sign in sign up
<<
Home , Lepton

Implications of lepton universality violation in B- decays

J. Martin Camalich

JLab Webinar

May 24th 2021 Why studying () Flavor in the 2020’s? 1 - Big Questions CKM and matrices: Parametrizations of flavor phenomena in the SM

The SM Flavor Puzzle

I Some quark seem “tuned” Anthropic arguments!

I Origin of CP-violation, , strong CP

I Why does need three generations at all?

However, the solution might not be accessible at the at reach. . .

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 2 / 28 2 - Precision physics programme Quark mixing in precision regime since decades ago! 15 I High-Luminosity LHC: Expected 10 bb¯ pairs !! HL-LHC Yellow Book A. Cerri et al., arXiv: 1812.07638

ca. 200 pages PDG

I 3 mixing angles and 1 CPV phase

I 6 quark masses

CKM: Cabibbo-Kobayashi-Maskawa paradigm

  1−λ2/2 λ Aλ3(ρ−iη)  2 2  VCKM '  −λ 1−λ /2 Aλ  Aλ3(1−ρ−iη) −Aλ2 1

λ = 0.2253(7), A = 0.808(22), ρ¯ = 0.132(22), η¯ = 0.341(13)

Jure Zupan, arXiv: 1903.05062

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 3 / 28 2 - Precision physics programme Quark mixing in precision regime since decades ago! 15 I High-Luminosity LHC: Expected 10 bb¯ pairs !! HL-LHC Yellow Book A. Cerri et al., arXiv: 1812.07638

ca. 200 pages PDG

I 3 mixing angles and 1 CPV phase

I 6 quark masses

CKM: Cabibbo-Kobayashi-Maskawa paradigm

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 3 / 28 Very sensitive probes of “generic” new physics

Theorem: FCNC are “Loop-Suppressed” in the SM!

I Prototypical example: FCNCs

Loop Weak z }| { Flavor z}|{ y 2 z }| { M ∼ G t (V ∗V )2 SM F 16π2 ts tb | {z } GIM

I Flavour observables probe (indirectly) very high scales!

M. Neubert at EPS 2011

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 4 / 28 Effective field theories and bottom-up approach to new physics

Legacy of Flavor Physics Program

I High precision measurements of SM mixing parameters

I Discover New Physics OR set the lower bounds on ΛFlavor

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 5 / 28 Flavor physics was instrumental in discovering and shaping the SM

I Nuclear β-decays: Discovery weak interactions and

I Rare -decays: Discovery of the

I Kaon decays: Discovery of CP violation =⇒ Discovery of 3 generations

Expect the unexpected

Thanks to Zoltan Ligeti for sharing this

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 6 / 28 3 - Flavor anomalies!

Gauge interactions are generation independent!

Hints for Lepton Flavor Universality Violation (LFUV)?

“The flavour of new physics”, JMC & J. Zupan, CERN courier, May/June 2019

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 7 / 28 Flavor Anomalies: b s`+`− →

Leptons do not feel the strong !

B(B¯→K (∗)µ+µ−) SM R (∗) = ' 1 K B(B¯→K (∗)e+e−)

Deficits on RK (∗) measured by LHCb

Signal of µ-to-e LFUV at 2-4σ

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 8 / 28 Flavor Anomalies: b s`+`− →

Leptons do not feel the strong force!

B(B¯→K (∗)µ+µ−) SM R (∗) = ' 1 K B(B¯→K (∗)e+e−)

Deficits on RK (∗) measured by LHCb

Other anomalies in branching ratios and angular observables

Claims of > 5σ signal of NP in global fit to ca. 100+ observables

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 8 / 28 Our previous analyses of b s`` → 1 Anatomy and model-independent parametrization of hadronic uncertainties Beneke et al. NPB592(2001)3, Jäeger and JMC JHEP1305(2013)043, PRDD93(2016)1,014028 0 I Were at the center of “P5 controversies” (hadronic effects vs. NP) Bharucha et al.JHEP 1009 (2010) 090, Descotes-Genon et al. JHEP 1412 (2014) 125, Lyon et al. arXiv:1406.0566, Ciuchini et al.

arXiv:1512.07157 + many, many more papers. . .

2 First theoretical analysis of the first RK measurement

3 Among first combined theoretical analyses of RK (∗)

Emphasize lepton-universality ratios and Bs → µµ to discover NP

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 9 / 28 Announcement of new LHCb results on RK and Bs → µµ!

Quick (last-second) reaction to analyse the new results

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 10 / 28 Announcement of new LHCb results on RK and Bs → µµ!

Collaboration between:

I Beihang University: Li-Sheng Geng, Shuang-Yi Li and Rui-xiang Shi

I UC San Diego: Benjamín Grinstein

I U. Sussex: Sebastian Jäger

I Instituto de Astrofísica de Canarias: JMC Other related work appeared around the same time:

Hiller et al. 2103.12724 [hep-ph], Angelescu et al. 2103.12504 [hep-ph], Altmannshofer et al. 2103.13370 [hep-ph], Cornella et al. 2103.16558

[hep-ph], Greljo et al. 2103.13991 [hep-ph], . . .

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 10 / 28 Effective field theory approach to b s`` decays → CC (Fermi theory):

bLLW s ∗ µ ⇒ GF Vcb Vcs C2 c¯Lγ bL s¯LγµcL c c FCNC: b s bRLW s RL t W e t t ⇒ ∗ ¯ σ µν 2 GF Vtb Vts mb C7 sL µν bR F γ γ 4π

b s b s b s LLW LLt LLt W t t W ∗ α µ ¯ Z, γ Z, γ ⇒ GF Vtb Vts C9(10) s¯Lγ bL `γµ(γ5)` ν 4π l l l l l l

0 ¯ I New-Physics also in Ci or e.g. Oi obtained PL → PR in sLb

l

l I Light fields active at long distances b s Nonperturbative QCD!

B K∗ F Factorization of scales mb vs. ΛQCD HQEFT, QCDF, SCET,.. .

d

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 11 / 28 A beautiful example: B0 `` q →

Marco Santimaria @ CERN, March 23rd 2021

b

Z l Bq

l q

2 2 ∗ 2 2 2 0 2 Bs` ∝ GF α |VtbVts | m` fBs |C10 − C10|

Semileptonic decay constants fBq can be calculated in LQCD FLAG averages SM −9 Include radiative corrections: Bsµ = 3.66(14) × 10 Beneke et al. JHEP10(2019)232

∼ 2 − 3σ deficit with respect to SM predictions!

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 12 / 28 A beautiful example: B0 `` q →

Marco Santimaria @ CERN, March 23rd 2021

2 2 ∗ 2 2 2 0 2 Bs` ∝ GF α |VtbVts | m` fBs |C10 − C10|

Semileptonic decay constants fBq can be calculated in LQCD FLAG averages SM −9 Include radiative corrections: Bsµ = 3.66(14) × 10 Beneke et al. JHEP10(2019)232

∼ 2 − 3σ deficit with respect to SM predictions!

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 12 / 28 A rich (and complex) example: B K ∗( K π)`` → →

d(4)Γ 9 s 2 c 2 2 = 32π (I1 sin θk +I1 cos θk dq d(cos θl )d(cos θk )dφ s 2 c 2 2 2 + (I2 sin θk +I2 cos θk ) cos 2θl +I3 sin θk sin θl cos 2φ

2 + I4 sin 2θk sin 2θl cos φ+I5 sin 2θk sin θl cos φ+I6 sin θk cos θl

2 2 + I7 sin 2θk sin θl sin φ+I8 sin 2θk sin 2θl sin φ+I9 sin θk sin θl sin 2φ) Anomalies in the angular observables . . .

I Blurred by hadronic uncertainties 0 I5 ATLAS, JHEP 10 (2018) 047 P5 = √ 2 −I2sI2c

I Cancel leading theory uncertainties

New physics? µ δC9 ' −1

Descotes-Genon et al. PRD88,074002

Most precise measurement: LHCb, PRL125(2020)011802

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 13 / 28 Large-recoil region (low q2)

I No LQCD (Sum Rules, models ...) and QCDf and SCET (power-corrections)

I Dominant effect of the pole Charmonium region

I Dominated by long-distance (hadronic) effects 2 2 I Starting at the perturbative cc¯ threshold q ' 6 − 7 GeV Low-recoil region (high q2)

I LQCD+HQEFT + OPE (duality violation)

I Dominated by semileptonic operators

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 14 / 28 Anatomy of the amplitude in a nutshell

Jäger and JMC, PRD93 (2016) no.1, 014028 Helicity amplitudes λ = ±1, 0

eff C9 z }| { n  m2  mˆ m o H (λ) = −iN C V˜ + B h − b B C T˜ , V 9 Lλ q2 λ q2 7 Lλ

˜ HA(λ) = −iNC10VLλ

Hadronic form factors: 7 independent q2-dependent nonperturbative functions

“Charm” contribution

Z 4 iq·y ¯ ∗ em,had,µ ¯ hλ ∝ d ye hK |T {j (y), O1,2(0)}|Bi

Charm and O9 are tied up by renormalization eff Only C9 is observable!

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 15 / 28 The lepton-universality ratios

Leptons do not feel the strong force!

B(B¯→K (∗)µ+µ−) SM R (∗) = ' 1 K B(B¯→K (∗)e+e−)

Very clean null tests of the SM Sensitive to muonic LH currents!

I Th. uncertainties negligible! Geng, Grinstein, Jäger, JMC, Ren, Shi, PRD96(2017)093006

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 16 / 28 Analyzing the data Frequentist statistics: 2 ~ 2 ~ 2 χ˜ (C, ~y) = χexp(C, ~y) + χth(~y)

~ I C vector of Wilson coefficients I ~y vector of 27 hadronic parameters

F From LQCD predictions to NDA estimates for values and errors of ~y

 2 2 X yi − y¯i χ (~y) = th δy i i Profile over all hadronic uncertainties

χ2(C~ ) = min χ˜2(C~ , ~y) ~y

µ µ We perform 2 fits to only C9 and/or C10

I Clean Fit: Only RK and RK ∗ and Bs → µµ ∗ I Global Fit: Add 94 measurements of angular observables of B → K µµ 2 I Data not included: Absolute BRs nor high-q (testing SM or nonperturbative QCD?)

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 17 / 28 The theoretically-clean fit

Assume NP is µ-specific (1σ,3 σ and5 σ contours) 4

R (*) 3 K

2 Bs → μ μ 1

μ 10 ★★

δ C 0

-1 Result in 2017 Gaussian in 2021 -2

-3 -4 -3 -2 -1 0 1 2 3 μ δC9

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 18 / 28 The theoretically-clean fit Assume NP is µ-specific (1σ,3 σ and5 σ contours) 4

R (*) 3 K

2 Bs → μ μ 1

μ 10 ★★

δ C 0

-1 Result in 2017 Gaussian in 2021 -2

-3 -4 -3 -2 -1 0 1 2 3 μ δC9 Significant changes since 2017

I SM in tension with data at4 .0σ (p-value= 7.6 × 10−5) from3 .5σ in 2017

I 2D fit performs better than SM at4 .6σ from3 .8σ µ µ I C10 and CL perform better than SM at ∼ 5σ

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 18 / 28 The global fit

Assume NP is µ-specific (1σ and3 σ contours) 2.0

1.5

1.0

μ 10 0.5

δ C ★★ 0.0 Clean Fit -0.5 Global Fit

-1.0 -2 -1 0 1 2 μ δC9

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 19 / 28 The global fit Assume NP is µ-specific (1σ and3 σ contours) 2.0

1.5

1.0

μ 10 0.5

δ C ★★ 0.0 Clean Fit -0.5 Global Fit

-1.0 -2 -1 0 1 2 μ δC9

Significant changes since 2017

I Best fit shifts towards more C9 or more LH

I 2D fit performs better than SM at4 .6σ from4 .2σ µ I CL perform better than SM at ∼ 5σ

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 19 / 28 Current-current interpretation!

C9(10) µ Leff ⊃ 2 (s¯γ PLb)(¯µγµ(γ5)µ) ΛNew−Physics ΛNew−Physics ∼ 30 TeV!!!

UV completions: Z 0’s and

Extra bounds from low energy e.g. Z 0 ¯ I Bs-Bs mixing I Neutrino trident production

b s

Z′

s¯ ¯b

0 0 I Requires small Z bs coupling! (e.g MFV) I Controls Z µµ coupling!

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 20 / 28 Current-current interpretation!

C9(10) µ Leff ⊃ 2 (s¯γ PLb)(¯µγµ(γ5)µ) ΛNew−Physics ΛNew−Physics ∼ 30 TeV!!!

I “Colored” gauged Lµ − Lτ Altmannshofer et al. PRD89(2014)095033

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 20 / 28 Current-current interpretation!

C9(10) µ Leff ⊃ 2 (s¯γ PLb)(¯µγµ(γ5)µ) ΛNew−Physics ΛNew−Physics ∼ 30 TeV!!!

0 − + I “Agnostic” simplified Z models: pp → µ µ X LHC and beyond...

Allanach et al. JHEP1903(2019)137

Could be far from production @ LHC (or FCC!!)

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 20 / 28 What’s next? Many ongoing analyses at LHCb

∗ I RK , Rφ, RΛb ,. . . 2 I High q

I LFUV ratios or differences of angular observables!

LHCb may very soon exceed 5σ with clean observables!

Need independent experimental confirmation!: Belle 2

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 21 / 28 Connecting to LFUV anomalies in b cτν decays →

B(B¯→D(∗)τ −ν¯) R (∗) = where ` = e, µ D B(B¯→D(∗)`−ν¯)

Babar, Belle and LHCb are independently in tension with SM!

I World average now at ∼ 3σ from SM. . .

“Naturally” produced by same operator!

I In the SU(2) × U(1) (SM)EFT

Q(3) = 1 (Q¯ i γµ~τQ j ) · (L¯ k γ ~τL l ) `q(ijkl) Λ2 L L L µ L 2 2 I Assume hyerarchical couplings ∼ mτ /mµ

ΛNew−Physics ∼ 3 TeV If true, it can be discovered at LHC!

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 22 / 28 Connecting to LFUV anomalies in b cτν decays →

B(B¯→D(∗)τ −ν¯) R (∗) = where ` = e, µ D B(B¯→D(∗)`−ν¯)

The vector lepqtoquark U1 = (3, 1)2/3 best possible solution

I Should be discovered at the LHC!

Buttazzo et al. ’17

Cornella et al. arXiv: 2103.16558 [hep-ph]

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 22 / 28 Conclusions

“Extraordinary claims require Extraordinary evidence” – C. Sagan

1 Theoretically-clean b → s`` observables point to NP! −5 I Data is at odds with the SM at 4.0σ or p − value = 7.6 × 10

I Left-handed scenarios and with axial µ-couplings favored over the SM @ ∼ 5σ I Global fits favor left-handed. Interpretation sensitive to hadronic uncertainties ...

F Interplay between hadronic effects and C9 2 Discovery of NP?

I Including future LHCb theoretically-clean data may very soon consolidate & 5σ I Confirmation/refutation should come from Belle 2 in a “few” years I “Natural” connection to b → cτν anomalies!

F New can be directly discovered at ATLAS and CMS F Shed light on Flavor Puzzle?

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 23 / 28 Description at low q2 using QCD factorization

Heavy-quark and large-recoil (K ∗) limit only 2 independent “soft form factors” 2E T+ = V+ = 0, T− = V− = ξ⊥, T0 = V0 = S = ξk mB

Dugan et al. PLB255(1991)583, Charles et al. PRD60(1999)014001

0 The observable P5 Matias et al.’12

 ( eff 2 2 mb mB eff C C + C C ,⊥=C (q )+ C 0 I5 10 9,⊥ 9,k 9 9 q2 7 P5|∞ = √ ' , q eff 2 2 mb E eff 2 −I2sI2c 2 2 2 2 C9,k=C (q )+ C7 (C9,k + C10)(C9,⊥ + C10) 9 q2

For the charm Beneke, Feldmann&Seidel, NPB612(2001)25

+ − ¯ ∗ ¯ h` ` Ka |Hw|Bi=Caξa+ΦB ⊗Ta⊗ΦK ∗ +O(ΛQCD/mb) Below cc¯ threshold! q2 ≤ 6 GeV2

No model-independent treatment for Λ/mb corrections

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 24 / 28 Robustness against hadronic uncertainties

Model-independent approach: 27 hadronic parameters

I Treat them as nuisance parameters and obtain profile likelihoods

I Use gaussian or flat priors (R-fit)

I Study the impact in the fit of increasing theoretical errors

I Repeat for ×2 and ×3 th. errors

I Theoretically-clean fit I Global fit 2.0 2.0

1.5 1.5

1.0 1.0 μ 10 μ 10 0.5 ★★ 0.5 δ C

δ C ★★ 0.0 0.0 Clean R- Fit in 2021 R- Fit in 2021 Clean R- Fit x2 in 2021 R- Fit x2 in 2021 Clean R- Fit x3 in 2021 -0.5 R- Fit x3 in 2021 -0.5 Clean gaussian in 2021 Clean gaussian in 2021 Gaussian in 2021 -1.0 -1.0 -2 -1 0 1 2 -2 -1 0 1 2 μ μ δC9 δC9

Theoretically-clean results robust w.r.t. theoretical uncertainties

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 25 / 28 Fitting strategy and summary new data since 2017 The theoretically-clean fit

I Only observables with controlled theoretical uncertainties: Bs → µµ and RK (∗)

I Major changes since 2017

F New measurements of Bs → µµ

expt SM R = Bsµ /Bsµ

R = 0.83(16)(1σ) ⇒ R = 0.78(9)(2.4σ)

−1 F New measurement of RK (9 fb ) R = 0.745(90)(2.4σ) ⇒ R = 0.846(44)(3.1σ)

F Belle measurements of R K (∗) The global fit

(0) I Add angular observables Pi : 94 data 0,+ ∗0,+ + − F New LHCb data on B → K µ µ LHCb, PRL125(2020)011802 and 2012.13241 [hep-ex]

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 26 / 28 The RK measurement (my interpretation)

LHCb, PRL122(2019)191801, arXiv: 2103.11769 [hep-ex], K. Petridis @ CERN March 23rd 2021 Different behavior of µ and e in detector

I : Energy losses through bremsstrahlung

I Reconstruct energy by identifying clusters in ECAL

Degraded -energy resolution due to reconstruction

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 27 / 28 + + + + LHCb strategy: Divide by 1 ' rJψ = B(B → K J/ψ(µµ))/B(B → K J/ψ(ee))

I Supress much of the systematic uncertainties

First cross-check: Measure Rψ(2S)

I They obtain Rψ(2S) = 0.997(11) (world-leading test of LFU in ψ(2S) → ``)

Second cross-check: Check efficiencies by rJ/ψ ' 1 in different kinematics

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 28 / 28 + + + + LHCb strategy: Divide by 1 ' rJψ = B(B → K J/ψ(µµ))/B(B → K J/ψ(ee))

I Supress much of the systematic uncertainties

Final LHCb result of RK with full run 1 and run 2 data sets

I Twice as much data as 2019 analysis! 3 fb−1 7,8 TeV 5 fb−1 (+2 fb−1 13 TeV) 9 fb−1 (+4 fb−1 13 TeV) PRL113(2014)151601 PRL122(2019)191801 arXiv: 2103.11769 +0.090 +0.060+0.016 +0.042+0.013 0.745−0.074 ± 0.036 0.846−0.054−0.014 0.846−0.039−0.012 √ I Expected 2 reduction in stat., amusingly identical c.v. Blind analysis

arXiv:2103.11769

J. Martin Camalich (IAC&ULL) Implications of LUV in B-meson decays May 24th 2021 28 / 28