RG flow of the Higgs potential. Holger Gies

Transcription

1 RG flow of the Higgs potential Holger Gies Helmholtz Institute Jena & TPI, Friedrich-Schiller-Universität Jena & C. Gneiting, R. Sondenheimer, PRD 89 (2014) , [arxiv: ], & S. Rechenberger, M.M. Scherer, L. Zambelli, EPJC 73 (2013) 2652 [arxiv: ] & R. Sondenheimer, arxiv: ; L. Zambelli, arxiv:14xx.yyyy Strong interactions in the LHC era

2 Search for the Higgs boson 4 Jul ATLAS & 14 Mar 2013, CERN press release:... the new particle is looking more and more like a Higgs boson... CMS 12 : ± 0.4(stat) ± 0.5(sys)GeV, ATLAS 12 : ± 0.4(stat) ± 0.4(sys)GeV Success of Experiment & Theory (ANDERSON 62; BROUT,ENGLERT 64; HIGGS 64; GURALNIK,HAGEN,KIBBLE 64)

3

4 Numbers matter standard model best before: Λ = M Planck?

5 Validity range of the standard model Λ: UV cutoff SM as effective theory scale of maximum UV extension Λ? scale of new physics: Λ NP Λ

6 Higgs boson mass and maximum validity scale upper bound m H λ R v 2 lower bound (HAMBYE,RIESSELMANN 97) SM: e.g. (KRIVE,LINDE 76; MAIANI,PARISI,PETRONZIO 78; KRASNIKOV 78; POLITZER, WOLFRAM 78; HUNG 79; LINDNER 85; WETTERICH 87; SHER 88; FORT,JONES,STEPHENSON, EINHORN 93; ALTARELLI,ISIDORI 94; SCHREMPP,WIMMER 96;... ; HOLTHAUSEN,LIM,LINDNER 09;... ) BSM: e.g., (CABBIBO, MAIANI,PARISI,PETRONZIO 79; ESPINOSA,QUIROS 91;... )

7 Lower bound of Higgs boson mass vacuum stability / meta-stability bound effective potential á la Coleman Weinberg: U eff (φ) = 1 2 µ2 φ λ(φ)φ4 e.g., λ(φ) from RG-improved perturbation theory: t λ = 3 4π 2 ( ht 4 + ht 2 λ + 1 [ 2g 4 + (g 2 + g 2 ) 2] 1 ) 16 4 λ(3g2 + g 2 ) + λ 2

8 Lower bound of Higgs boson mass effective potential á la Coleman Weinberg: U eff (φ) = 1 2 µ2 φ λ(φ)φ4 e.g., λ(φ) from RG-improved perturbation theory: t λ = 3 4π 2 ( ht 4 + ht 2 λ + 1 [ 2g 4 + (g 2 + g 2 ) 2] 1 ) 16 4 λ(3g2 + g 2 ) + λ 2 (GABRIELLI ET AL. 13)

9 Lower Bound of Higgs boson mass meta-stability: tunneling time > age of universe Near critical standard model: (BUTTAZZO ET AL. 13) NNLO calculation (DEGRASSI ET AL. 12) earlier calculations, e.g., (ISIDORI,RIDOLFI,STRUMIA 01)

10 Lower Bound of Higgs boson mass Near critical standard model: (BUTTAZZO ET AL. 13: UPDATE V4) Stability seems to prefer m H 130GeV or m top 171GeV

11 Lower Bound of Higgs boson mass Near critical standard model: (BUTTAZZO ET AL. 13: UPDATE V4) Stability seems to prefer m H 130GeV or m top 171GeV The Higgs potential has the worrisome feature that it might become metastable at energies above 100bn gig-electron-volts,

12 2nd thoughts on the lower bound True minimum of U eff (φ) at φ GeV > M Pl? UV IR RG flow?

13 2nd thoughts on the lower bound True minimum of U eff (φ) at φ GeV > M Pl? UV IR RG flow? simple top-higgs Yukawa model: lattice simulation vs. 1-loop PT with cutoff vs. 1-loop Λ-removed PT (HOLLAND,KUTI 03; HOLLAND 04) Criticism: to few scales? (EINHORN,JONES 07)

14 Z 2 symmetric model S int = ihφ ψψ Top-Higgs Yukawa models chiral model S int = i h t ( ψ L φ C t R + t R φ C ψ L) includes relevant top quark + Higgs field (+ largest Yukawa coupling) discrete model no Goldstone bosons (as in SM) gauged models: (POSTER: R. SONDENHEIMER)

15 Top-Higgs Yukawa models generating functional: Z [J] = DφD ψdψe S[φ, ψ,ψ]+ Jφ Λ = Dφ e S[φ, ψ,ψ] S F,Λ [φ]+ Jφ Λ (HOLLAND, KUTI 03; HOLLAND 04) top-induced effective potential U F (φ) = 1 2Ω ln det Λ( 2 + ht 2φ φ) det Λ ( 2, ) CAVE: cutoff Λ / regularization dependent

16 Top-induced effective potential exact results for fermion determinants for homogeneous φ e.g., sharp cutoff: U F,t (φ) = Λ2 8π 2 h2 t φ 2 }{{} <0 (mass-like term) π 2 [ ) )] ht 4 φ 4 ln (1 + Λ2 ht 2 + h 2 φ 2 t φ 2 Λ 2 Λ 4 ln (1 + h2 t φ 2 Λ 2 }{{} >0 (interaction part) mass-like term: contributes to χsb = v 246GeV interaction part: strictly positive = cannot induce instability for any finite Λ (HG,SONDENHEIMER 14)

17 Rederiving the instability try to send Λ : U F,t (φ) = Λ2 8π 2 h2 t φ ( 16π 2 h4 t φ [ln 4 Λ2 h 2 ht 2 + const. + O t φ 2 )] φ 2 Λ 2

18 Rederiving the instability try to send Λ : U F,t (φ) = Λ2 8π 2 h2 t φ ( 16π 2 h4 t φ [ln 4 Λ2 h 2 ht 2 + const. + O t φ 2 )] φ 2 Λ 2 renormalization: trade (Λ, m Λ, λ Λ ) for (µ, v, λ v )? U F,t (φ) 1 ) 16π 2 h4 t φ (ln 4 h2 t φ 2 µ 2 + const.

19 Rederiving the instability try to send Λ : U F,t (φ) = Λ2 8π 2 h2 t φ ( 16π 2 h4 t φ [ln 4 Λ2 h 2 ht 2 + const. + O t φ 2 )] φ 2 Λ 2 renormalization: trade (Λ, m Λ, λ Λ ) for (µ, v, λ v )? U F,t (φ) 1 ) 16π 2 h4 t φ (ln 4 h2 t φ 2 µ 2 + const. instability occurs beyond h2 t φ 2 Λ 2 > 1 (HG,SONDENHEIMER 14) similar problems for other reg s implicit renormalization conditions would violate unitarity (HOLLAND, KUTI 03; HOLLAND 04) (BRANCHINA,FAIVRE 05; GNEITING 05)

20 Contradiction? t λ = 3 4π 2 h4 t +... = λ < 0 interaction part of U F,t (φ) > 0 strictly positive

21 Contradiction? t λ = 3 4π 2 h4 t +... = λ < 0 U eff (φ) 1 2 λ(φ)φ4 interaction part of U F,t (φ) > 0 strictly positive

22 Contradiction? t λ = 3 4π 2 h4 t +... = λ < 0 U eff (φ) 1 2 λ(φ)φ4 interaction part of U F,t (φ) > 0 strictly positive ( h 2 t CAVE: O )-terms φ 2 Λ 2 for finite Λ

23 Summary, Part I no in-/meta-stability from top (fermion) fluctuations... if cutoff Λ is kept finite but arbitrary no in-/meta-stability at all? U eff (φ) = U Λ (φ) + U B (φ) + U F (φ)

24 Summary, Part I no in-/meta-stability from top (fermion) fluctuations... if cutoff Λ is kept finite but arbitrary no in-/meta-stability at all? U eff (φ) = U Λ (φ) + U B (φ) + U F (φ) }{{}}{{} arbitrary generically stable... in-/meta-stabilities from the bare action/uv completion lower Higgs mass bounds? = nonperturbative methods recommended if not needed extensive lattice simulations: (FODOR,HOLLAND,KUTI,NOGRADI,SCHROEDER 07) constraining 4th generations: implications for dark matter models: (GERHOLD,JANSEN ) (GERHOLD,JANSEN,KALLARACKAL 10; BULAVA,JANSEN,NAGY 13) (EICHHORN, SCHERER 14) (POSTER: M.M. SCHERER)

25 Higgs boson mass bounds as a UV to IR mapping Λ

26 Higgs boson mass bounds as a UV to IR mapping Λ

27 Higgs boson mass bounds as a UV to IR mapping Λ

28 Higgs boson mass bounds as a UV to IR mapping Λ

29 Higgs boson mass bounds as a UV to IR mapping Λ

30 Higgs boson mass bounds as a UV to IR mapping Λ QFT + RG

31 Higgs boson mass bounds as a UV to IR mapping Λ QFT + RG = consistency bounds

32 Higgs boson mass bounds as a UV to IR mapping microscopic action at cutoff Λ: (HG,GNEITING,SONDENHEIMER 13) S Λ = S Λ (m 2 Λ, λ Λ, λ 6,Λ,..., h Λ,... ) = RG: mapping to IR observables RG = v 246GeV, m top 173GeV, m H = m H [S Λ ] e.g, renormalizable (?) UV bare potential U Λ (φ) = 1 2 m2 Λφ λ Λφ 4, λ Λ 0 trade: m Λ, h Λ v, m top = m H = m H (λ Λ, Λ)

33 Nonperturbative tool: functional RG (WETTERICH 93) t Γ k = 1 2 Tr tr k (Γ (2) k + R k ) 1 RG trajectory: Γ k =Λ = S Λ = 1 2 ( φ)2 + U Λ (φ) +...

34 Nonperturbative tool: functional RG (WETTERICH 93) RG trajectory: t Γ k = 1 2 Tr tr k (Γ (2) k + R k ) 1

35 Nonperturbative tool: functional RG (WETTERICH 93) RG trajectory: t Γ k = 1 2 Tr tr k (Γ (2) k + R k ) 1

36 Nonperturbative tool: functional RG (WETTERICH 93) t Γ k = 1 2 Tr tr k (Γ (2) k + R k ) 1 RG trajectory: Γ k 0 = Γ = U eff +...

37 Higgs boson mass bounds from functional RG Z 2 Yukawa model + φ 4 bare potential (HG,GNEITING,SONDENHEIMER 13) Systematic derivative expansion: ( ) Γ k = d d Zφk x 2 µφ µ φ + U k (φ 2 ) + Z ψk ψi / ψ + ihk φ ψψ m H GeV GeV λ Λ = 0

38 Higgs boson mass bounds from functional RG Z 2 Yukawa model + φ 4 bare potential (HG,GNEITING,SONDENHEIMER 13) Systematic derivative expansion: ( ) Γ k = d d Zφk x 2 µφ µ φ + U k (φ 2 ) + Z ψk ψi / ψ + ihk φ ψψ m H GeV GeV λ Λ = 0 λ Λ = 0.1

39 Higgs boson mass bounds from functional RG Z 2 Yukawa model + φ 4 bare potential (HG,GNEITING,SONDENHEIMER 13) Systematic derivative expansion: ( ) Γ k = d d Zφk x 2 µφ µ φ + U k (φ 2 ) + Z ψk ψi / ψ + ihk φ ψψ 800 m H GeV λ Λ = 0 λ Λ = 0.1 λ Λ = 1 GeV

40 Higgs boson mass bounds from functional RG Z 2 Yukawa model + φ 4 bare potential (HG,GNEITING,SONDENHEIMER 13) Systematic derivative expansion: ( ) Γ k = d d Zφk x 2 µφ µ φ + U k (φ 2 ) + Z ψk ψi / ψ + ihk φ ψψ 800 m H GeV λ Λ = 0 λ Λ = 0.1 λ Λ = 1 λ Λ = 10 GeV

41 Higgs boson mass bounds from functional RG Z 2 Yukawa model + φ 4 bare potential (HG,GNEITING,SONDENHEIMER 13) Systematic derivative expansion: ( ) Γ k = d d Zφk x 2 µφ µ φ + U k (φ 2 ) + Z ψk ψi / ψ + ihk φ ψψ 800 m H GeV λ Λ = 0 λ Λ = 0.1 λ Λ = 1 λ Λ = 10 λ Λ = 100 GeV

42 Higgs boson mass bounds from functional RG Z 2 Yukawa model + φ 4 bare potential (HG,GNEITING,SONDENHEIMER 13) Systematic derivative expansion: ( ) Γ k = d d Zφk x 2 µφ µ φ + U k (φ 2 ) + Z ψk ψi / ψ + ihk φ ψψ 800 m H GeV λ Λ = 0 λ Λ = 0.1 λ Λ = 1 λ Λ = 10 λ Λ = 100 GeV = conventional lower bound for λ Λ = 0 ( mean-field result) agreement with lattice (HOLLAND 04; FODOR,HOLLAND,KUTI,NOGRADI,SCHROEDER 07; GERHOLD,JANSEN )

43 Conventional lower Higgs boson mass bound chiral top-bottom-higgs Yukawa model (+Goldstone decoupling): (HG,SONDENHEIMER 14) mhgev GeV FRG: NLO derivative expansion λ 2Λ = 0 λ 2Λ = 1 λ 2Λ = 10 λ 2Λ = 100 = conventional lower bound close to Z 2 model:... bottom quark has little quantitative influence

44 General microscopic actions S Λ is a priori unconstrained. Consider, e.g., U Λ = λ 1Λ 2 φ2 + λ 2Λ 8 φ4 + λ 3Λ 48 φ6 (HG,GNEITING,SONDENHEIMER 13) for λ 3Λ > 0 we can choose λ 2Λ < 0:

45 General microscopic actions S Λ is a priori unconstrained. Consider, e.g., U Λ = λ 1Λ 2 φ2 + λ 2Λ 8 φ4 + λ 3Λ 48 φ6 (HG,GNEITING,SONDENHEIMER 13) mhgev for λ 3Λ > 0 we can choose λ 2Λ < 0: GeV λ 3Λ = 0, λ 2Λ = 0 λ 3Λ = 3, λ 2Λ = 0.08 lower bound relaxed = consistency bound < conventional bound string models with λ < 0: (HEBECKER,KNOCHEL,WEIGAND 13)

46 Renormalizable field theories seeming contradiction with common wisdom...?... observables are determined by renormalizable operators... y-axis: m H observable, x-axis: Λ?

47 RG mechanism for lowering the lower bound g i g 2 Theory Space g 1

48 RG mechanism for lowering the lower bound g i Gaußian fixed point g 2 Theory Space g 1

49 RG mechanism for lowering the lower bound g i renormalizable operators span a hyperplane g 2 Theory Space g 1

50 RG mechanism for lowering the lower bound g i non-renormalizable operators are exponentially damped towards the IR g 2 Theory Space g 1

51 RG mechanism for lowering the lower bound g i non-renormalizable operators are exponentially damped towards the IR BUT: RG time (scales) is needed for this damping g 2 Theory Space g 1

52 Consistency bounds from generalized bare actions e.g., U Λ = λ 1Λ 2 φ2 + λ 2Λ 8 φ4 + λ 3Λ 48 φ6 (HG,GNEITING,SONDENHEIMER 13) mhgev GeV = consistency bound shifted Λ axis λ 3Λ = 0, λ 2Λ = 0 λ 3Λ = 3, λ 2Λ = 0.08

53 Lowering the lower Higgs boson mass bound chiral model with λ 6,Λ (φ φ) 3 interaction comparison with lattice data: mhgev = RG mechanism confirmed GeV (HG,SONDENHEIMER 14) (HEDGE,JANSEN,LIN,NAGY 13) 10 GeV at Λ GeV O(1) variations of bare λ n,λ : m H 5 GeV at Λ GeV 2 GeV at Λ GeV (POSTER: R. SONDENHEIMER)

54 Summary, Part II Consistency bounds on the Higgs boson mass (or any other physical IR observable) arise from a mapping S micro O phys... provided by the RG For effective quantum field theories (with a cutoff Λ): bounds on O phys = f [S Λ ]... full S Λ not just the renormalizable operators lowering the conventional lower Higgs boson mass bound is possible... without in-/meta-stable vacuum

55 TODO list Λ study of general S Λ evolution of several minima instabilities?

56 Implications if m H < conventional lower bound: new physics at lower scales first constraints on underlying UV completion if m H exactly on the convenional lower bound: (e.g. if m top 171GeV) underlying UV completion has to explain absence of higher dimensional operators... criticality flat interaction potential

57 Candidates Asymptotically free YM-Higgs-Yukawa models? possible, pheno-viable models?? general perturbative UV prediction: (REV.: CALLAWAY 88) λ g analysis relies on the deep Euclidean region standard model + asymptotically safe gravity (WEINBERG 76; REUTER 96) gravity fluctuations induces a UV fixed point λ 0 (PERCACCI ET AL 03 09) = m H put onto conventional lower bound (WETTERICH,SHAPOSHNIKOV 10) (BEZRUKOV,KALMYKOV,KNIEHL,SHAPOSHNIKOV 12) asymptotically safe ( free) particle physics models? (SHROCK 13 14) = guaranteed in gauged Yukawa models (LITIM,SANNINO 14) (TALKS: R. SHROCK, D. LITIM)

58 Asymptotically free UV gauge scaling solutions? (RECHENBERGER,SCHERER,HG,ZAMBELLI 13; HG,ZAMBELLI 14IP) (Almost) flat potentials: = large amplitude fluctuations If flatness is driven by asymptotically free gauge sector: gauge rescaling of fields: X = g 2P Z φ φ 2 = solution of fixed-point equation for effective potential (P = 1): SU(2) Yang-Mills-Higgs : ( ) 2 [ ( )] 3 X V (X) = ξx 2 2X + X 2 ln 16π 2 + X = Coleman-Weinberg type, one-parameter family ξ k 2

59 Asymptotically free UV gauge scaling solutions gauge scaling towards flatness (RECHENBERGER,SCHERER,HG,ZAMBELLI 13; HG,ZAMBELLI 14IP) g = g = 0.04 g = 0.03 g = 0.02 g = 0.01 approach to UV k : φ 2 g 2 0, φ min 2 1 g 2, λ g4 0, = deep Euclidean region is sidestepped m 2 W k 2 const. marginal-relevant direction: self-similar & polynomially bounded (MORRIS 98)

60 Estimates of IR Observables (HG,ZAMBELLI 14IP) Higgs to W boson mass ratio: 10 3 mh2 4mW 2 m 2 H m 2 W ξ UV-IR mapping of physical parameters: Ξ v m H m W g 2 δm 2 ξ marginal-relevant relevant exactly marginal = pheno-relevant parameter regime is accessible = SU(2) non-abelian Higgs model can be UV complete

61 Summary, Part III Numbers matter... m top, m H QFT is more than a collection of recipes... new insight from new tools vacuum stability: no reason for concern... so far... UV complete models approaching flat potentials... appealing also in view of current data

62

63 The IR window for the Higgs boson mass m H vλ R (WETTERICH 87) mapping: λ Λ λ R not surjective on R + e.g. for φ 4 bare potential, fix Λ = 10 7 GeV mhgev (HG,GNEITING,SONDENHEIMER 13) convergence check of derivative expansion NLO / LO strong coupling U eff solver (polynom. exp.) Λ

64 chiral Yukawa model: Towards the standard model (HG,SONDENHEIMER 14) [ S = µ φ µ φ + U(φ φ) + ti / t + bi / b + ih b ( ψ L φb R + b R φ ψ L ) + ih t ( ψ ] L φ C t R + t R φ C ψ L) φ = ( ) φ1 + iφ 2 φ 4 + iφ 3 ψ L = ( tl b L ) enforce decoupling of Goldstone bosons (m G = 0) k 2 k 2 + m 2 G k 2 k 2 + m 2 G + gv 2 k choose gauge boson masses gv 2 k = (80.4 GeV)2 cf. lattice model (GERHOLD,JANSEN )

65 Lowering the lower Higgs boson mass bound generalized bare potential with λ 6,Λ (φ φ) 3 interaction: (HG,SONDENHEIMER 14) mhgev λ 3Λ = 0, λ 2Λ = 0 λ 3Λ = 3, λ 2Λ = GeV = same RG mechanism at work

66 Gauge-invariant regulator ζ function regularization (interpolating reg.: propertime dim.reg.) (HG,SONDENHEIMER 14) U F,t (φ) = 4µ4 d Λ d 2 (d 2)(4π) d/2 h2 t φ 2 }{{} <0 (mass-like term) + 2µ4 d dt (4π) d/2 1/Λ 2 ( ) e h2 T 1+(d/2) t φ 2T + ht 2 φ 2 T 1 }{{} >0 (interaction part) interaction part: strictly positive = cannot induce instability for any finite Λ, µ, d

67 Gauge-invariant regulator ζ function regularization (interpolating reg.: propertime dim.reg.) (HG,SONDENHEIMER 14) U F,t (φ) = 4µ4 d Λ d 2 (d 2)(4π) d/2 h2 t φ 2 }{{} <0 (mass-like term) + 2µ4 d dt (4π) d/2 1/Λ 2 ( ) e h2 T 1+(d/2) t φ 2T + ht 2 φ 2 T 1 }{{} >0 (interaction part) BUT: Limit Λ and expansion about d = 4 ɛ? U F,t (φ) # ɛ h4 t φ 4 1 ) 16π 2 h4 t φ (ln 4 h2 t φ 2 µ 2 + const. = instability : artifact of dim.reg use dim.reg. in the presence of large fields: (BROWN 76,LUSCHER 82)