semileptonic decays at Belle Christoph Schwanda Institute of High Energy Physics Austrian Academy of Sciences Outline of this talk Measurement of B 0 D t n with semileptonic ID: 627323
Download Presentation The PPT/PDF document "Tree-level New Physics searches in" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Tree-level New Physics searches in semileptonic decays at Belle
Christoph SchwandaInstitute of High Energy PhysicsAustrian Academy of SciencesSlide2
Outline of this talkMeasurement of B
0 ® D*-t+
n with semileptonic tag (winter 2016 preliminary)
arXiv:1603.06711
Search for B
0 ® p-t+n with hadronic tagPhys. Rev. D93, 032007 (2016)
2Slide3
3
1999 – 2010: B factory at KEK (Japan)
Linac
KEKB double
ring e
+
e
-
collider
e
+
e
-
®
Y(4S)
® BB
Belle detector
World largest B meson sample
~771 million BB events
Over 450 Belle physics publicationsSlide4
Tagging techniques for Y(4S) eventsTagging provides:
Background suppressionInformation on Bsig(4-momentum)4
PURITY
EFFICIENCY
Untagged
No requirement on
B
tag
High efficiency,
low puritySemileptonic tagBtag ® D*lnEfficiency ~O(0.2%)
Hadronic tagB
tag ® hadronsEfficiency ~O(0.1%)Slide5
B ® D*
tn with semileptonic tag
5Slide6
New Physics in B ® D*tn
Semitauonic B decays of type b ® c
tn are sensitive probes to search for New Physics. NP can change the branching ratio and the D*/t
polarization.
6
Type II 2HDMA charged Higgs of spin 0 mediates the decay instead of the W
Can enhance or decrease the BR of B
®
D*
tnLeptoquark models
LQs are bosons which couple to a lepton-quark pairCarry
color and electric charge, baryon and lepton numberLQ models which generate an effective tensor operator lead to an effect in B
® D*tnSlide7
Principle of the measurementSimultaneously reconstruct signal and normalization events
D* reconstruction:D*+ ® D0p+, D+
p0 (~100%)10 D0 modes (~37%)
5 D
+
modes (~22%)Semileptonic tag: combineD*+ with an oppositely charged lepton, calculate cos qB,D*lRequire two tagged B candidates per event of opposite charge7
Signal event
Normalization eventSlide8
Principle of the measurement (2)Neural network to separate
signal and normalizationevents using:signal-side cos qB,D*lmissing mass squaredvisible energy
Determine the number of signal and normalization events by a two dimensional maximum likelihood fitFit variable 1: neural network outputFit variable 2: sum of energies of neutral clusters not associated to reconstructed particles EECL
Signal, normalization and B
®
D**ln yields are floated in the fit, other components are fixed to MC expectation 8Cos q
M
2
miss
EvisSlide9
Fit result9Slide10
Systematic uncertainty / stability10Slide11
Comparison to other measurementsThe difference with the SM prediction is at the level of 4.0 sigma for all four measurements combined
11
This measurementSlide12
B ® ptn
with hadronic tag12Slide13
MotivationWithin the type II 2HDM, the branching fraction of B
® ptn can be modified similarly to B ® D(*)tnCurrent experimental situation: B ®
tn branching fraction is SM-like while B ® D(*)
tn
exhibits a ~4
s anomalyB ® ptn thus provides additional insight and in particular is an independent probe of the b ® utn transition13Slide14
Principle of the measurementBtag
reconstructed in a hadronic modet reconstruction (~71%):t ® e/
mnnt ®
p
/r(pp0) n Background is suppressed with a multivariate discriminator (boosted decision trees)Signal is extracted from a one-dimensional fit to EECL14Slide15
Boosted decision trees (BDTs)15
One BDT for every t modeDiscriminant
variables:Missingenergy and momentumKL veto
…
Background: dominant
b ® c decaysSlide16
Fit result16
Electron mode
Pion
mode
Rho modeSlide17
Systematic uncertainty17Slide18
SUMMARY
18Slide19
SummaryNew preliminary Belle result for R(D*) obtained with semileptonic
tagging [arXiv:1603.06711]R(D*) = 0.302 +/- 0.030(stat) +/- 0.011(syst)Confirms the excess seen by other experiments and brings to tension with the SM to 4.0
sSearch for B0
®
p-t+n at Belle[Phys. Rev. D93, 032007 (2016)]Br(B0 ® p-t+n
) < 2.5 x 10
-4
@ 90% C.L.
We need the Belle II data to clarify the experimental situation19Slide20
backup
20