LHC – A “WHY” FACILITY Gordy Kane Taipiei October 2010

1. LHC – A “WHY” FACILITY Gordy Kane Taipiei October 2010

2. Why a “why” facility? Standard Model already gives a beautiful, well-tested description of all we see – but it doesn’t tell us why it is what it is, and it doesn’t answer some questions Most important immediate questions: Is there a signal of physics beyond the SM? Then: • How is the EW symmetry broken, Higgs physics? • How does the dark matter relate to the other particles? Stable particle produced in collisions that carries momentum away? • Is the supersymmetric extension of the SM the correct one? • How is the supersymmetry broken? Then – what are the clues pointing to a high scale, short distance theory?

3. Several DM-LHC connections OUTLINE • Introduction – supersymmetry? • What if CDMSII were a signal (or Xenon soon?) • Light h, light gluino • Higgs boson – Mh 163 GeV – Tevatron? – lighter than LEP limits? • Discovery of new physics at LHC – is it indeed supersymmetry? – measure spin early! • What do we really hope for from LHC? – not just signals or even masses – low scale Lagrangian? – high scale theory? – 10D underlying theory? – how is supersymmetry broken? – how is electroweak symmetry broken? – are gaugino masses unified? – can use “footprint analysis” to study such issues • M-theory based construction  wino-LSP  PAMELA positrons and antiprotons  cosmological history  LHC predictions – complete theory with related predictions for LHC, dark matter, CP violation, axions, etc – 4-top signal of new physics at LHC

4. REMINDER: Strong indirect evidence laws of nature are supersymmetric • Stabilizes hierarchy! • Data from LEP suggest -- high scale unification of gauge coupling strengths as predicted in supersymmetry -- light Higgs boson ( 2MZ )as required in supersymmetry -- deviations from Standard Model predictions small, predicted in supersymmetry All predicted ahead of time from supersymmetry! Simultaneously: • Supersymmetry provides a dark matter candidate (LSP) • Supersymmetry allows a derivation of the Higgs mechanism – supersymmetry breaking  EWSB • Supersymmetry allows an explanation of the matter asymmetry of the universe • Strong theoretical motivation Input that superpartner masses  TeV, and derive all the rest! If indeed nature is supersymmetric, can calculate predictions perturbatively from high scale for collider scale, or extrapolate data implications to high scale – opens a window to the Planck scale

5. A. Dark matter – Higgs – LHC connections Recently CDMSII (detector designed to see recoil of nucleus hit by dark matter particle) reported data – 2 events with background  0.8 events – no reason to think that is a signal of dark matter particle  But as a pedagogical exercise, imagine it were (Xenon100 data soon…)  favors (a) light Higgs boson, perhaps below LEP limit (b) lighter gluino (a)Reported cross section fairly large, N 10-44 cm2, and (b) LSP   wino +  bino +  higgsino, and increases with 2 -- higgsino content increases for smaller -parameter, and smaller gluino mass makes it easier to satisfy the EWSB conditions for smaller 

6. B. HIGGS BOSON – crucial to complete SM • Global fit to LEP data, plus effect of Higgs radiative corrections to MW – Mtop relation, imply Mh 160GeV • So far signal/background has been improving much faster than N at Tevatron because systematics improved, more connected channels added – expected to continue • Tevatron can cover that entire region with about 12 fb-1 per detector integrated luminosity, which it should get in about 2 years if collider and detectors and experimenters keep working • Now Tevatron , because of stimulus money – running 2010-2011 -- manpower available for analysis since LHC slowdown • Very likely Higgs boson will be detected at Tevatron in  2 years LHC Higgs analysis difficult – could see 2-3  effect in each of several channels after 2-3 years running at fairly high luminosity But Tevatron will at best detect h – LHC essential for properties

7. Precision ==> Higgs constraints Standard Model Higgs Expected now with all constraints : MH = 90 +36-27GeV MH < 163 GeV @ 95 % CL

8. Briefly look at “lighter Higgs”, below LEP limits – should explore experimentally There are at least 17 distinct configurations of the MSSM Higgs sector consistent with the LEP results: Theoretical implications of the LEP Higgs search.G.L. Kane, Ting T. Wang, Brent D. Nelson, Lian-Tao Wang Phys.Rev.D71:035006,2005; hep-ph/0407001 Trace literature, constraints from: Light MSSM Higgs boson scenario and its test at hadron colliders.Alexander Belyaev, Qing-Hong Cao, Daisuke Nomura, Kazuhiro Tobe, C.-P. Yuan, Phys.Rev.Lett.100:061801,2008; hep-ph/0609079

9. MSUGRA? x

10. Supersymmetry requires two complex SU(2)xU(1) doublets, so 8 degrees of freedom – three become longitudinal polarization stated of W±, Z so 5 physical states remain h,H,A,H If h lighter than Standard Model LEP limit value, then associated supersymmetry higgs states are not too heavy  can probably also see them at LHC

11. C. WHEN THERE IS A DISCOVERY AT LHC --Is it supersymmetry (or whatever)?? – classic way to tell is to measure the spin of a partner – usual approaches very difficult, use decay angular correlations Much easier: Measure gluino spins from production cross section with little data -- 0805.1397 GK, Petrov, Shao, Wang Use production cross section information to measure spins of larger effects – works because spin quantized and effects large Clearly works for isolated effects – i.e. for many ways the world might be – requires few hundred pb-1 or less – assume color, etc, check later • Paper shows also works for cases where really observe mass difference, by using other data – still works for  1 fb-1 • Currently studying effects of producing several states of similar mass, and needed luminosity in some familiar benchmark models -- GK, Ian-Woo Kim, Eric Kuflik, Liantao Wang, in progress

12. Works to measure top quark spin Spin 1/2top quark Spin-zero “top”

13. Basic way to proceed is to simply test hypothesis that new signal corresponds to spin ½ (color octet etc) new particle, consistent with being gluino Since spin quantized, and associated cross sections very different, that is definitive!

14. KK partner, same spin gluino A plot of the cross sections for gluino (spin-half) pair production (solid line) and KK-gluon (spin-one) pair production (dashed line) at LHC. In the calculation, the masses of extra color triplet quark partners are taken to be 5 TeV.

15. More tests – branching ratios, topologies very different too Is it SUSY?hep-ph/0510204 AneshDatta, GK, Manual Toharia Note ratios

16. More on spin analyses: A Review of Spin Determination at the LHC.Lian-Tao Wang, Itay Yavin, Int.J.Mod.Phys.A23:4647-4668,2008, arXiv:0802.2726 More on determining masses at LHC: Algebraic Singularity Method for Mass Measurement with Missing Energy, Ian-Woo Kim, arXiv:0910.1149

17. D. ULTIMATELY WE WANT TO LEARN MUCH MORE FROM LHC -- is the same dark matter being observed at LHC and in indirect, direct detection? -- what is the composition of the dark matter? -- not just masses, spins but what is the full spectrum, what is the Lagrangian at the collider scale – degeneracies! -- not just the collider scale, but the associated high scale theory -- are gaugino masses unified? -- not just the high scale theory but the 10D underlying theory And supersymmetry breaking! -- If supersymmetry unbroken, no new parameters beyond SM ones -- If supersymmetry unbroken the EW symmetry also unbroken, no Higgs mechanism – no quark and lepton and W,Z masses -- Understanding supersymmetry breaking essential for understanding how string ground state arises, how moduli (so force strengths, mass values) are determined

18. Each axis a Lagrangian parameter Supersymmetry and the LHC inverse problem.Nima Arkani-Hamed, Gordon L. Kane, Jesse Thaler, Lian-Tao Wang, JHEP 0608:070,2006. hep-ph/0512190

19. Also, when studies are done for the reach of LHC, for the relevance of models, for limits, for possible signals, etc, people usually select values of some parameters and vary one or two parameters – we know little about whether the results are generic or special Can study all these issues with “footprint analysis” GK, P.Kumar, J. Shao Phys.Rev.D77:2008, arXiv:0709.4259 – illustrate here with examples

20. Studying Gaugino Mass Unification at the LHC.Baris Altunkaynak, Phillip Grajek, Michael Holmes, Gordon Kane, Brent D. Nelson, JHEP 0904:114,2009, arXiv:0901.1145 Testing hypotheses about relations among gaugino masses is of great interest once there is data But measuring soft gaugino masses M1, M2, M3 unlikely, and measuring high scale ones very unlikely – soft-breaking mass parameters are not experimental mass eigenstates! We show that nevertheless it is not hard to test the hypotheses using footprint analyses, since observables are sensitive to Mi Method can be generalized to many issues about implications of data for underlying theory – e.g. squark mass sum rules, LSP wavefunction

21. Assume a mirage mediation model, where  parameterizes mixture (at collider scale) -- make footprint plots for several signatures  measured – this is one example – essential point is that signatures are sensitive to  -- universality for =0 Get overlaps for any single pair, but can separate by using several signature pairs Number of events Y Any two signatures Number of events of type X at LHC (5fb-1)

22. Luminosity needed to measure mixture  (“typical result”) fb-1

23. The Footprint of F-theory at the LHC.Jonathan J. Heckman, Gordon L. Kane, Jing Shao, Cumrun Vafa, JHEP 0910:039,2009, arXiv:0903.3609 (Fenomenology) All these predict PAMELA positron excess is not DM annihilation, and no current direct DM detection

24. Studying DM at LHC glggg gluino If a signal is seen the LSP is not higgsino – size distinguishes wino, bino LSP Holmes, GK, Kuflik, Nelson, in progress – here only shapes, not normalized

26. hig

27. They do standard cut analysis. Footprint analysis would give result at lower luminosity

28. “Problem” with footprint analyses -- should be done by experimenters -- but requires combining analyses of different channels -- hampered by sociology

29. E. M-theory and LHC Now turn to an interesting (exciting?) set of developments – dark matterLHCcosmological historystring-based models In past few years we’ve pursued development of an M-theory based model M-theory compactified on 7D manifold of G2 holonomy – moduli stabilized in fluxless sector, and supersymmetry broken at high scales simultaneously (gaugino condensation) – moduli geometrical, generically in gauge kinetic function, enter superpotential non-perturbatively – calculate metastable de Sitter minimum of potential [Acharya, Bobkov, Kane, Kumar, Shao, Vaman, Watson, Kuflik, Lu]

30. Generically generates TeV scale physics by the conventional gaugino condensation mechanism M3/2 ten(s) of TeV • scalar masses  M3/2 • tree level gaugino masses (gluino, chargino, neutralinos) suppressed, compressed (by form of supersymmetry breaking) so same order as anomaly mediation contribution wino LSP generic • Moduli masses, lifetimes ( 10-3 sec), branching ratios calculable – non-thermal cosmological history – dark matter arises from moduli decays (relic density in model few times observed value) • No phenomenological problems – addresses many issues together • Can describe PAMELA positron excess (and antiprotons) • Normalize wino mass to PAMELA  Mgluino 950 GeV, gluino signatures good for detection at LHC If you prefer you can think of this as construction of a comprehensive weak scale model with all the above features simultaneously, and the G2 construction as its UV completion

31. Is non-thermal cosmological history a problem? NO, success! • Generically, string theories with TeV physics have “light moduli” with masses  M3/2 – they decay at about 10 MeV, after thermal freezeout temperature, produce much entropy and many LSP’s – arguably such a history implied by string theories with TeV physics • In such theories get about right relic density and “WIMP miracle” IF the wimp annihilation cross section is that of wino, which is large!– with thermal history and wino LSP annihilation rate so large that relic density small[ (v)wino ≈ 2x10-24 cm3 /sec] • that large annihilation rate also allows possibility of describing PAMELA data without unreasonable “boost factors” wino W±e±, ± , quarks … chargino wino W±

32. Non-thermal cosmological history Can argue that any of these generically Implies the others “Wimp miracle” nLSP ≈ H(reheat temp)/v Wino-like LSP, gives  correct relic density String-based models for TeV physics  moduli masses  M3/2 Allows description of PAMELA data

33. Assumes no dark matter density fluctuations, and , and local density =0.35 GeV/cm3 Theory normalized to local relic density, no “boost factors” – Wino LSP annihilation cross sections (10-24 cm3 /sec) 2.5, 2.2, 1.9

34. Note predicted decrease

35. Inject astrophysical electrons x

36. Wino plus background x

37. x

38. Thus PAMELA may have been lucky and seen the first direct evidence of dark matter and supersymmetry (the positron excess) How can this be tested/confirmed? Satellite data : • AMS2 can confirm and extend PAMELA data itself, and test interpretation at higher energies • Fermi could see gammas from wino annihilation, both continuous spectrum (diffuse, and dwarf galaxies) and particularly monoenergetic lines from Z and  wino w  wino W Z,  at E = Mwino , Mwino- 12 GeV (rate somewhat larger for Z) W chargino

40. Direct detection of LSP : • Pure wino has very small rate to scatter on nuclei – probably excluded if CSMSII, Xenon100 see signal • But with  10% higgsino mixture can have direct scattering in CDMSII, Xenon100 region, and still good description of PAMELA data, so here the data is not qualitatively definitive – mixture also reduces annihilation to Z – mostly wino LSP ok • G2-MSSM has upper limit cross section on Xenon of about 10-45 cm2 • Xenon100 has begun taking data with 50kg fiducial volume – expected background ≈ 0 – so far about 300 days! • If CDMSII had a signal then Xenon100 should see many events

41. LHC with wino-like LSP Assume G2-MSSM does describe PAMELA data, so fix wino mass For wino-like LSP, chargino and neutralino essentially degenerate – main LHC source of events with good signatures is gluino production, followed by gluino decay to chargino, neutralinos In usual models 2 Mgluino /Mwino 9 – in many string constructions spectrum compressed In M-theory normalized to describe PAMELA, ratio is 3.1-4.6

42. Dan Feldman, GK, Ran Lu x

43. 7 TeV

44. LHC phenomenology of light wino LSP well known Early  1999, 2000 • Moroi-Randall • Feng Moroi Randall Strassler • Ghergetta, Giudice, Wells More recent • Moroi, Yanagida et al ph/0610277 • Acharya et al 0801.0478 • Buckley, Randall, Shuve arXiv:0909.4549 

45. Spectrum predictions from M-theory G2 compactification Mgluino > MN2 > MC1 > MN1 10 TeV stop 1 TeV gluino c1 LSP=N1 PAMELA

46. An example spectrum and notation: Gluino G, mass 900 GeV Chargino C1, mass 173 GeV LSP N1, mass 173 GeV 2nd neutralino N2, mass 253 GeV Stop  8700 GeV, mainly RH Higgs boson= 120 GeV (gravitino  35 TeV) C1  wino, N1  wino, N2  bino

47. Production cross sections 14 TeV (7 TeV): G Dominant: g   1/2 pb ( 1/10 pb) G Also C1 Z C1   1/2 pb C1 and N1 are essentially degenerate, so C1 decays soft, so C1 + N1 channel has large cross section but nearly unobservable C1  N1 + W*, W*e, , quarks C1 + N2 and N1 + N2 small since C1 – W – bino coupling small C1 + G, N1 + G, N2 + G all small since need squark exchange

48. Gluino decays tbar 4 tops for gluino pair! G stop top 2 b’s, 3 W’s for each gluino! N2 ( C1+W* ) MODE BRlifetime  10-19 sec Gt tbar N2  ½ t bbar C1  ¼ b bbar N1  ¼ g N1+N2  1% N2 rather than N1 because N1 mostly partner of W so doesn‘t couple to RH stop MN2 – MC1 = 80 GeV for this model so essentially real W

49. Chargino, LSP nearly degenerate, so hard to see – missing energy from “escaping” chargino, LSP • mass difference  200 MeV(!), so decay has 1-2 soft pions • find via triggers: • -- gluino pair • -- large missing energy plus no energetic jets or leptons or b’s (one trigger with N2, another without) • probably can use vertex detector on events with gluino production to see N2, C1

50. General features of signatures for G2-MSSM: • lots of leptons but always with jets – no “trileptons” • ALL prompt leptons are from W decay!  no flavor correlations for leptons • Leptons from gluino production so no charge asymmetries (compared to squarks…) • Assuming the model is right, distributions can measure three mass differences, G-N2, G-N1, G  can solve for three masses • Should not see squarks, or virtual squarks such as q G squark qbar N1, N2, C1