the reversed suppression of the D decays relative to the B decays results in much more equal amplitudes → big interference effects ~ O(1)
Counting experiment: no need for flavor tagging or proper time determination → measure of BR ~ O(10-7) or smaller
DIS2007 Munich 16-20 April 12E. Santovetti
K
DiKD er
KDA
KDA
KDA
KDA
favouredA
DCSA
0
0
0
0
Interference parameters From the B decays: – because have b→u , b→c interference
rB – the ratio in magnitude of two diagrams (≤0.1 for DK±)
δB – a CP conserving strong phase difference
The D decays introduce: rDK – the ratio in magnitude of two diagrams (0.060)
δDK – a CP conserving strong phase difference
)cos(2)())((
)cos(2)(1))((
)cos(2)())((
)cos(2)(1))((
22
2
22
2
KDB
KDB
KDBD
KDB
KDB
KDBD
KDB
KDB
KDBD
KDB
KDB
KDBD
rrrrKKB
rrrrKKB
rrrrKKB
rrrrKKB (1) 56000*(1) 56000*
(2) 700(2) 700
(3) 56000(3) 56000
(4) 700(4) 700
rates (1) and (3) are favored + CP eigenstate = GLW method
rates (2) and (4) are suppressed + another D decay = ADS method
but these suppressed rates have order 1 interference term, as rB~rD
4 B4 B±± → (K → (K))DDKK±± decays decays
*both charges
The BThe B → D(K → D(K)K)K decays decays
DIS2007 Munich 16-20 April 13E. Santovetti
BB →DK →DK strategy: ADS+GLW strategy: ADS+GLW D0→K
3 observables from the relative rates of the 4 processes, depend on 4 unknowns , rB, B, D
K
rDK
is already well measured
may benefit from cos(D) measurements in CLEO-c and/or BES III
Need another D0 decay channel to solve for all unknowns
D0→K (BR ~ 8%): provides 3 observables which depends on 4 unknowns , rB, B, D
K3 only D
K3new, rDK3
is already well measured
D0→KK/ each CP mode provides one more observables with no new unknowns
)cos(21))((
)cos(21))((2
2
BBBD
BBBD
rrKhhB
rrKhhB
DIS2007 Munich 16-20 April 14E. Santovetti
-25 -16.6 -8.3 0 8.3 16.6 25-180 3.8 3.1 3.0 3.9 3.2 2.8 2.7-120 2.8 2.5 2.5 2.5 2.2 2.2 1.9-60 3.9 3.6 3.4 4.3 4.2 3.7 3.60 4.5 4.2 4.4 5.6 8.0 6.1 4.960 3.3 3.1 3.3 4.6 6.6 9.4 11.0120 3.4 3.6 3.8 4.1 3.9 3.6 3.3180 3.5 3.0 2.9 3.8 3.2 2.8 2.6
-25 -16.6 -8.3 0 8.3 16.6 25-180 8.6 7.5 6.5 6.8 7.2 7.3 6.0-120 6.0 6.3 6.3 6.4 6.2 6.2 4.7-60 8.0 7.9 8.1 7.8 7.4 6.7 6.20 10.3 11.1 12.4 11.5 12.1 13.1 13.060 9.1 10.6 11.2 12.9 13.4 15.0 15.2120 11.6 11.3 11.8 11.0 10.9 11.1 10.8180 8.5 7.4 6.5 6.8 7.1 7.3 6.5
δDK, δD
K3
δDK, δD
K3
w/o bkgrd
estimated bkgrd
LHCb sensitivityLHCb sensitivity Toy MC to simulate 2 fb-1 signal data:
→ fit results return input values Combine K with:
K3 similar yields and identical background level
KK 4300 B+ and 3300 B- with B/S ~ 2
() ~ 5º-15º in 2 fb-1 depending on rB, δD
K and δDK3
(Highlighted are RMS quoted from non-Gaussian distribution of fit results due to close ambiguous solutions, will disappear as statistics increase. Global analysis using all modes will also help.)
DIS2007 Munich 16-20 April 15E. Santovetti
ADS with BADS with B →D*K →D*K decays decays
Attractive feature D*→D00 (BR~2/3) – has strong (CP con.) phase δB
D*→D0 (BR~1/3) – strong phase δB+ → if can distinguish the two decays, powerful
additional constraint ! Preliminary studies (without background) show that,
including D*K improves precision of previous analyses to = 2º - 5º (favored mode: 17k/2 fb-1)
However… Reconstruction efficiency is small for soft while
background is enormous Non trivial to separate D00 and D0 Fit DK mass shape to get and components
ignoring neutrals
1 bb event here represents 5k/2fb-1
D0K mass
Charged reconstruction
B±→D*(D00)K±
B±→D*(D0)K±
B±→D*(D0)K±
bb sample
Signal/bkgrdArbitrary norm.
B± mass
DIS2007 Munich 16-20 April 16E. Santovetti
GLW method - GLW method - from B from B0 0 →→ DD00K*K*00
Dunietz variant of Gronau-Wyler method makes use of interference between two color-suppressed diagrams interfering via D0 common final states (, K, KK)
Measuring the 6 decay rates, B0 D0(K,,KK)K*0 + CP conjugates, allows to be extracted without flavor tagging or proper time determination
d
b
d
s
u
c
B0
D 0
K*0
D0
d
b
d
s
c
u
B0
K*0
A1 = A(B0 D0K*0): bc transition, phase 0
A2 = A(B0 D0K*0): bu transition, phase +
A3 = 2 A(B0 DCPK*0) = A1+A2, because DCP=(D0+D0)/2
Mode (+ cc) Yield* Bbb/S (90%CL)
B0 (K)D K*0 3400 < 0.3
B0 (K)D K*0 500 < 1.7
B0 (KK)D K*0 600 < 1.4
() ~ 7º - 10º (2 fb-1) depending on , δD
K and δDK3
*in 2 fb-1, both charges
DIS2007 Munich 16-20 April 17E. Santovetti
BBss→ D→ DssKK Interference between tree level decays via BS
mixing (s-version of Bd→D*): time dependent analysis (clean, no penguins)
Measures +S (S from BS→J/Ψ Φ)
Expect 5400 events/2fb-1
Excellent proper-time resolution (t ~ 40 fs) allows to resolve BS oscillations
() ~ 13° from 2fb-1 data [ms=17.3 ps–1]
Main background from DS(BR×10 ) Suppressed using kaon ID from RICH detector B/S <1 @ 90% CL
Parallel analysis possible with Bd→D±
(~790k events/2fb-1 with B/S ~ 0.3, extraction requires rD or combined Bs→ DsK U-spin analysis)
DIS2007 Munich 16-20 April 18E. Santovetti
from Bfrom B0 0 hhhh B0 → originally proposed to measure but the influence of penguin diagrams
makes the task difficult
)sin()cos()( mtAmtAtA mixdirCP
),,,()(
),,()(
),,,()(
),,()(
4
3
20
10
dfKKBA
dfKKBA
dfBA
dfBA
smix
sdir
mix
dir
Assuming U-spin flavor symmetry (u↔s) d=d’ and ’ and taking from Bd → J/Ks and from Bs → J/ we can solve for
From the time dependent CP asymmetry
Extract four asymmetriesdei = ratio of penguin and tree amplitude in B→d’ei’ = ratio of penguin and tree amplitude in B→KK
DIS2007 Munich 16-20 April 19E. Santovetti
from Bfrom B0 0 hhhh
26k Bd → events with 2fb-1, B/S < 0.7
37k Bs → KK events, B/S = 0.3 ~ 5º + uncertainty from U-spin symmetry breaking Sensitive to new physics
DIS2007 Munich 16-20 April 20E. Santovetti
Summary of performances on Summary of performances on B mode D mode Method ()
B+→DK+ K +KK/ + K ADS+GLW 5°-15°
B+→D*K+ K ADS+GLW under study
B+→DK+ KS Dalitz 15°
B+→DK+ KK 4-body “Dalitz” 15°
B+→DK+ K 4-body “Dalitz” under study
B0→DK*0 K + KK + ADS+GLW 7°-10°
B0→DK*0 KS Dalitz under study
BS→DSK KK tagged, A(t) 13°
B → ,KK 4°-10°
Combining all modes, with a nominal year of data (2 fb-1), LHCb will be able to extract with ~ 4° resolution, and compare the B→DK direct measurement with the indirect determination (B0 +–,Bs K+K–) to make a stringent test of the SM
Signal only, no accept. effect
DIS2007 Munich 16-20 April 21E. Santovetti
BBs s mixing phase, mixing phase, ss, from B, from Bss→→J/J/((++--) )
? t,c,u
t,c,u
W W
s,d
b s,d
b
0B0B
In the SM sSM = -2 = -2 (small) and from UT fits we get s
= -0.037± 0.002 Direct measurements not very precise: recent D0: -0.79+0.47 -0.39
CP violating decay can proceed directly or through mixing.
The “mixing box” can have contributions from NP particles, resulting in s = s
SM + sNP ≠ -22
Measure proper time dependenceMeasure time-dependent CP asymmetry
= ± 1 CP
Tagged Bs
Tagged Bs
DIS2007 Munich 16-20 April 22E. Santovetti
BBs s mixing phase, mixing phase, ss, from B, from Bss→→J/J/((++--) ) Because the final state contains two vector
particles, it is a mixture of CP odd and CP even
Use tr angle between + and normal to decay plane to do an angular analysis to identify the states.
With: 2 fb-1 of dataPrecision on s s 0.023
Simultaneous fit to time and angular distributions
CP even
CP odd
bkgnd
Total
From pure CP states Bs → J (s)= 0.059
Combining (s)= 0.021 (UT fit value: -0.037)
DIS2007 Munich 16-20 April 23E. Santovetti
BBss→→ ++--
Very small branching ratio in SM: (3.4 ± 0.5) × 10-9
Present limit from Tevatron at 95% CL(1 fb-1): < 7 × 10-8 (expected final limit at 95% CL (8 fb-1): < 2 × 10-8)
Sensitive to New Physics through loops
Could be strongly enhanced by SUSY.
W
W
b
s
t
?
?
MSSM
DIS2007 Munich 16-20 April 24E. Santovetti
BBss→→ ++--LHCb Sensitivity (signal+bkg is observed)
Integrated Luminosity (fb-1)B
R (
×10
-9)
5
3
SM prediction
with L=2fb-1
3 observation if at SM value
LHCb should have good prospect for significant measurement: 17 SM events/2fb-1
Difficult to get reliable estimate of expected background
No background events selected in sample of 33M events but estimation limited by statistic
Combinatorial: B to +X, B to -X known as the main source of background, addressed by very good mass resolution 18 MeV/c2
Bd, Bs to , K, KK and mis-id. addressed by particle identification and mass resolution.
DIS2007 Munich 16-20 April 25E. Santovetti
ConclusionsConclusions
LHCb will be ready to collect data with its full detector as LHC turns on. Physics at 14 TeV starting in 2008
It will make precision measurements that will severely constrain the unitarity triangle fits and probe rare decays
These measurements could either limit New Physics contributions to B decays or, more optimistically, uncover them
DIS2007 Munich 16-20 April 29E. Santovetti
The BThe B → DK → DK decays (GLW method) decays (GLW method)
b
0D
K
Bc
s
u
uu
b
0D
KB
u
s
c
uu
Consider the following diagrams and the c.c.
210
3
2002
001
2 AAKDBAA
eKDBAKDBAA
KDBAKDBAAi
11 AA
2A 2A3A
3A
2
colour favouredcolour favoured
colour suppressedcolour suppressed
Measuring the three different decay rates (relative) and the c.c. will allow to extract the gamma angle in a clean way, but…
two amplitudes (A1 and A2) very different → large error
need the D0 tag
DIS2007 Munich 16-20 April 30E. Santovetti
The ADS method The ADS method
Decays of D0, D0 to same final state allows the two tree diagrams (theoretically clean!) to interfere. Consider the decay D → K-+
For these decays the reversed suppression of the D decays relative to the B decays results in much more equal amplitudes → big interference effects ~ O(1)
Counting experiment: no need for flavor tagging or proper time determination
→ measure of BR ~ O(10-7) or smaller
Cabibbo favouredCabibbo favoured
KfDB
KfDB
i
i0
0
K
0D
s
d
u
uu
c
Cabibbo suppressedCabibbo suppressed
B+→D0K+ (colour suppressed) B+→D0K+ (colour favoured)K
0
Dd
u
s
uu
c