Exotica at LHCb More on LongLived Particles David

Exotica at LHCb: More on Long-Lived Particles David Kaplan (Johns Hopkins U) and Matt Strassler (Rutgers U) LHCb May 2009 M J Strassler 1

Question: What didn’t we cover yet? n We’ve considered events with 2 highly displaced vertices q From Higgs X X with X long-lived q From SUSY with LSP long-lived : n LSP visible, or LSP visible + invisible n The importance of the Hidden Valley scenario, and related classes of models, for LHCb is q Additional 2 -vertex examples in Higgs decays, SUSY decays q Enhanced likelihood of new long-lived particles q Enhanced likelihood of events with > 2 displaced vertices n In other words: q more particles per model AND more particles per event LHCb May 2009 M J Strassler 2

Hidden Valley Scenario MJS + Zurek hep-ph/0604261, hep-ph/0605193 MJS hep-ph/0607160, ar. Xiv/0801. 0629 We have very few limits on light particles that interact weakly with SM Could an entire sector of such particles await us? Such sectors easily generate neutral long-lived particles, often more than one type, and often produced with multiplicity > 2. If nature provides us with such phenomena, LHCb may scoop ATLAS/CMS LHCb May 2009 M J Strassler 3

Energy A Conceptual Diagram Entry into Valley via Narrow “Portal” Slow Decay Back to SM Sector via Narrow Portal Multiparticle Production in Valley Some Particles Unable to Decay Within Valley LHCb May 2009 M J Strassler Inaccessibility 4

General Predictions of HV Scenario n New neutral particles q q q n Often several of them in each model Exhibiting many possible decay modes Masses are usually of order (and above) the valley floor (mass gap) Long-lived resonances q q n MJS + Zurek hep-ph/0604261 Displaced vertices common n Great opportunity for LHCb if rates are sufficiently high Lifetimes for some new particles in the right range if barrier ~ Te. V scale Multiparticle production with unusual clustering q Valley dynamics can give events with of ~ 4, ~ 10, even ~ 30 vertices n Possibly highly correlated in direction n Possibly widely dispersed in direction LHCb May 2009 M J Strassler 5

Motivation for the HV scale n Why should nature set barrier and valley floor at these scales? q q q Some dynamics generates the electroweak scale – n maybe SUSY breaking, n maybe the true Planck scale, n etc. The same dynamics may easily generate similar scales in other sectors and couplings to those scales in the Te. V range Developing models of this type is very easy (too easy!) for theorists LHCb May 2009 M J Strassler 6

HV: typical lifetimes are long Examples: particles in 1 – 100 Ge. V range, coupled to SM by ~ (1 Te. V) scale n Dimension 6 effective operator [e. g. two SM fermions, two valley fermions] [ (20 Ge. V)5/(1 Te. V)4 ] yb 2 times phase space 10 -10 sec lifetime n Dimension 8 effective operator [e. g. two gluons, two valley gauge bosons] (100 Ge. V)9/(1 Te. V)8 times phase space 10 -10 sec lifetime n Dimension 4 effective operator with a 2 -loop mixing angle [e. g. photon mixing with valley gauge boson] 1 Ge. V * (mixing angle)2 times phase space 10 -12 sec lifetime Note the lifetimes are very sensitive to parameters; may easily be too long or too short to observe … but … LHCb May 2009 M J Strassler 7

HV: multiple metastable particles n Consider: q q q n QCD has many hadrons which are long lived on QCD time-scales against electromagnetic and weak decays n So may a hidden sector have particles with suppressed decays The lifetimes of these hadrons span many orders of magnitude n So may a hidden sector have particles with many lifetimes Many hadrons may be simultaneously produced n So may hidden sector particles be produced with high multiplicity A hidden valley that is no more complicated than QCD itself can easily put vertices into LHCb’s reach LHCb May 2009 M J Strassler 8

HV: production through decays/radiation Cross sections for multiple vertices may well be high enough for LHCb n Any new particle may decay wholly or in part into a hidden valley q q q n Higgs or any new scalar LSP (or LKP or LTP) New colored particles Quirkonium Z’ [cross section ~ few pb] [cross-section ~ few pb] [cross section ~ 10 - 100 fb] Also possible to radiate hidden valley particles off new particles LHCb May 2009 M J Strassler 9

HV: Examples 2 high-multiplicity vertices, similar lifetimes LHCb May 2009 M J Strassler 10

HV: Examples Jets. Pairs, Pion Kaon Pairs, Taus, Muons, Electrons… many low-multiplicity vertices, similar lifetimes LHCb May 2009 M J Strassler 11

HV: Examples Dijets or Trijets many high-multiplicity vertices, similar lifetimes LHCb May 2009 M J Strassler 12

HV: Examples many vertices, different multiplicities and lifetimes Et cetera LHCb May 2009 M J Strassler 13

A broad scenario, but… n The number of reasonable models is enormous n However, number of possible phenomena, while large, is smaller n Benchmark models, Monte Carlo tools appropriate for the range of phenomena are still under development (stay tuned) q q n Triggers, backgrounds are detector dependent So are benchmarks Discussions on this matter are welcome But for immediate purposes, simple models, simple studies using PYTHIA will probably suffice LHCb May 2009 M J Strassler 14

Simulation of Multiple Vertex Decays For many purposes, can use standard Pythia with small adjustments n q q Use a standard Pythia production process Alter only the decays using PYUPDA or PYSLHA calls to n n n Then Pythia will generate kinematically consistent events n q q n Add new particles Add new decays to old/new particles Caution: Pythia often cannot calculate total cross section correctly The new decays are distributed according to phase space Topology of the events determined by decay chain kinematics PYUPDA examples below; PYSLHA is better (newer, more stable) q Let’s try Higgs multiple vertices, two very simple examples LHCb May 2009 M J Strassler 15

Example: Higgs decays to many particles PYTHIA n n h 0 X X followed by X Y Y, Y fermions h 0 Y Y Y, etc. q q q q q PMAS(35, 1)=140. D 0 ! set H 0 mass MSUB(152)=1 ! MSUB(171)=1 ! MSUB(172)=1 ! MSUB(173)=1 ! MSUB(174)=1 ! ! ! MJS & Zurek 06 See also Chang, Fox & Weiner 05 gg H 0 [we use H 0 because h 0 is treated specially by Pythia] Z H 0 W H 0 VBF of H 0 VBF' of H 0 OPEN (UNIT=60, FILE=‘demo. pyupda. in', STATUS='OLD') CALL PYUPDA(3, 60) ! upload new-particle records CLOSE(60) LHCb May 2009 M J Strassler 16

The PYUPDA file [watch formatting!] Mass Change Old Particle 35 H 0 1 0 0 0 0 140. 00000 0 0 0. 399800 0. 300000 0. 000100 6000111 6001022 21 Width 0. 01000 Max Delta-E Lifetime 0. 10000 0. 00001 E+00 2 1 6000111 0 0 6001022 6001022 21 0 0 Decaying particle 6000111 ~v_pion 0 0 0 4. 00000 0. 01000 0. 10000 1. 00000 E-03 2 1 1 0 1. 000000 6001022 0 0 0 6001022 1 1 1 ~Uboson 0 0. 400000 0 0. 200000 Add New Particles 11 13 211 0 0 0 -11 -13 -211 0. 60000 0. 01000 0 0 0 0 LHCb May 2009 M J Strassler 0. 10000 5. 00000 E+00 2 1 17

The PYUPDA file [watch formatting!] v-pion H 0 U-boson Mass Change Old Particle 35 H 0 Decay flag 1 0 0 0 0 140. 00000 0 0 0. 399800 0. 300000 0. 000100 Branching Fraction 6000111 6001022 21 Width 0. 01000 Max Delta-E Lifetime 0. 10000 0. 00001 E+00 2 1 6000111 0 0 6001022 6001022 21 0 0 Decaying particle Particles in Final State 6000111 ~v_pion 0 0 0 4. 00000 0. 01000 0. 10000 1. 00000 E-03 2 1 1 0 1. 000000 6001022 0 0 0 6001022 1 1 1 ~Uboson 0 0. 400000 0 0. 200000 Add New Particles 11 13 211 0 0 0 -11 -13 -211 0. 60000 0. 01000 0 0 0 0 LHCb May 2009 M J Strassler 0. 10000 5. 00000 E+00 2 1 Electrons Muons Pions 18

The PYUPDA file [watch formatting!] U-boson H 0 Mass Change Old Particle 35 H 0 Decay flag 0 0 1 0 0 0 140. 00000 0 0 0. 399800 0. 300000 0. 000100 Branching Fraction 6000111 6001022 21 Width 0. 01000 Max Delta-E Lifetime 0. 10000 0. 00001 E+00 2 1 6000111 0 0 6001022 6001022 21 0 0 Decaying particle Particles in Final State 6000111 ~v_pion 0 0 0 4. 00000 0. 01000 0. 10000 1. 00000 E-03 2 1 1 0 1. 000000 6001022 0 0 0 6001022 1 1 1 ~Uboson 0 0. 400000 0 0. 200000 Add New Particles 11 13 211 0 0 0 -11 -13 -211 0. 60000 0. 01000 0 0 0 0 LHCb May 2009 M J Strassler 0. 10000 5. 00000 E+00 2 1 Electrons Muons Pions 19

In backup slides n Comment on “dark photon” branching fractions n Comment on simulating SUSY decays n Comment on models for which a more advanced Monte Carlo is needed, with examples (specific Z’ and quirkonium models) LHCb May 2009 M J Strassler 20

Conclusions n n n n Long-lived particles are very common in theoretical literature The Hidden Valley scenario illustrates that we may have underestimated the likelihood of long-lived particles. HV dynamics also can lead to a high multiplicity (per event) of such particles Tevatron experiments have not been able to put strong limits on long-lived neutral particles decaying in flight ATLAS/CMS will not find it easy in many cases due to q Triggering limitations q Reconstruction challenges LHCb therefore may have a special opportunity to discover new physics HLT 1 and secondary interactions appear to be the bottleneck Theorists are available to provide benchmark models, MC assistance LHCb May 2009 M J Strassler 21

Backup Slides LHCb May 2009 M J Strassler 22

Lepton-Jets comment n Getting the branching fractions right, as a function of mass, for a “hidden photon” (or “U-boson” or “dark photon”) requires careful use of data on off-shell photon decays. n Software to get the branching fractions right will be available before the LHC turns on q Contact e. g. Matt Reece (Princeton) or others LHCb May 2009 M J Strassler 23

SUSY decays Same idea as for Higgs q Set a standard PYTHIA card for MSSM production q Use PYUPDA/PYSLHA to add new hidden sector particles q Change the LSP from stable to unstable, set lifetime, and add decay modes to the new sector n e. g. Neutralino vpion + invisible n e. g. Stau + vpion + invisible q Add decay modes within the hidden sector and from the hidden sector back to visible particles n E. g. vpion b quarks n E. g. vpion U U ; U mu+ mu- LHCb May 2009 M J Strassler 24

More complex physical processes n n n Some processes cannot be simulated properly using phase-space cascade decays q Decay of particle into a parton shower has special kinematics q Other production mechanisms (e. g. quirk relaxation and annihilation) may produce multiple particles with odd distributions For these, specialized software is needed q Software for decays of a heavy particle to a v-quark pair in a QCD-like hidden sector exists q Software for decays to a v-quark pair in a higgsed hidden sector is more or less available However, for the immediate future this does not seem critical q The clustering of the new particles’ momenta is more or less obtainable from the PYUPDA approach LHCb May 2009 M J Strassler 25

HV: Effects on Narrow-Width Particles: n hep-ph/0604261 hep-ph/0605193 Possible big effect on Higgs q Long lived particles: H XX, X decays displaced new discovery mode n q not unique to HV!!! Chang Fox Weiner 05 / Carpenter Kaplan Rhee 06 High multiplicity decays: H XXX, XXXX, etc n not unique to HV!!! Chang Fox Weiner 05 n Also Z’, other resonances… n Big effect on SUSY, UED, Little Higgs – any theory w/ new global charge q LSP (or LKP or LTP) of our sector can decay to the valley LSP/LKP/LTP n n n Plus SM particles or Plus hidden particles which decay back to SM particles or Plus both hep-ph/0607160 Either the hidden particles or the LSP/LKP/LTP may be long-lived; q LSP may have high-multiplicity decays; q SUSY events have significantly reduced MET Generalizes well known work from 90 s [GMSB, Anomaly, Hidden Sector] q LHCb May 2009 M J Strassler 26

The PYUPDA file [watch formatting!] v-pion H 0 U-boson Mass Change Old Particle 35 H 0 Decay flag 1 0 0 0 0 140. 00000 0 0 0. 399800 0. 300000 0. 000100 Branching Fraction 6000111 6001022 21 Width 0. 01000 Max Delta-E Lifetime 0. 10000 0. 00001 E+00 2 1 6000111 0 0 6001022 6001022 21 Particles in Final State 0 0 Decaying particle Warning: Need to keep H 0 g g to not forbid g g H 0 6000111 ~v_pion 0 0 0 4. 00000 0. 01000 0. 10000 1. 00000 E-03 2 1 1 0 1. 000000 6001022 0 0 0 6001022 1 1 1 ~Uboson 0 0. 400000 0 0. 200000 Add New Particles 11 13 211 0 0 0 -11 -13 -211 0. 60000 0. 01000 0 0 0 0 LHCb May 2009 M J Strassler 0. 10000 5. 00000 E+00 2 1 Electrons Muons Pions 27

q q Q Q : v-quark production v-quarks q Q Z’ q Q Same for g g new neutral particles of spin 0 or 2 Analogous to e+e- hadrons LHCb May 2009 M J Strassler 28

qq QQ v-gluons q Q Z’ q Q Analogous to e+e- hadrons LHCb May 2009 M J Strassler 29

qq QQ q Z’ q v-hadrons Q Q Analogous to e+e- hadrons LHCb May 2009 M J Strassler 30

qq QQ Some v-hadrons are stable and therefore invisible v-hadrons q q Z’ Q Q But some vhadrons decay in the detector to visible particles, such as bb pairs, qq pairs, leptons etc. Analogous to e+e- hadrons Same structure for gg H v-quark LHCb May 2009 pairs M J Strassler 31

Cross sections and lifetimes? n The basic idea of a new particle decaying to a shower of hidden valley particles in a hidden sector is general q If these particles are long lived, get a shower of clustered displaced vertices q If these particles decay promptly to b’s, get a shower of many B’s q Could be highly collimated OR widely dispersed n Typically Z’ production rates are too small for LHCb But other resonances (including Higgs) could have similar decays So could the LSP, or even new exotic colored particles q In these cases, cross sections could be very large (1 – 10 pb) n n LHCb May 2009 M J Strassler 32

Quirks and v-glueballs Ge. V-Te. V–confinement quirk production/annihilation MJS + Zurek hep-ph/0604261 Juknevich, Melnikov, MJS hep-ph/09…… YM glueball spectrum Morningstar Peardon 99 Quirk: Matter charged under SM and hidden confining group… n Hidden confining string cannot break Quirkonium n Quirk loops induce couplings of SM and hidden gauge bosons See also Low-confinement-scale “quirks”: Kang, [Nasri], Luty 2006, 2008 g g g q g g photon Q photon q Q quirks g g v-gluons LHCb May 2009 M J Strassler v-glueballs g 33