Stefania Gori. The University of Chicago & Argonne National Laboratory. Theory Seminar, EPFL, Lausanne, June 5th 2012

Transcription

1 Exploring Exploring the the Higgs Higgs portal portal with with the the first first LHC LHC data data Stefania Gori The University of Chicago & Argonne National Laboratory Theory Seminar, EPFL, Lausanne, June 5th 2012

2 Outline 1. Introduction: status of the Higgs searches with 10fb-1 Exclusion for a heavy SM-like Higgs boson A small hint for a light Higgs boson 2.The Higgs portal and the Higgs phenomenology A model independent analysis A scalar model for the Higgs boson Connection with EWPTs 3.Direct searches of exotic scalars Current LHC bounds on extra scalars Based on: 4. Conclusions Updated results Exploring the Higgs portal with 10fb-1 at the LHC B. Batell, S. Gori, L.T. Wang arxiv: Higgs searches and status of EWPTs B. Batell, S. Gori, L.T. Wang, work in progress 2/38

3 A heavy SM-like Higgs in trouble ATLAS-CONF ATLAS Moriond results CMS CMS PAS HIG All channels are updated with full luminosity ( fb-1) At 95% C.L.: 3/38 Exclusion At 95% C.L.:

4 A heavy SM-like Higgs in trouble ATLAS-CONF At 95% C.L.: 4/38 ATLAS Moriond results Exclusion CMS CMS PAS HIG At 95% C.L.:

5 A hint for a light Higgs Moriond results Consistency of the background only hypothesis ATLAS-CONF ATLAS 5/38 CMS CMS PAS HIG σ local significance 2.8σ local significance (2.8σ from the γγ channel; 2.1σ from the ZZ channel) (the ZZ and the γγ channel do not correspond exactly to the same mass)

6 Best fit values ATLAS CMS mh = 126 GeV Results of 6/38 1. last December 2. Moriond/February

7 Best fit values CMS ATLAS 125 mh = 126 GeV Results of 6/38 1. last December 2. Moriond/February 3. Moriond MVA

8 What can we learn from these results? 7/38

9 Possible reactions ATLAS & CMS results Take the signal seriously SM Higgs boson Higgs with 1. enhanced gamma-gamma rate Study of the implications for new physics scenarios 8/38

10 Possible reactions ATLAS & CMS results Take the signal seriously SM Higgs boson Higgs with 1. enhanced gamma-gamma rate This is a statistical fluctuation Higgsless theories & theories with heavy Higgs are still viable 2. Study of the implications for new physics scenarios 8/38

11 The two interpretations of data What are the implications of 1. A light Higgs at ( ) GeV Chi square, assuming a Gaussian form in R: It depends on the number of degrees of freedom and on the C.L. (for 3 d.o.f.: #~3.5 at 1σ, #~8 at 2σ and #~14 at 3σ) ~ 2. A hidden heavy Higgs at the 95% C.L., combination of both experiments Suppression of its gluon gluon production cross section 9/38

12 Limitations of the fit Current information given by experimental collaborations is often not sufficient to allow theorists to make a rigorous fit/exclusion bounds Ex. 1. exact likelihood (departure from Gaussians...) 2. correlations among channels For more sophisticated fits: 3. cut efficiencies for each Higgs production mode and event category Carmi, Falkowski, Kuflik,Volansky, arxiv: Azatov, Contino, Galloway, arxiv: Espinosa, Grojean, Muhlleitner, Trott, arxiv: Giardino, Kannike, Raidal, Strumia, arxiv: Ellis and You, arxiv: Farina, Grojean, Salvioni, arxiv: Klute, Lafaye, Plehn, Rauch, Zerwas, arxiv: Best if fit is done by experimentalists; theorists can give support on how to perform calculations In addition: it is rather premature to fit the present LHC data at 125GeV; at the end it is not a discovery yet warm-up exercise in preparation for better statistics 10/38

13 Higgs at Hadron Colliders A. Djouadi, All three may be affected by the presence of new physics γγ The good point being at 125 GeV: several channels can be measured: ( maybe 11/38, Gainer, Keung, Low, Schwaller, 2011)

14 NP in the Higgs production and decays The hierarchy problem suggests the existence of new particles at the TeV scale that couple with the Higgs If new physics restores naturalness, Higgs interactions are probably modified Three main processes can affect the Higgs signals at the LHC: Hiding Higgs Loop effects 12/38 Mixing effects The Higgs golden channel in the low mass range (γγ) can be easily modified since arising at the loop level in the SM

15 The Higgs portal A easy way to couple the Higgs to NP: (name coined in Patt, Wilczek, 2006) Lorentz invariant gauge singlet ONP can be made of NP states carrying SU(3) x SU(2) L x U(1)Y quantum numbers Many specific examples, including Higgs-partner coupling in natural theories Possible couplings: Fermion Vector Need low cut-off to have large effect Manohar, Wise, 2006 Hur, Jung, Ko, Lee, 2007 Low,Vichi, 2010 Bai, Fan, Hewett, 2011 Dobrescu, Kribs, Martin, 2011 (Lebedev, Lee, Mambrini, 2012) Can introduce mixing between SM and new gauge boson (more constrained) Scalar One well motivated example: scalars of Susy 13/38

16 The Higgs portal A easy way to couple the Higgs to NP: (name coined in Patt, Wilczek, 2006) Lorentz invariant gauge singlet ONP can be made of NP states carrying SU(3) x SU(2) L x U(1)Y quantum numbers Many specific examples, including Higgs-partner coupling in natural theories Possible couplings: Fermion Vector Need low cut-off to have large effect Manohar, Wise, 2006 Hur, Jung, Ko, Lee, 2007 Low,Vichi, 2010 Bai, Fan, Hewett, 2011 Dobrescu, Kribs, Martin, 2011 (Lebedev, Lee, Mambrini, 2012) Scalar One well motivated example: scalars of Susy 13/38 Can introduce mixing between SM and new gauge boson (more constrained) NP effects: After EWSB: h SS Loop effects in gg h and h γγ

17 Model independent approach Assumptions: The scalar S is relatively heavy it can be integrated out It does not introduce new sources of CP violation It generates effects on the Higgs golden channel gg h γγ (Manohar, Wise, 2006) Effects in h ZZ, WW can be neglected (see also Low, Rattazzi, Vichi, 2009) Higgs production cross section SM: Decay of the Higgs into two photons SM: Q 14/38

18 Model independent fit of the LHC data 1. (see also the recent paper: Espinosa, Muhlleitner, Grojean,Trott, ) 1σ 2σ Assuming small cg and cγ coefficients, to have small NP effects in the Higgs total width 15/38

19 A (still) possible heavy Higgs 2. Can a heavy Higgs be hidden because of the reduced production cross section? Strong constraint coming from the ATLAS h ZZ 4l Above the top threshold we cannot have an arbitrary suppression of the gg production cross sections 16/38

20 A model of scalars We restrict our attention to those scalar representations which allow a renormalizable coupling with some SM fermions See also Bai, Fan, Hewett, 2011 Requirements Potential bounded from below Perturbativity of the coupling constants Lower bound on the Higgs portal coupling Imposing the cut off scale to be at least at (2-3) TeV: λ 4 S does not participate to EWSB, namely S never mixes with the SM Higgs H 17/38

21 Scalar representations and their signatures (SU(3), SU(2), Y) Simplifying assumption: We choose Y to allow simple renormalizable couplings to SM fields (generalization to other representations is straightforward) 18/38

22 Scalar representations and their signatures (SU(3), SU(2), Y) Simplifying assumption: We choose Y to allow simple renormalizable couplings to SM fields (generalization to other representations is straightforward) 18/38

23 Modification of the Higgs phenomenology 1. Assumption: λ is relatively small & mh< 2 ms The Higgs total width receives only negligible corrections no splitting between the several components of the multiplet Analogously the gg Higgs production cross section: See Bonciani, G. Degrassi, and A. Vicini, 2007 & Boughezal, Petriello, 2010 For the NNLO computation for the 8 representation 19/38

24 Modification of the Higgs phenomenology 1. Assumption: λ is relatively small & mh< 2 ms The Higgs total width receives only negligible corrections no splitting between the several components of the multiplet Analogously the gg Higgs production cross section: Note: 19/38 λ>0 implies a positive NP contribution to the production cross section and a negative NP contribution in the branching ratio to two photons See Bonciani, G. Degrassi, and A. Vicini, 2007 & Boughezal, Petriello, 2010 For the NNLO computation for the 8 representation

25 A light Higgs boson signal: colored scalars 1. (8,2,1/2) ATLAS CMS (*) (*) (*) (*) # degrees of freedom=3 A strong suppression of the gg fusion is not producing a good fit of the CMS data (*) Enhancement of the gg fusion is rather constrained 20/38 2σ 1σ No effects on the total width

26 A light Higgs boson signal: colored scalars 1. (3,2,1/6) ATLAS CMS (*) (*) (*) (*) # degrees of freedom=3 A strong suppression of the gg fusion is not producing a good fit of the CMS data (*) Enhancement of the gg fusion is rather constrained 21/38 2σ 1σ No effects on the total width

27 A light Higgs boson signal: color neutral 1. (1,1,2) ATLAS CMS Large negative contribution to the gamma gamma width (*) (*) (*) # degrees of freedom=3 The gamma-gamma rate is enhanced by a factor of 2 Both ATLAS and CMS prefer λ<0 22/38 1σ 2σ enhancement of the gamma-gamma rate

28 Modification of the Higgs phenomenology 2. Assumption: λ is relatively small & mh> 2 ms The Higgs total width can receive sizable corrections no splitting between the several components of the multiplet New decay mode: Second player in hiding a heavy Higgs at the LHC Effects in the gluon gluon fusion 23/38

29 A hidden heavy Higgs λ = -1 mh = 300GeV 2. (3,2,1/6) (8,2,1/2) (6,1,4/3) (8,2,1/2) (8,2,1/2) (6,1,4/3) (3,2,1/6) (6,1,4/3) (3,2,1/6) The triplet representation should be rather light to hide efficiently a heavy Higgs Already been excluded by direct searches? 24/38 See also Dobrescu, Kribs, Martin, 2011 See later...

30 Light exotic scalars, so what? Higgs phenomenology are consistent with (or even favor) very light exotic scalars These states can be searched for at the LHC Complementary to the measurement of the Higgs phenomenology QCD pair production with sizable rates even at the 7 TeV LHC Note: single production of these scalars through gluon gluon fusion can have similar rates only for very heavy (ms>1tev) scalars Gresham, Wise, /38

31 Production of the new scalars (see also Del Nobile, Franceschini, Pappadopulo, Strumia, 2009) Copiously pair produced at the 7TeV LHC SU(3)c triplet SU(3)c sextet SU(3)c octet 105 events at 10fb-1 SU(2)L singlet SU(2)L doublet SU(2)L triplet 26/38

32 Decay modes Decays mediated by the renormalizable couplings Ex. s d S If ηij are generic O(1) couplings, large FCNCs are induced, s however: 1. One can impose the Minimal Flavor Violation (MFV) principle 2. One can impose that ηij are small 1. 27/38 Only the representation (8, 2, 1/2) can have MFV coupling d Work for the future Manohar, Wise, 2006 Bottom and top rich final states

33 Small η regime Lifetime of a real scalar decaying to two (relatively light) fermions: 2. If η 10-7 the decay is prompt in the detector R-hadron = long lived charged and hadronizing particle (CMS-PAS-EXO ) L = 1.09 fb-1 Reminder: Present limit on sgluons: M > 900 GeV We deduce the bounds for the color octets: (8,2,1/2) Present limit on scalar tops: M > 620 GeV We deduce the bounds for the color triplets: (8,2,1/2) (6,1,4/3) (3,2,1/6) (6,1,4/3) (3,2,1/6) λ=-1 Very difficult to hide a Higgs if the scalars are long lived 28/38

34 Promptly decaying exotic scalars In most cases, no dedicated S searches We estimate the bounds based on similar final state searches Assumption: The scalar decays 100% in one of the possible final states If several decay modes are open, the constraints would be less stringent We compare the LO cross section with the LHC excluded rate (this is an estimation, the kinematic may be different for example) However, since rate falls very fast with mass, a factor of 2 error on the constraint of rate only translate into 10-20% error on the constraint of mass Strongest constraint Several LHC searches: 29/38

35 Leptoquarks and 2j+missing energy searches Only for color triplets: L ~5 fb-1 ATLAS collaboration:atlas-conf : similar to Susy searches In the limit, Mquark > 1200 GeV We deduce the bounds for the color triplets: Reminder: (8,2,1/2) SU(2)L gauge invariance requires that this final state co-exists with a lepto-quark-like (8,2,1/2) (6,1,4/3) (3,2,1/6) final states (6,1,4/3) (3,2,1/6) λ=-1 ms > 650 GeV 30/38 (for first generation lepto-quarks) L = 1.03 fb-1 ATLAS collaboration: L = 1.8 fb-1 CMS PAS EXO

36 4 jet searches A very common final state (for triplets, sextets and octets) A rather difficult channel because of the large QCD background Uncovered mass range L = 34 pb-1 ATLAS collaboration: arxiv: /38 L = 2.2 fb-1 CMS PAS EXO (last January public note)

37 4 jet searches & Higgs phenomenology (3,2,1/6) (8,2,1/2) CMS fit for a light Higgs (mh~125 GeV) 32/38

38 4 jet searches & Higgs phenomenology If the exotic scalars are decaying exclusively to 2 jets: (3,2,1/6) Allowed (8,2,1/2) CMS fit for a light Higgs (mh~125 GeV) 32/38

39 4 jet searches & Higgs phenomenology (3,2,1/6) (8,2,1/2) (6,1,4/3) Heavy hidden Higgs 33/38

40 4 jet searches & Higgs phenomenology If the exotic scalars are decaying exclusively to 2 jets: (3,2,1/6) (8,2,1/2) Allowed (6,1,4/3) Heavy hidden Higgs 33/38

41 What next? A look into EWPTs A Light Higgs has been hinted by EWPTs since years BUT from GFitter group Old tension in the fit: Bottom forward backward asymmetry 2.3σ discrepancy for mh = 124GeV (larger for heavier Higgs bosons) A puzzle? After all, discrepancies come and go all the time! But, if attributed to experimental error, or statistical fluctuation, electroweak fit prefers a lighter Higgs, in (slight) tension with LEP bound! 34/38

42 Connection Higgs phenomenology-ewpts Can the anomaly and Higgs rates be due to the same underlying new physics? Two possibilities: 1. NP that alters Higgs searches and status of EWPTs B. Batell, S. Gori, L.T. Wang, work in progress (taking seriously that measurement) Choudhury, Tait, Wagner, 2001 Revision of the beautiful mirrors model Additional vector like b quarks mixing with the b quark of the SM Not for this talk 2. Throw away the asymmetry & put NP in the other EW observables to get mh=125 GeV as a good fit (statistical fluctuation of the measurement) In both cases, electroweak data can suggest NP 35/38

43 New fit of EW data mh = 125 GeV A small positive NP contribution to the T parameter? 36/38

44 Fit with a scalar doublet neutral under SU(3) Requires mass splittings ~ GeV 37/38

45 Conclusions 2012 is going to be the year for the Higgs: Confirm a light Higgs signal, or Rule out SM-like weakly coupled Higgs. Rich implications for NP particles interacting with the Higgs (Higgs portal) 4jets searches are close to mass scales needed to hide a heavy Higgs. What about the ( )GeV range? Complementarity between exotic scalar searches and Higgs phenomenology Most exciting scenario! If deviations from SM-like Higgs are observed, the light exotic matter coupling through the Higgs portal will provide a promising and experimentally testable explanation. To keep an eye on EWPTs! Let s wait for more data! 38/38

46 SM Higgs decay modes Main Higgs decay modes for the LHC Backup

47 Coupling with SM fermions, an example Example: color octet 1. It interacts with This Lagrangian mediates the decays With LHC signatures from SS(*) production: a,b,c,... SU(3)c indices i,j,k,... flavor indices α,β,γ,... Lorentz indices 2. It interacts with This Lagrangian mediates the decays With LHC signatures from SS(*) production: Some of these signatures are already pretty constrained by the LHC! Backup

48 Fit: double solution Backup