Upload
others
View
4
Download
0
Embed Size (px)
Citation preview
LFV in the Higgs sector and t Hq decays at LHC→
Luca Fiorini [email protected]
(IFIC U. of Valencia)
Workshop on Flavour Physics in the LHC era24th Nov 2015
Introduction● Lepton flavor violation exists in Nature (neutrino oscillations), but LFV in the charged sector is extremely suppressed in the SM. FCNC is also highly suppressed in the SM by GIM mechanism.
● The Higgs sector is a new reality in particle physics and may be a portal for beyond SM phenomena.● A number of models beyond SM predict LFV in charged sector and FCNC related to the Higgs sector at levels observable at LHC.
● Low energy results (e.g. e→ , eee, → e conversion, etc.) provide ‐indirect constraints, but there are often assumptions.
Luca Fiorini 2
l
l '
Luca Fiorini 3
LHC Luminosity and pileupRun1 Results based on 2011+2012 data
Luminosity is measured with forward/tracking detectors and calibrated with beam separation scans
~29 fb1 of data delivered during Run 1 and about 4 fb1 during 2015
● Pileup already at the design level and above during 2012, thanks to the excellent performance of the LHC.● Peak luminosity [cm2 s1]: 7.7x1033 (2012), 5.2x1033 (2015).
Higgs Production Modes at LHC
Total xsection:
17 pb @ 7 TeV22 pb @ 8 TeV51 pb @ 13 TeV
Dominant mode ggF(87%)
VBF (7%)
Associated VH (5%)
Associated bbH, ttH (<1%)
Luca Fiorini 4
Higgs Decay Modes
Luca Fiorini 5
They can also provide indirect measurement of couplings to quarks at LHC (via virtual loops)
125 GeV
W, Z
W, Z
f
f
Bosonic modes: , ZZ, WW
Fermionic modes: , bb, ...
h
h
Signal Strength: Production and Decay
Luca Fiorini 6
●
●
● Higgs results of Run1 data show good agreement with SM within uncert. ● Largest difference in mttH: 2.3 excessσ with respect to SM
SM BR assumed SM production assumed
SM pvalue: 25%
SM pvalue: 60%
ATLA
S C
ON
F2 015044 , CM
SP
AS
15 002
ATLA
S C
ON
F2 015044 , CM
SP
AS
15 002
Constraints on BSM couplings
Luca Fiorini 7
● Only x BRs can be measured. Without further assumptions on the σwidth of the Higgs boson BRBSM cannot be measured: assume kV≤1 (as in 2HDM)● kt measurement dominated by ttH process● BRBSM < 0.34 at 95% C.L. (assuming kV≤1)
SM pvalue: 11%
LFV h → 'ℓℓ
Luca Fiorini 8
General Higgs interaction to fermions in mass basis.
In the SM:
e
e
Indirect limit on BR(H→ℓ�) are loose O(10%)Stringent indirect limits on Ye from e→ BR(H→e)< 108, but with assumptions on NP contributions in the loop.
CMSPASHIG14040 Phys. Lett. B 749 (2015) 337
CMSPASHIG14040
arXiv:1508.03372
CMS H →
Luca Fiorini 9
● CMS analyzed e and had final states● Analysis employs categorization in 0jet, 1jet, 2jet final states● Using binned collinear mass spectrum for the statistical analysis.● Main backgrounds:
● Z → ● W+jets and with misidentified leptons
● Analysis exploits H → studies and different kinematics of the signal wrt backgrounds: harder muon spectrum, MET alignment with direction wrt backgrounds.
CMS H → mass distributions
Luca Fiorini 10
Leptonic e categories have in general more sensitivity than had categories.
CMS H → results
Luca Fiorini 11
Best fit BR(H→)=(0.84+0.390.37‐ )%
Observed (expected) 95% CL limit is BR(H→)<1.51% (<0.75%)
● Small excess observed in 3 out of 6 categories. Combined:
● 2.4 excess
ATLAS H →
Luca Fiorini 12
● ATLAS analyzed the had final states● Analysis employs 2 signal categories and 1 control region● Using binned MMC (missing mass calculator) spectrum for the statistical analysis.● Main backgrounds:
● W+jets main backgrond in SR1● Z → main background in SR2
SR1
SR2
WCR
ATLAS H → mass distributions
Luca Fiorini 13
SR1 is the most sensitive region. Large Z → contribution to SR2 allows to constraints true taus systematics
ATLAS H → results
Luca Fiorini 14
● No significant excess found (1.3σ). ● Observed (expected) 95% CL limit:BR(H→)<1.85 (1.24)%
● Best fit of BR(H → ) = (0.77 ±0.62)%
CMS H e→
Luca Fiorini 15
● CMS uses the same analysis strategy as H → ● Leptonic e are the most sensitive categories
No excess observed:95% CL limit is BR(H→e)<0.69%
CMS H e→
Luca Fiorini 16
● CMS uses unbinned fit of the e mass spectrum.● 11 categories (2 VBF + 3x3 barrel/endcap combination x number of jets), similar to H → and H → analyses● Background modeled by polynomial, exponential and power law (category dependent).● Signal modeled by the sum of two gaussians.
No excess observed: 95% CL limit is BR(H→e)<0.036%
FCNC t → u/c+H
Luca Fiorini 17
Modeling via anomalous couplings in effective Lagrangians
In the SM:
ATLAS and CMS analyzed ttbar decays with final states with H gg, H bb and “multileptons” final states → →( H ZZ, WW, → ) and t bW(→ →ℓ, qq).→
JHEP 1406 (2014) 008arXiv:1509.06047 arXiv:1509.06047
CMSPASTOP13017 CMSPASTOP14019 CMSPASTOP14020
ATLAS FCNC t u/c+H (multilep)→
Luca Fiorini 18
● Reinterpretation of ATLAS ttH analysis in multilepton final states● 8 different categories defined by lepton multiplicity and jet multiplicity:
● (ee, , e) x (4j,>=5j) ● 3 light leptons● 2 light leptons + 1 tau
The observed (expected) 95% CL upper limits on the branching ratios are:
● BR(t Hc→ ) < 0.79% (0.54%) ● BR(t Hu→ ) < 0.78% (0.57%),assuming BR(t → Hu), BR(t → Hc) =0 respectively
CMS FCNC t c+H→
Luca Fiorini 19
● Analysis of 3 light leptons or 2 same sign leptons events, selecting the final states ( H ZZ, WW, → ) and t bW(→ →ℓ) ●Two categories, no jet splitting:
● 2 light leptons SS ● 3 light leptons
The observed (expected) 95% CL upper limits on the branching ratio is:● BR(t Hc→ ) < 0.93% (0.89%)
ATLAS FCNC t u/c+H(→ )
Luca Fiorini 20
● 7 TeV and 8 TeV datasets are used.● Main backgrounds are diphoton nonresonant background and ttH● Analysis considers both hadronic and leptonic final states of the W decay:
● diphoton+jets (cut on mtop)● diphoton+lepton+jets (cut on mT(W) )
Observed (expected) 95% CL upper limits on the branching ratios:● BR(t Hq→ ) < 0.79% (0.51%)
CMS FCNC t u/c+H(→ )
Luca Fiorini 21
● 8 TeV datasets is used.● Similar approach to ATLAS analysis, considering both hadronic and leptonic final states of the W decay:
Observed (expected) 95% CL upper limits on the branching ratios:● BR(t Hu→ ) < 0.42% (0.65%) ● BR(t Hc→ ) < 0.47% (0.71%)
ATLAS FCNC t u/c+H(bb)→
Luca Fiorini 22
● Search for ttbar WbHq ( )b(bb)q→ → ℓν● Requiring one light lepton, >= 4 jets and >=2 bjets● 9 signal and bkgenriched event categories:
● (4j, 5j, ≥6j) x (2b, 3b, ≥4b)● main background is SM ttbar( WbWb)+jets→● Using Likelihood discriminant including mass constraints and btagging information for signal and bkg hypotheses.
Observed (expected) 95% CL upper limits on the branching ratios:● BR(t Hu) < 0.61% (0.64%) → BR(t Hc) < 0.56% (0.42%) →
ATLAS FCNC t u/c+H combination→
Luca Fiorini 23
● Combnation of ATLAS searches in H bb, H → → and multilep● BR(t uH) < 0.45% (0.29% exp)→● BR(t cH) < 0.46% (0.25% exp)→
● Best fit of BR(t uH) vs BR(t cH) → →compatible within ~1 with null hypothesis.
CMS FCNC t u/c+H(bb)→
Luca Fiorini 24
●Result released on the 19/11/2015● Requiring one light lepton, >=4 jets and >= 2bjets● Using Boosted decision Tree discriminant with kinematic variables of the Higgs and top candidates and and combining it in a NeuralNetwork Likelihood discriminant including btagging information of the jets.
Observed (expected) 95% CL upper limits on the branching ratios:● BR(t Hu) < 1.92% (0.85%) → BR(t Hc) < 1.16% (0.89%) →
Luca Fiorini 25
● The LHC experiments performed comprehensive measurements of the Higgs sector and couplings of the Higgs boson.
● Searching for LFV effects in the Yukawa couplings of the H125 boson are ongoing at LHC. ● No large lepton flavor violation observed.● A small deviation from null hypothesis for H → ( CMS: 2.4 , ATLAS: 1.3) and compatible with <~1% BR will be verified with LHC Run2 data.
● Limits set on Higgsmediated flavorchanging neutral currents in top decays at LHC below 0.3% for t cH and t uH.→ →● Mostly limited by statistics: stay tuned for updates with more luminosity at higher center of mass energy with Run2 data!
Summary
Luca Fiorini 27
Thanks for the attention
Bonus SlidesBonus Slides
Luca Fiorini
ATLAS Detector
Luca Fiorini 4
= 1.6
45 m
24 m
7000 Tons
ATLAS Collaboration 38 Countries 175 Institutions 3000 Scientific Authors total(~2000 with a PhD)
CMS Detector
Luca Fiorini 5
29 m
15 m
14000 Tons
CMS Collaboration 42 Countries 182 Institutions 3300 Scientific Authors total(~900 students)
Luca Fiorini 6
DAQ and Trigger
ATLAS L1/HLT rates: 75 kHz/400 Hz*
*: promptly reconstructed, ATLAS has 200 Hz more of deferred data.
Both experiments have significantly improved their DAQ and trigger systems for Run2.They are now able to reach >= 100 kHz at L1 and >= 1 kHz HLT ouput
600+300 (4x102+2x102)
(103)
Luca Fiorini 8
Computing and SimulationComputing and SimulationThe fast duty cycle of the LHC analyses is possible thanks to the Tier0
and GRID resources
● Just in 2012, both CMS and ATLAS experiments have produced 34 billions of MC events on the GRID and processed ~3 billions of data events at Tier0.
● On a single machine, it would require more than 15 thousands years (without considering user and group analyses, calibrations, reprocessings, ...).
●GRID is a crucial asset of the LHC experiments to provide physics results in a timely manner.
Luca Fiorini 9
Summary of SM results
Preliminary measurements of the crosssections down to few pb (~tens of fb in some cases if we include also the BR). Comparable to Higgs production total
● Good agreement with SM expectations within uncertainties.
● Experimental uncertainties are in some cases at the level of the theor. predictions
H@125 GeV total
Luca Fiorini 10
Higgs Mechanism Birth
● EW Gauge bosons in previous formulation of the SM were massless.● Four seminal papers in 1964 proposed a spontaneous symmetry breaking mechanism in relativistic gauge theory.
●The introduction of a complex scalar doublet allow to give mass to the W and Z bosons after symmetry breaking.● 3 deg. of freedom go in the longitudinal polarizations of the W± and Z● Remaining d.o.f. is a new scalar particle → the Higgs boson● Yukawa couplings to fermions was later introduced in the formulation of the SM● In SM all Higgs parameters are determined once the Higgs mass is known.
24/11/15
Higgs Potential (2)Higgs Potential (2)
Luca Fiorini
The minima of the potential are on a circumference of radius:
We rewrite the Lagrangian around a minimum:
The Lagrangian now becomes:
where the third and forth terms represent the self coupling of the Higgs field:
Luca Fiorini 5
top quark productionMain ttbar final states:
● Full hadronic 45%: 6 jets● High bkgs (mainly QCD)
● Semileptonic 30%: + MET+ 4 jetsℓ● (ℓ=e,)● Moderate bkgs (mainly W)
● Dileptonic 11%: 2 + MET+ 2 jetsℓ● (ℓ=e,)● Low bkgs (mainly Z+jets)
● t qH searches analyse ttbar final →states and tag the event requiring a second top in the event decaying to hadrons or leptons:
● t bW(qq')→● t bW(lnu)→
Luca Fiorini
Higgs prod. Rates at TeVatron Higgs prod. Rates at TeVatron
ggH is the dominant production mode.VH is the subleading production mode
78% 7% 15% <1%
H →
H → Analysis● Main background is Z → modelled by tauembedding of data Z → ● ATLAS employs MVA analysis. CMS cutbased analysis includes VH.
Luca Fiorini 20
arXiv:1501.4943
arXiv:14 01.5041
Analysis includes leptonic an hadronic decay channels of the taus. Events are divided in 0jet, 1jet and 2jet categories to enhance sensitivity to the VBF production mode and to ggF production of highly boosted Higgs bosons.
H → Results●Evidence of Higgs fermionic decays:
● Excess wrt expected background observed by both experiments● Highest significance among fermionic channels thanks to sensitivity to VBF process.
Luca Fiorini 21
arXiv:1501.4943
ATLAS Results Significance 4.5 (3.4) σ
CMS Results Significance 3.2 (3.7) σ
μ=0.78±0.27
arXiv:14 01.5041
H →H → Analysis BDT Analysis BDT
● Data is divided in 6 signal regions and 9 control region to simultaneously fit signal and backgs.Luca Fiorini 34
24/11/15
Higgs Mass widthHiggs Mass width
Luca Fiorini
● The Higgs width also depends on its mass value, spanning several orders of magnitude:
Width becomes equal to mass around 1.4 TeVFor mass >~1 TeV, the concept of Higgs resonance would disappear.
For mH=125 GeV, the width is about 4 MeV
For high mass, the width grows as
ttH (H>bb, WW, ZZ, , )● ATLAS and CMS covered broad range of Higgs boson final states and ttbar decay modes.● btagging and toptagging used to suppress backgrounds.● The analyses are characterised by large number of categories and control region.● CMS recently added a new search for single top tqH production.
Luca Fiorini 26
ATLAS CMS=2.8±1.0 @125.6 GeVLocal Significance: 3.4 (1.2)
=1.8±0.8 @125.36 GeVLocal Significance: 2.5 (1.5)
arXiv 140 8.1682
arXiv:1503.05066
Significance of Combined Measurements
Luca Fiorini 35
● Improved sensitivity thanks to the combination● Comparing likelihood of the bestfit with the μprod=0 and μdecay=0 hypothesis we obtain:
● VBF production and H → now established with more than 5 significance.σ● ggF and H ZZ,→ ,WW already established by individual experiments
V vs F Contour
Luca Fiorini 37
● Couplings are grouped: κV = κW = κZ ; κF = κt = κb= κτ● Assumptions:– gg → H and H → only through SM particles → only SM particles contribute to decay● All results in agreement with SM (κV = κF = 1) within 1σ
SM pvalue: 59%
V vs F and Couplings scaling
Luca Fiorini 36
● Measure ggF+ttH production and VBF+VH production for each decay mode (No assumption on SM production or decay rates needed for individual channels). ● Measurement of the combined ratio: μVBF+VH/μggF+ttH = 1.06 +0.35 0.27
● Within current precision Higgs couplings scale with particle masses
SM pvalue: 88%