1. Search for New Light Bosons in Models with Hidden Sectors at CMS Yuriy Pakhotin, Jim Pivarski, Alexei Safonov, Aysen Tatarinov LPC Topic of the Week November 16th, 2011

2. Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Motivation: Satellite Experiments 2 Nature 458 (2009) 607-609 arXiv:1109.0521v1

3. Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Motivation: NMSSM 3 Both motivations suggest a new light boson weakly coupling to charged SM particles Not observed in direct production  weak coupling leads to small cross section, need very high luminosity Heavy particles made at hadron colliders might decay to the new light boson, that eventually decays to SM particles

4. D0 Collaboration, Search for dark photons from supersymmetric hidden valleys Phys.Rev.Lett.103 (2009) 081802, arXiv:0905.1478 D0 Collaboration, Search for events with leptonic jets and missing transverse energy in ppbar collisions at sqrt(s)=1.96 TeV Phys.Rev.Lett.105 (2010) 211802, arXiv:1008.3356 CDF Collaboration, Search for Anomalous Production of Events with a W or Z Boson and Additional Leptons in 5.1 fb-1 of Data public note 10526 ATLAS Collaboration, A Search for Lepton-Jets with Muons in pp Collisions at sqrt(s) = 7 TeV in the ATLAS Detector conference report ATLAS-CONF-2011-076 CMS Collaboration, Search for Light Resonances Decaying into Pairs of Muons as a Signal of New Physics JHEP 07 (2011) 098, arXiv:1106.2375 Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Experimental Searches at Colliders 4

5. Search for groups of collimated leptons cascades of low-mass particles expected to be collimated due to high-momentum boost use only muons for now, but neither require nor exclude other particles search for m1 or m2 resonance peaks  resolve combinatorics within each group by finding the most consistent combination of dimuon masses Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Signature: Lepton Jets 5

6. Desires: Sensitivity to many kinds of new physics  Standalone light boson or whole shower Results interpretable in other models Implementation: Identify muon candidates Cluster nearby muons into “muon jets” Categorize events by topology  number of muon jets  number of muons in jets Do not use:  Isolation (there could be electrons, pions, taus nearby) Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Analysis Strategy 6

7. Challenge: muons in Lepton Jets are crossing each other StandAloneMuons, and hence GlobalMuons, are inefficient when crossing Solution: TrackerMuon is used pT > 5 GeV/c, |η| < 2:4, arbitrated seg. >= 2, tracker hits >=8, χ2/Ndof < 4 Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Muon Reconstruction 7

8. Efficiency to reconstruct dimuon using TrackerMuons analysis cuts applied Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Tracker Muons 8

9. Only single-muon non-isolated barrel trigger doesn’t depend on whether muons cross in the muon system Accepting endcap-triggered events would introduce a strong efficiency dependence on dimuon kinematics; therefore, we require at least one muon above trigger threshold pT > 15 GeV in barrel |η| < 0.9 minimal unprescaled muon trigger in 2010 Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Trigger 9

10. Start from highest pT muon Add a new muon (  i+1 ) to the jet if the jet contains another muon (  i ) and if: Opposite charge Common vertex Invariant mass(  i,  i+1 )< 9 GeV Ignore single muons Advantages: Efficient independent of boost Insensitive to pile-up Mass cut can be dialed for optimal distribution of backgrounds among categories Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Clustering Muons into Jets 10

11. Data: 35 pb -1 of 2010 LHC data Inclusive single muon trigger p T >15 GeV Offline: Require at least 1 muon with p T >15 GeV, |  |<0.9 Identify all other muons with p T >5 GeV, |  |<2.4 Reconstruct muon jets and categorize Target the case with new bosons produced on-shell: Consider all pairings of muons and pick one with most consistent mass Plot pair masses Signal will be on the diagonal Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Selections 11

12. Search for three different topologies Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Topologies 12 namedescription Lead  -Jet p T Backgrounds R12R12 Single dimuon+X>80 GeV/c 2  ’s from a b-jet, Drell Yan R14R14 Single quadmuon+Xno explicit cut 2  ’s from a b-jet + 2 fakes R 2 2 2 Two dimuons+Xno explicit cut bb-bar+X, 2  ’s from each b R N 5 + All other categoriesno explicit cutRare, from bb-bar+X/fakes

13. Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Dimuon Backgrounds 13

14. Fairly elaborate and fully data driven procedure was developed Use templates to build multidimensional distributions predicting background shapes  Non-diagonal side-band data is used to fit for normalization Background shape templates must be individually constructed due to contribution of different physics sources Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Background Modeling 14

15. Topology R 1 2 “One muon jet with two muons” Apply p T > 80 GeV cut on “single dimuon” to reduce background Derive shape from high-statistics low-pT data, test in medium-pT, and use in high-pT signal search Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Single Dimuon 15

16. Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Two Dimuons 16  =

17. Topology R 1 4 “Four Nearby Muons” Dominant SM backgrounds: decays-in-flight and mis- reconstruction (fakes) Simulate fake muons by putting non- muon tracks into mu-jets:  Background enriched: 2 muons, 2 tracks  Control: 3 muons, 1 track  Signal: 4 muons Plots of control region with template shape overlaid Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Four Muons 17

18. Considering the case of m2 -> m1m1 -> 4µ with m1 off- shell but m2 on-shell, template for the four-muon mass was prepared important case for search for h dark   dark  dark the second bump is due to events with two muons from a J/ψ boosting the distribution for these events to higher mass Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Four Muons: Mass Shape 18 Background enriched: 2 muons, 2 tracks Control: 3 muons, 1 track

19. Muon system: Dominant factor for p T (  )<150 GeV,  ~92-95%  Losses due to non-instrumented gaps and muon overlaps in the muon system (endcap) Tracking: Important for high momentum dimuons  Losses when tracks have large overlap of hits in the tracker  m(  )~0.4-0.6:  ~85/75% at p T (  )=250/350 GeV  Elsewhere  ~95/85% at p T (  )=250/350 GeV High multiplicity muon jet efficiency factorizes into efficiencies of finding dimuons in it Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Signal Efficiencies 19

20. Single dimuon+X: narrow bump would indicate signal The two-dimuon+X (right): zero diagonal events Background normalization assumes that new resonances are produced on-shell Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Data in Signal Regions -1 20 No data events in any of the higher multiplicity topologies

21. Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Two Dimuons Event 21 The two dimuons are not close enough in mass to be decays of the same on-shell m1 boson (and therefore belong to the off-diagonal control) mass = 2.7 GeV mass = 1.5 GeV

22. Quadmuon +X: zero diagonal events Background normalization assumes that new resonances are produced on-shell  If produced off-shell, signal can be off-diagonal  4  mass (search for h dark   dark  dark ) Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Data in Signal Regions -2 22 No data events in any of the higher multiplicity topologies

23. Consistent with two true muons and two tracks misidentified as muons Likely a bb-bar event with  in one of the b-jets Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 The Quadmuon Event 23

24. Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Systematic Uncertainties 24

25. Use three simplest topologies to set “conservative” model independent limits: Dimuon+X Two-dimuon+X Quadmuon+X Limits of applicability: Mean p T (  -jet)≤250GeV Instructions for future uses: Follow analysis steps to calculate branching and acceptance for specific final state Assuming an ideal detector and compare with the limit plot Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Model-Independent Interpretation 25

26. MSSM with squarks/ gluinos accessible by LHC: MSSM LSP is 400 GeV neutralino decaying to dark neutralino (300 GeV) and light  dark /h dark MSSM LSP is squark decaying to q and light dark fermion and  dark /h dark NMSSM higgs: pp  h 1  a 1 a 1  m(a 1 )<2m(  ), m(h1)~100 GeV – Moderate boost (  ~50/3) Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Benchmark Models 26

27. MSSM neutralino decays to a heavy stable dark neutralino and  dark /h dark (h dark  dark  dark ) Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Model Specific Limits (1) 27 Theory reference: Y. Bai and Z. Han, Phys. Rev. Lett. 103 (2009) 051801.

28. MSSM LSP is a squark decaying to a quark, light dark fermion and either  dark (left) or h dark (right) Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Model Specific Limits (2) 28 Theory reference: J. T. Ruderman et al., JHEP 04 (2009) 014. Y. Bai and Z. Han, “Measuring the Dark Force at the LHC”, Phys. Rev. Lett. 103 (2009) 051801.

29. Limits derived using two-dimuon topology Scan over NMSSM parameter space for m(a 1 )<2m   WMAP and LEP constraints applied Limits are blurred due to varying acceptance as function of m(h 1 ) and m(a 1 ) Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 NMSSM Higgs Limits with 2010 data 29 Following Belyaev, Pivarski, A.S. et al., Phys. Rev. D 81 (2010) 075021.

30. Most part of NMSSM Higgs masses phase space is expected to be restricted with full 2011 data Acceptance is reduced due to lowest muon trigger threshold 40 GeV Isolation might be required to apply to keep negligible background contribution into signal region Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Estimated NMSSM Higgs Limits with 2011 data 30 limit projection to 5 fb -1

31. Search with CMS of cascade decays in a hidden sector using lepton (muon) jets in final states is presented No discovery, but new ground in sensitivity to SUSY models with light squarks and gluinos Complementary to Tevatron searches sensitive to SUSY with light charginos Results allow a quazi model independent interpretation for phenomenological use NMSSM light CP-odd Higgs search will surpass the Tevatron sensitivity when new data is analyzed New results with ~5 fb-1 of data collected in 2011 are to come soon. Stay tuned! Yuriy Pakhotin (Texas A&M)LPC Topic of the Week, November 16th, 2011 Conclusions 31