Prospects for SUSY at the LHC

Transcription

1 Prospects for SUSY at the LHC LHC details Sparticle production Sparticle decay Event generation LHC reach and year 1 Multi-muons + jets RT-S dijet signal precision measurements Howard Baer University of Oklahoma OUTLINE Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

2 The role of the CERN Large Hadron Collider (LHC) The LHC is a proton-proton collider (pp) Each beam will have E = TeV (trillion electron volts) Center-of-mass energy E s = 7 14 TeV The collider is on a circular tunnel 27 km in circumference It is nearly ready: turn-on expected in November 2009! Protons are not fundamental particles: made of quarks q and gluons g The quark and gluon collisions should have enough energy to produce TeV-scale superparticles at a large enough rate that they should be detectable above SM background processes LHC should be able to discover SUSY or other new physics: but probably can t rule SUSY out if just a Higgs or nothing new is found Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

3 Layout of the LHC:two main detectors: Atlas and CMS Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

4 The Atlas detector Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

5 The CMS (Compact Muon Solenoid) detector Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

6 For a hadronic reaction, Parton model of hadronic reactions A + B c + d + X, where c and d are superpartners and X represents assorted hadronic debris, we have an associated subprocess reaction a + b c + d, whose cross section can be computed using the Lagrangian for the MSSM. To obtain the final cross section, we must convolute the appropriate subprocess production cross section dˆσ with the parton distribution functions: dσ(ab cdx) = a,b dx a dx b f a/a (x a, Q 2 ) f b/b (x b, Q 2 ) dˆσ(ab cd). 0 where the sum extends over all initial partons a, b whose collisions produce the final state c + d. Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

7 Parton Distribution Functions (PDFs) Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

8 Calculating subprocess cross sections/decay rates in QFT The fundamental calculable object in QM is the amplitude M for a process to occur A pictorial representation of M is given by a Feynman diagram Feynman rules for many theories can be found in standard texts: e.g. Peskin& Schroeder, Introduction to Quantum Field Theory In the MSSM, an additional complication occurs due to presence of Majorana spinors Methods for handling these given e.g. in Weak Scale Supersymmetry (HB, X. Tata), or book by M. Drees, Godbole& Roy total amplitude M is sum of all different ways a process can occur M is a complex number; M 2 gives probability must normalize and sum (integrate) over all momentum configurations to gain cross section, usually in femtobarns: Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

9 Calculating subprocess cross sections/decay rates in QFT dˆσ = 1 2ŝ 1 (2π) 2 d 3 p c 2E c d 3 p d 2E d δ 4 (p a + p b p c p d ) F color F spin M 2, Must sum (integrate) over all final state momentum configurations May be done analytically for simple processes e.g. 2 2 Usually done using Monte Carlo method for n 3 Monte Carlo well suited for adding on particle decays so one has really 2 n processes where n can be very large Convolution of subprocess cross section with PDFs must be done numerically, since PDFs distributed as subroutines Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

10 Gluino pair production Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

11 Squark pair production Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

12 Gluino-squark associated production Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

13 Production at LHC Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

14 Gluino decays: g q q or 3-body Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

15 Gluino decays: branching fractions Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

16 Squark decays ũ L u Z i, d W + j, u g, d L d Z i, u W j, d g, ũ R u Z i, u g, d R d Z i, d g. Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

17 Chargino decays W j W Z i, H Zi, ũ L d, dl u, c L s, s L c, t 1,2 b, b1,2 t, ν e ē, ẽ L ν e, ν µ µ, µ L ν µ, ν τ τ, τ 1,2 ν τ, and W 2 Z W 1, h W 1, H W 1 and A W 1. Charginos may decay to a lighter neutralino via W j Z i + f f, (1) Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

18 Neutralino decays Z i W W j, H W j, Z Z i, h Z i, H Z i, A Z i q L,R q, q L,R q, ll,r l, ll,r l, ν l ν l, ν l ν l. If 2-body modes are closed,then the neutralino can decay via Z i Z i + f f (2) Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

19 Sparticle cascade decays Mass scale g ~ (2060) q ~ L ( 1857) q ~ R ( 1770) b ~ 2 (1690) t ~ 2 (1689) b ~ 1 (1619) 1400 t ~ 1 (1449) W ~ ± 2 (1067) Z ~ 4 (1066) Z ~ 3 (1056) W ~ ± 1 (754) Z ~ 2 (754) τ ~ 2 (726) ν ~ τ (712) 400 τ ~ 1 (442) Z ~ 1 (395) Br (%) Z ~ 1 qq (27.0 %) Z ~ 1 τνwbb (12.1 %) Z ~ 1 ττwwbb (8.4 %) Z ~ 1 WWbb (7.4 %) Z ~ 1 τνqq (5.9 %) Z ~ 1 τνwwwbb (4.1 %) Z ~ 1 ττbb (2.9 %) Z ~ 1 ττqq (2.9 %) Z ~ 1 τνzwbb (2.8 %) Z ~ 1 τνhwbb (2.6 %) Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

20 A realistic picture of what SUSY matter looks like at LHC Counting different flavor states (which are potentially measurable), there are well over 1000 subprocess reactions expected at LHC from the MSSM on average, each sparticle has 5-20 decay modes rough estimate of distinct SUSY 2 n processes: this is actually a gross underestimate since each daughter of a produced sparticle has multiple decay modes, and so on... the way forward: Monte Carlo program calculate all prod n cross sections: generate according to relative weights calculate all branching fractions, and generate decays according to them interface with parton shower, hadronization, underlying event computer generated events should look something like what we would expect from the MSSM at the LHC Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

21 Event generation for sparticles p 2 ~ g 3 ~ Z 1 1 ~ q Hadrons 4 ~ Z 1 p Remnants 5 Event generation in LL - QCD 1) Hard scattering / convolution with PDFs 2) Intial / final state showers 3) Cascade decays 4) Hadronization 5) Beam remnants Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

22 Event generations for SUSY Isajet (HB, Paige, Protopopsecu, Tata) IH, FW-PS, n-cut Pomeron UE Pythia (Sjöstrand, Lönnblad, Mrenna) SH, FW-PS, multiple scatter UE, SUSY at low tanβ only Herwig (Marchesini, Webber, Seymour, Richardson,...) CH, AO-PS, Phen. model UE, Isawig, Spin corr.! SUSYGEN (Ghodbane, Katsanevas, Morawitz, Perez) mainly for e + e ; interfaces to Pytha SHERPA (Gleisberg, Hoche, krauss, Schalicke, Schumann, Winter) C + + code for various 2 n processes CompHEP, CalcHEP, Madgraph: for automatic Feynman diagram evaluation: interface via LHA Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

23 Briefly: particle interactions with detector Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

24 SUSY scattering event: Isajet simulation SUSY event with 3 lepton + 2 Jets signature m 0 = 100 GeV, m 1/2 = 300 GeV, tanβ = 2, A 0 = 0, µ < 0, m(q) ~ = 686 GeV, m(g) ~ = 766 GeV, m(χ~ 0 2) = 257 GeV, m(χ~ 0 1 ) = 128 GeV. Jet2 Jet1 e - ~ χ 0 1 ~ χ 0 1 µ + µ - Leptons: p t (µ + ) = 55.2 GeV p t (µ - ) = 44.3 GeV p t (e - ) = 43.9 GeV Jets: E t (Jet1) = 237 GeV E t (Jet2) = 339 GeV Sparticles: ~ p t (χ 0 1 ) = 95.1 GeV p t (χ~ 0 1) = 190 GeV Charged particles with p t > 2 GeV, η < 3 are shown; neutrons are not shown; no pile up events superimposed. Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

25 GMSB AMSB Search for SUSY at LHC: model dependent MM-AMSB (mirage mediation) hypercharged-amsb (HCAMSB) deflected AMSB deflected mirage mediation gravity-mediated models msugra or CMSSM NUHM1, NUHM2 non-universal gaugino masses: MWDM, BWCA, LM3DM, HM2DM,... normal scalar mass hierarchy (m 0 (1, 2) > m 0 (3)) compressed SUSY Split SUSY, pmssm, NMSSM, Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

26 Right or wrong, most analyses work in msugra model m 0, m 1/2, A 0, tan β, sign(µ) Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

27 Search for SUSY at CERN LHC g g, g q, q q production dominant for m < 1 TeV lengthy cascade decays are likely E T + jets 1l+ E T + jets OS 2l+ E T + jets SS2l+ E T + jets 3l+ E T + jets 4l+ E T + jets BG: W + jets, Z + jets, t t, b b, WW, 4t, Grid of cuts gives optimized S/B Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

28 Pre-cuts and cuts E T > 200 GeV N j 2 (where p T (jet) > 40 GeV and η(jet) < 3 Grid of cuts for optimized S/B: N j 2 10 E T > GeV E T (j1) > GeV E T (j2) > GeV S T > muon isolation S > 10 events for 100 fb 1 S > 5 B for optimal set of cuts Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

29 Sparticle reach of LHC for fb m 1/ msugra with tanβ = 10, A 0 = 0, µ > 0 Z l + l - 2l OS 1l 4l γ 0l 2l SS 3l m(u ~ L )=2 TeV m(g ~ )=2 TeV E T miss m 0 m 1/ msugra with tanβ = 30, A 0 = 0, µ > 0 3l 2l OS Z l + l - 1l 4l γ 0l 2l SS m(u ~ L )=2 TeV m(g ~ )=2 TeV E T miss m 0 HB, Balazs, Belyaev, Krupovnickas, Tata: JHEP 0306, 054 (2003) Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

30 Sparticle reach of all colliders with relic density 1600 msugra with tanβ = 10, A 0 = 0, µ > 0 Ωh 2 < LEP2 excluded 1600 msugra with tanβ = 45, A 0 = 0, µ < 0 Ωh 2 < LEP2 excluded m 1/ LHC m 1/ LHC 600 LC LC LC LC Tevatron 200 Tevatron m m 0 HB, Belyaev, Krupovnickas, Tata: JHEP 0402, 007 (2004) Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

31 Sparticle reach for various integrated luminosity N jets 2 (with E T miss cuts, optimized) fb -1 (ofit) 1000 fb -1 (ofit) m~ Z1 = 700 GeV m Z1 ~ = 400 GeV m 1/ fb -1 (ofit) 10 fb fb -1 ~ m Z1 = 200 GeV 1 fb fb -1 (ofit) fb fb fb TeV 14 TeV m h = 114 GeV m 0 HB, Barger, Lessa, Tata (2009) Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

32 Direct and indirect detection of neutralino DM 1600 msugra, A 0 =0 tanβ=10, µ> msugra, A 0 =0, tanβ=50, µ< not LSP Z ~ 1 LHC DD µ 1200 not LSP Z ~ 1 DD m 1/ m 1/ LHC 600 LC LC LC LC500 µ 200 TEV LEP no REWSB 200 LEP no REWSB m 0 Φ(p - )=3x10-7 GeV -1 cm -2 s -1 sr -1 Φ(γ)=10-10 cm -2 s -1 Φ sun (µ)=40 km -2 yr -1 (S/B) e+ =0.01 m h =114.4 GeV m 0 Φ(p - ) 3e-7 GeV -1 cm -2 s -1 sr -1 (S/B) e+ =0.01 Φ(γ)=10-10 cm -2 s -1 Φ sun (µ)=40 km -2 yr -1 m h =114.4 GeV 0<Ωh 2 <0.129 σ(z ~ 1 p)=10-9 pb Φ earth (µ)=40 km -2 yr -1 σ(z ~ 1 p)=10-9 pb 0< Ωh 2 < HB, Belyaev, Krupovnickas, O Farrill: JCAP 0408, 005 (2004) Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

33 Issues in early search for SUSY: beam energy σ g ~ g ~ (fb) LHC Start-up m g ~ = 300 GeV m g ~ = 400 GeV m g ~ = 500 GeV NLO LO m q ~ ~ 10 TeV s (TeV) Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

34 Early SUSY discovery at LHC with just 0.1 fb 1? To make E T cut, complete knowledge of detector needed dead regions hot cells cosmic rays calorimeter mis-measurement beam-gas events Can we make early discovery of SUSY at LHC without E T? Expect SUSY events to be rich in jets, b-jets, isolated ls, τ-jets,... Use multiplicity of isolated muons rather than E T HB, Prosper, Summy, PRD77, (2008); HB, Lessa, Summy, PLB674, 49 (2009) HB, Barger, Lessa, Tata, JHEP0909, 063 (2009) Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

35 D0 saga with missing E T Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

36 Possible SM sources of multi-muon events: s = 10 TeV σ (fb) j j bb W j Susy (SPT2) Q 2 = s/36 Q 2 = s Q 2 = 4s NLO s = 10 TeV Z j bbbb tt VV Susy (SPS1a ) j = u, d, s, c, g V = W +-, Z bb Z bb W bb tt tt Z Jet and b cuts: R > 0.4 η < 4.0 p T > 20 GeV LO pdf: CTEQ5L NLO pdf: CTEQ6L tt W VVV tt VV tttt VVVV Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

37 Simple cuts: 4 jets plus isolated muons: no E T -cut SPT2 point: (m 0, m 1/2, A 0, tanβ, sgn(µ)) = (450 GeV, 170 GeV, 0, 45, +1) note: dis-allowed by Z 1 CDM but allowed for mixed a/ã CDM σ (fb) OS SS Signal OS SM OS Signal SS SM SS SM Background tt + jets tttt tt + VV tt + W tt + Z + jets VV VVV VVVV W + jets Z + jets QCD jets bb bbbb bb tt bb + W bb + Z + jets Signal (SPT2) n(µ) Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

38 Require n(jets) 4 and µ + µ pair: no E T OS(µ), N jets 4 (no E T miss cuts) 1 fb fb fb fb m 1/ m 0 Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

39 Case of isolated µ + µ events Background SPT2 SPS1a 25 σ/bin (fb/gev) m(µ + µ ) msugra = (450, 170, 0, 45, +1) Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

40 Require n(jets) 4 and µ ± µ ± pair: no E T SS(µ), N jets 4 (no E T miss cuts) 1 fb fb fb m 1/ m 0 Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

41 Require n(jets) 4 and 3µ: no E T N(µ) = 3, N jets 4 (no E T miss cuts) 1 fb m 1/ m 0 Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

42 Cuts optimized with n(jets) 2 and n(µ) 2: no E T N(µ) 2, N jets 2 (no E T miss cuts, optimized) 1 fb fb fb fb m 1/ m 0 Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

43 Randall-Tucker-Smith dijet signal propose simplest thing: can we see SUSY in di-jet channel knee-jerk reaction: no, QCD di-jet BG is too large reality: SUSY di-jets from squark pair production: do not lie in one plane L. Randall and Tucker-Smith exploit this: PRL101 (2008) φ(j1, j2) α = E T (j2)/m(jj) MT2(jj+ E T ) (though not needed) Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

44 Randall-Tucker-Smith dijet signal σ/bin (fb/0.02) 1e Total Background Signal (350,500,0,45,+) QCD dijet W + jj Z + jj tt α (m 0, m 1/2, A 0, tan β, sign(µ) = 350, 500, 0, 45, +1) at s = 10 TeV We require that E T (j 1 ) + E T (j 2 ) > 700 GeV, but make no restriction on E T. Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

45 Cuts optimized for Randall-Tucker-Smith dijet signal 550 N leps = 0, N jets = 2 (no E T miss cuts, optimized) fb fb fb fb -1 m 1/ m 0 HB, Barger, Lessa, Tata, arxiv: Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

46 Precision measurements at LHC M eff = E T + E T (j1) + + E T (j4) sets overall m g, m q scale m(l l) < m ez2 m ez1 mass edge m(l l) distribution shape combine m(l l) with jets to gain m(l lj) mass edge: info on m q further mass edges possible e.g. m(l ljj) Higgs mass bump h b b likely visible in E T + jets events in favorable cases, may overconstrain system for a given model methodology very p-space dependent some regions are very difficult e.g. HB/FP Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

47 Conclusions msugra: right or wrong? LHC slepton pair reach to m l 350 GeV LHC clean trilepton from W 1 Z 2 production (no spoiler modes) LHC reach via cascade decay to multi-leptons reach at s = 10 TeV vs. s = 14 TeV early reach via OS, SS, 3µ or jj events without need for E T cut Precision measurements possible for LHC Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

48 Paige, Hinchliffe et al. case studies: examined many model case studies in msugra, GMSB, high tanβ... classic study: pt.5 of PRD55, 5520 (1997) and PRD62, (2000) m 0, m 1/2, A 0, tan β, sign(µ) = (100, 300, 0, 2, 1) in GeV dominant g g production with g q q L qq Z 2 q 1 q 2 l 1 l q1 q 2 l 1 l 2 Z1 (string of 2-body decays) can reconstruct 4 mass edges; allows one to fit four masses: m ql, m ez2, m l, m ez1 to 3 12% can also find Higgs h in the SUSY cascade decay events if enough sparticle masses measured, can fit to MSSM/SUGRA parameters Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

49 M eff = E T (j1) + E T (j2) + E T (j3) + E T (j4)+ E T rough estimate of m g, m q can be gained from max of M eff dσ/dm eff (mb/400 GeV) M eff Atlas TDR (F. Paige) Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

50 m(l + l ) mass edge from Z 2 l + l Z1 kinematically, m(l + l ) < m e Z 2 m e Z 1 for Z 2 l + l (l + Z 1 )l, have m(l + l ) < m ez2 1 m2 l 1 m2 ez 1 m 2 m Z2 e 2 l < m ez2 m ez1 400 signal SM backg SUSY backg events/4 GeV/30 fb m ll Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

51 m(l + l q) mass edge from q q Z 2 q L q Z 2 q l ± l ql ± l Z Events/20 GeV/100 fb m llq Atlas TDR (F. Paige) Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

52 m(lq) mass edge from q q Z Events/20 GeV/100 fb m lq Atlas TDR (F. Paige) Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

53 m(lq) mass edge from q q Z Events/20 GeV/100 fb m llq Atlas TDR (F. Paige) Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,

54 m(b b) Higgs mass bump in SUSY jets + E T events signal SM backg SUSY backg Events/8 GeV m bb Atlas TDR (F. Paige) Howie Baer, Karlsruhe/Freudenstadt meeting, October 1,