c LFV in Tau experiments K Hayasaka Niigata

c. LFV in Tau experiments K. Hayasaka (Niigata) 4/Jun/2019 International School on c. LFV 1

Introduction to tau lepton 4/Jun/2019 International School on c. LFV 2

Tau lepton Features • Charged lepton＝electron • Third lepton • triton (third) = τριτον • The heaviest lepton • Heavier than π （Muon is lighter than π. ） ＝possible to decay includingπ • Weak decay • Various decays are possible. （Muon has at most 3 decays） 4/Jun/2019 International School on c. LFV 3

Who discovers tau lepton? https: //en. wikipedia. org/wiki/Martin_Lewis_Perl 4/Jun/2019 International School on c. LFV 4

Discovery of tau lepton It has been discovered at the 4 -Ge. V machine on SLAC, called SPEAR, with 64 tau-pair candidates. e++e- t++t- e±+m∓+missing energy 4/Jun/2019 International School on c. LFV At e+e- collider, since bean energy is known, via energy conservation law, the energy missing from the measurement can be estimated. It is called missing energy. 5

Tau lepton Features • Charged lepton＝electron • Third lepton • triton (third) = τριτον • The heaviest lepton • Heavier than π （Muon is lighter than π. ） ＝possible to decay includingπ • Weak decay nt t- • Various decays are possible. （Muon has at most 3 decays） • Always neutrino is included in decay products. 4/Jun/2019 International School on c. LFV W- 6

From PDG http: //pdg. lbl. gov/ →PDGLive →τ 4/Jun/2019 International School on c. LFV 7

From PDG 1 prong: 85% =decays including only one charged track 4/Jun/2019 International School on c. LFV 8

From PDG 1 prong: 85% 3 prong: 15% 5 prong: 0. 1% 4/Jun/2019 International School on c. LFV Almost all tau decays have 1 - or 3 -prong 9

Contents of tau decay • Leptonic: 35% • 1 -prong: 85% • 3 -prong: 15% • 5 -prong: 0. 1% • ΔS=1: 3% (including odd number of K） 4/Jun/2019 International School on c. LFV 10

Tau production at e+e- collider 4/Jun/2019 International School on c. LFV 11

Various events at e+e- collider Quark Lepton 2 photon（ee γγ eeff） e, m, t f=e, m, t, u, d, s, c, b e γ γ e f ee ee is called “Bhabha”. For events having g, put Since quarks cannot appear “Radiative” on the head. alone, products appear via （Radiative Bhabha “hadronization”. There, many =ee eeγ） particles appear. Depending on √s, heavy quarks are not allowed. 4/Jun/2019 International School on c. LFV When beam particle passes each other, gs are emitted and they collide and then fermions are produced. 12

B-factory/Belle II 4/Jun/2019 International School on c. LFV 13

B-factory • Originally, it is built to prove that B meson CPV is larger than Kaon’s. • To produce B meson-pair via Y(4 S), CMS energy is set as Y(4 S) mass. (=10. 58 Ge. V) • To evaluate time-dependent CPV, flying B is better than stopping B. Therefore, beam energy is set asymmetrically. 4/Jun/2019 International School on c. LFV 14

Various events at B-factory Quark Lepton 2 photon（ee γγ eeff） e, m, t f=e, m, t, u, d, s, c, b e γ γ e f Cross section depends on √s 4/Jun/2019 International School on c. LFV 15

Initial State Radiation (ISR) • 4/Jun/2019 International School on c. LFV 16

Beam background • Electron/positron that does not join the collision sometimes away from beam bunch directly hits detector or flies into material and make some secondary particles and then they hit detector… • Sometimes detected as charged track, sometimes as photon… Beam current increase= Beam background increase 4/Jun/2019 International School on c. LFV 17

ＫＥＫ/KEKＢ accelerator KEK(High Energy Accelerator Research Organization), Tsukuba, Japan 4/Jun/2019 International School on c. LFV Circumference = 3 km 18

KEKB/Belle B-factory: E at CM = Υ(4 S) e+(3. 5 Ge. V) e-(8 Ge. V) Highest instant luminosity = 2 x 1034(cm-2 s-1) Belle finished data taking on Jun. 30, 2010. total: >1 ab-1 U(4 S): 711 fb-1 U(5 S): 121 db-1 U(3 S): 3. 0 fb-1 U(2 S): 24 fb-1 U(1 S): 5. 7 fb-1 Off-resonance: 87 fb-1 4/Jun/2019 Asymmetric multi-purpose detector Good track reconstruction and particle identification Lepton ID eff. ~ (80 -90)% Fake rate ~ (0. 1 -3)% (Integrated luminosity)x(production cross Section)=produced number of events International School on c. LFV 19

Super. KEKB • Current: x 2 • Beam size: x 1/20 In total, x 40 higher performance To make beam stably, beam energy Is changed from KEKB: e- 8 Ge. V, e+3. 5 Ge. V to Super. KEKB: e- 7 Ge. V, e+4 Ge. V→Observed B life time has been reduced. 4/Jun/2019 International School on c. LFV 20

Luminosity expectation Around 2027, it is expected that Int. Lum. reaches 50 ab-1 corresponds to 4. 5 x 1010 t-pair production. 4/Jun/2019 International School on c. LFV 21

Belle II • According to increase of the performance for accelerator, the detector also should be upgraded. • Increase of collision rate • Time span for collision gets shorter • BG originated from beam gets higher. • Beam energy asymmetry gets lower. • PID perfomance Simulation for tracker signals. 画像はイメージです 4/Jun/2019 International School on c. LFV 22

Belle II collaboration and a region. 4/Jun/2019 International School on c. LFV 23

From China • Liao. Ning Normal University(LNNU) • Peking Univ. (PKU) • Beihang Univ. (BUAA) • Soochow University • IHEP-China • Univ. of Science and Technology of China(USTC) • Fudan Univ. 4/Jun/2019 International School on c. LFV 24

Tau-pair feature in measurement level 4/Jun/2019 International School on c. LFV 25

Belle MC samples # of observed charged tracks ττ μμ eeμμ 4/Jun/2019 International School on c. LFV 26

Total observed energy (CMS) ττ μμ eeμμ 4/Jun/2019 International School on c. LFV 27

Pros/Cons for tau analyses • 4/Jun/2019 International School on c. LFV 28

Thrust vector Find which maximizes T. Thrust vector is defined as the direction is and its length is T. Direction where visible tracks and photons Sum of all visible tracks and photons Since SM tau decays always include neutrino(s) as a decay products and tau can not fully reconstructed, it is difficult to evaluate tau direction. Thrust vector is often used as a substitute for tau direction CMS τ Thrust axis As similar ones, acoplanarity and sphericity can be defined. 4/Jun/2019 International School on c. LFV e e τ 29

Size of thrust ττ spherical μμ Jet-like eeμμ 4/Jun/2019 International School on c. LFV 30

Features of tau-pair events • ~ ~ Since tau cannot be reconstructed, using above conditions and so on, more tau-pair-like events are selected. 4/Jun/2019 International School on c. LFV 31

Tau helicity • Since τ comes from pair-production… Helicity correlates each other. If tau helicity information is important in your analysis, from another tau, you can take it. 4/Jun/2019 International School on c. LFV 32

t Decay and helicity tgpn ht: helicity, xp=Ep/Ebeam If you do not see the helicity, distribution gets flat. (averaged) A. Shtal: Physics with Tau Leptons ISBN: 978 -3 -540 -66267 -9 (Print) 978 -3 -540 -48458 -5 (Online) 4/Jun/2019 International School on c. LFV 33

Correlation between t-gp+gp+n n/t t-rest frame MC n t p q hel. of t- =1 MC hel. of t- =-1 4/Jun/2019 Black (high) ⇔White(low） International School on c. LFV 34

Generic tau analysis • Since t-pair is produced, CMS Thrust axis Signal side τ e e τ Tag side 4/Jun/2019 1. t decaying into what we like to see is called “on signal side”, while anothe t is called “on tag side”. 2 -a. When signal side t makes a hadronic decay, we require a leptonic decay for tag side t. 2 -b. When signal side tau decay includes a lepton as a decay product, it is avoided to appear same lepton on tag side. ３． Missin momentum or mass should be lar ４．Size of the thrust should be around 1 but less than 1. 5. Decay products on signal side and tag side are well separated. International School on c. LFV 35

Pros/Cons for tau analyses • # of charged tracks : low • # of gammas : low Selection criteria gets simple. It is difficult to add original idea for the selection. 4/Jun/2019 International School on c. LFV 36

Analysis of tau LFV 4/Jun/2019 International School on c. LFV 37

Experimental advantage for LFV search • Various attempts to find New Physics (NP) exist. • Deviation from SM expectation is evaluated to find NP. • Usually, SM expectation has ambiguity from QCD, limitation of theoretical calculation and so on…as well as uncertainties from measurements Difficult to judge the existent of NP • If SM expectation is exactly zero, judgement get much easier. • LFV is exactly forbidden in SM. 4/Jun/2019 International School on c. LFV 38

Targets • Any possible decays are new physics candidates. • But, at least, spin and charge conservation lows should be kept. tgℓg, tgℓℓℓ, tgℓhh, tgℓP 0/S 0/V 0, … Of course, we can consider, for example, t-gm-e+e+nn, but here, we consider only neutrinoless decays. 4/Jun/2019 International School on c. LFV 39

Homework/discussion • Actually, LHb published search result for tgpmm. 4/Jun/2019 International School on c. LFV 40

Rough sketch for tau LFV processes • ~ ~ ~ n t n~m (or n~e) n~t n~m (or n~e) Here, I show the simplest process, no one knows what process induces tau LFV decay. 4/Jun/2019 International School on c. LFV cf. no corresponding m decay 41

Rough sketch for tau LFV processes • 4/Jun/2019 GUT conserves B-L… International School on c. LFV 42

Comments on m LFV SUSY SU(5) GUT with seesaw MEG(2011) MEGA(2002, 1999) Belle(2008) Ba. Bar(2010) 200 Ge. V-1 Te. V Correlation between m and t LFVs depends on model. So, generally, they should be considered as “independent”. (Probably, in NP, lepton universality is also violated. ) On the other hand, experimentally, t LFV decay measurements can be discussed with same systematics. 4/Jun/2019 International School on c. LFV MEG(2011)= Phys. Rev. Lett. 107, 171801 (2011) MEGA(2002)= Phys. Rev. D 65, 112002 (2002) MEGA(1999)= Phys. Rev. Lett. 83, 1521 (1999) Belle(2008)= Phys. Lett. B 666, 16(2008) Ba. Bar(2010)=Phys. Rev. Lett. 104, 021802 (2010) 43

Signal discriminant（１） • Neutrinoless t LFV decay can fully reconstruct mother t from daughters. The mass evaluated from sum of 4 -momenta for daughter should be around t mass in signal events. 4/Jun/2019 International School on c. LFV 44

Signal discriminant（2） Neutrinoless t LFV decay can fully reconstruct mother t from daughters. We have one more variable for t-pair produced at Bfactory. CM系 4/Jun/2019 International School on c. LFV 45

Lessons from Belle experiences 4/Jun/2019 International School on c. LFV 46

Analysis method of tau LFV e ta sid g e si sid gna e l Signal Extraction m g t t ~ t mass e generic 1 -prong decay DE~0 Blind analysis ⇒Blind signal region Estimate BG level: sideband data and MC Signal extraction: UEML fit/counting on the Mmg-DE plane. If no excess: set upper limits @ 90%CL Differently from muon case, charge conjugation is also considered. UEML=Unbinned Extended Maximum Likelihood fit 4/Jun/2019 International School on c. LFV signal MC distribution. Size of boxes shows density. 47

tgℓg Simplest, most promising… 4/Jun/2019 International School on c. LFV 48

Signature of signal and background m signal t e t mm event g Only tag side has neutrino! t t m e m Both sides have no neutrino total energy is equal to beam energy. qq event (q=u, d, s, c) g ISR g m e g ISR g e e SM tt event fake n n e … q e Both sides have neutrinos missing helps us 4/Jun/2019 to reject this kind of International BG’s. School on c. LFV … Both sides have many tracks and photons. No. of photons 49 decreases this kind of BG’s.

Concepts for signal selection criteria (tgmg) • Require 1 -1 prong events (2 charged tracks well separated) • Require muon by PID and photon close to muon (using opening angle) • Veto muon for another charged track to reject mm events • Require zero missing mass and large missing momentum, i. e. , only one neutrino is expected. 4/Jun/2019 International School on c. LFV 50