Analysis of CP Violating Higgs in MSSM using

Analysis of CP Violating Higgs in MSSM using the process t->b. H+, H+->WH 1, H 1 ->bb (Generetor level analysis) Arun Nayak June 07 (TIFR) Les Houches meeting 1

LEP limits for CPX scenario : Because of the suppressed H 1 ZZ coupling, LEP can not exclude the presence of a light Higgs boson at low tanb (~ 3. 5 to 10) (LEP Higgs working group, hep-ex/0602042) June 07 Les Houches meeting 2

Higgs Production in CPX scenario : Phenomenological analysis by Ghosh, Godbole, Roy hep-ph/0412193 Because of the suppressed H 1 VV coupling one of the pseudo-scalar Higgs state is very light Since there is correlation between the mass of charged Higgs and that of the pseudo-scalar Higgs state in MSSM, that implies a light charged Higgs, with MH+< Mt. A suppressed H 1 VV coupling implies increase in H 1 H+W coupling because of the sum rule g 2 Hi. VV + | g. Hi. H+W- |2 = 1 And that implies an increase in branching ratio of H+->H 1 W ( > 47% over the entire range) The traditional decay mode H+->tn is suppressed over an order of magnitude. Hence both light H 1 & H+ can be probed using this process pp -> t + t~ + X b. H+ b~W W H 1 qq~(ln) ln(qq~) bb~ June 07 Les Houches meeting (hep-ph/0412193) 3

Generator Level Analysis : Total no. of events generated : 260000 for signal, (~30 fb-1) (generated using Pythia) 100000 for ttbb background(~30 fb-1) 2000000 for ttjj background cross section : signal cross section : parameters : F(CP) = 90 o, tan(b) = 5 M(H 1) = 51 Ge. V, M(H+) = 133 Ge. V, M(top) = 175 Ge. V * BR = 2 * 840 pb (ttbar NLO cross section) * 0. 01(BR(t->b. H+), using CPsuper. H) * 0. 567(BR(H+->H 1 W), using CPsuper. H) * 0. 99(BR(t->b. W)) * 0. 92(BR(H 1 ->bb), using CPsuper. H) = 8. 675 pb ttbb bkg cross section : 3. 285 pb Generated using Comp. HEP with preselections : Pt(b)>15 Ge. V, | (b)|<3. , Delta. R(b 1, b 2)>0. 3, ttjj cross section : 506. 99 pb Comp. HEP generated events with preselections : Pt(j)>15 Ge. V, | (j)|<3. , Delta. R(j 1, j 2)>0. 3 June 07 Les Houches meeting 4

Reconstructions and selections: 1) The events were generated with all possible W decay. So the first selection was, the events with one W decaying to electron or muon and neutrino and other W decaying to 2 light quarks were selected. The final state now contains 4 b quarks, 2 light quarks, one charged electron or muon, and missing energy due to neutrino. 2) jets reconstructed in pythia using parameters MSTU(51)=100 , no. of eta bin MSTU(52)=70 , no. of phi bin PARU(51) =5. 0 , total eta range PARU(52)=0. 1 , tower threshold PARU(53)=20 , jet Et threshold PARU(54)=0. 5 , cone size 3) events with Pt(b-quarks)>20 Ge. V, | (b-quarks)|<2. 5 selected. events with 6 or more jets has been selected, b-jets found by matching jets to b-quarks in 0. 4 cone, events with 4 b-jets found are considered. Remaining jets are treated as light quark jets. June 07 Les Houches meeting 5

4) Missing Et reconstructed from jets and charged lepton. By adding transverse momenta of jets and leptons vectorially. 5) one W reconstructed from charged lepton and missing Et. W boson momentum and energy calculated applying constraint on W mass as 80 Ge. V. This gives two solutions. Incase the solutions are imaginary, the event is neglected. 6) other W reconstructed from two light jets. Two light jets pair chosen from all possible combination of pair of light jets(if more than 2 light jets present) where the di-jet invariant mass lies within 65 to 95 Ge. V. If more than one jet pair fall within this mass window, the pair with invariant mass close to 80 Ge. V has been considered. 7) One top quark reconstructed from one of the b-jets and one of the W. total 12 combinations for quadratic ambiguity on W reconstruction from lepton and missing ET. The pair with inv. mass close to 175 Ge. V and within 150 to 200 Ge. V has been considered. June 07 Les Houches meeting 6

8) If the W used in 1 st top reconstruction is one of Ws from ln, then the other one has been considered as the wrong one and was neglected. On the other hand if the W used in 1 st top reconstruction is from light jets, then there is quadratic ambiguity in the 2 nd top reconstruction. 9) other top has been reconstructed from remaining 3 b-jets and remaining one W. (If quadratic ambiguity in W reconstruction and the W used in first top reconstruction is from light jets, then there are two combinations for 2 nd top. The one close to top mass has been considered) The inv. mass cut for bbb. W is 150 to 200 Ge. V. 10) H 1 reconstructed from two b-jets ( out of 3 -bjets in 2 nd top, total 3 possible combinations. ) 11) As seen in figure, peak of H 1(2 -bjets) mass distribution (distribution includes all backgrounds and also combinatorial backgrounds) lies around 50 Ge. V. So H 1( or 2 -bjets) with inv. mass close to 50 and falls within 35 to 65 Ge. V was used for H+ reconstruction. H+ has been reconstructed from H 1 and W (in 2 nd top). June 07 Les Houches meeting 7

Top & W mass reconstruction : The plots shows the reconstructed W mass for W->jj, and top mass for t->bln & t->bjj The combinations with inv. mass close to W or top mass has been considered. For t->bjj case, a mass window for jj inv. mass has been applied. 65 Ge. V < mjj < 95 Ge. V June 07 Les Houches meeting 8

Full simulation Jets using Iterative cone algorthim with cone size 0. 5 ( , f) June 07 Les Houches meeting 9

Top mass contd…. Distribution of top mass for t->bbbln & t->bbbjj , the combination with inv. mass close to top mass was considered. For t->bbbjj case, a mass window on jj inv. mass was applied. 65 Ge. V < mjj < 95 Ge. V June 07 Les Houches meeting 10

H 1 & H+ mass reconstruction : Reconstructed mass of H 1(2 b jets) & H+(2 b jets and W). Plot includes both signal and background events. All combinatorial backgrounds are also included in 2 b-jet (H 1) inv. mass Two b-jets with inv. mass close to peak of H 1 mass distribution and lies within 15 Ge. V around the peak were considered for H+ reconstruction. Plots for 30 fb-1 data June 07 Les Houches meeting 11

p. T difference of b-jets 2 lowest p. T b-jets June 07 Les Houches meeting 12

ttjj background analysis : selection of events similar to the signal events, one W-> lnu and other W->2 jets. Here the final state consists two b-jets from two top quark, atleast 4 light quark jets, one charged lepton. One W reconstructed from charged lepton and mising Et with W mass constraint. Other W from two light jets( making all possible combination), W mass window 65 to 95 Ge. V. One top reconstructed from one of two b jets and one of two (or three) W. Out of remaining light jets, two jets are selected as b-jets (to do a fake mis tagging) Assuming a maximum possiblity, two light jets which in combination with remaining one b and remaining one W gives the best top mass (within 175 ± 25 Ge. V), has been used as two b-jets. | |(two selected jets)<2. 5 Since I put a mass window on bbb. W, I select 2 -fake b-jets in this way. other light jets, if taken (mis tagged) as b-jets, will not survive in the selection cuts. This way of choosing two fake b-jets maximise the possible background from ttjj events. June 07 Les Houches meeting 13

Now there are 4 b-jets (including two fake b-jets), so one can reconstruct two top and H 1 and H+ using the same procedure as signal events with same mass window for top, W and H 1. Since two fake b-jets has been considered, we multiply a constant (corresponding to mistagging rate) to the final selected events. Assuming that mistagging rate per jet is 1%, for two jets to be mistagged, a constant 0. 01*0. 01 = 0. 0001 is multiplied to final selected ttjj events. We assume a 50% b-tagging efficiency per jet. So we multiply a factor of 0. 0625 (efficiency for tagging 4 b-jets) to signal and ttbb events and 0. 25(efficiency for tagging 2 b-jets) to ttjj events. Cross section of events after final selections : (i. e. no. of events in H+ mass distribution plots) signal : ~ 1. 063 fb ttbb bkg : ~ 0. 051 fb ttjj bkg : ~ 0. 0296 fb June 07 Les Houches meeting 14

Signal & background contribution : The major background to 2 b-jet invariant mass comes from the combinatorial backgrounds which may make the distribution very wide and have long tail. But a clear peak on top of it helps to chose the correct b-jet pair for the H+ reconstruction. Plots for 30 fb-1 data June 07 Les Houches meeting 15

Selection details Cross W ln | (b)| section & < 2. 5 pb W qq PT(b)> 20 Ge. V (no. of events) Atleast 6 jets 76100 14265 signal 8. 675 17803 4 b-jets matched to bquarks Reco W(l+MET) & W(2 jets) 65<mjj<95 6579 2634 Top H 1 (b-jet +W) 150<mt <200 (3 b-jets + W) 150<mt <200 (2 b-jets) 35<mh 1 <65 2529 515 510 1. 063 fb (260000) ttbb 3. 285 29037 15563 13124 6672 2242 2212 27 0. 051 fb (100000) ttjj ** 506. 99 (2000000) 564403 458889 25 389639 340402 (2 b-jets only) 166713 152193 5109 4675 0. 0296 fb multiply b-tag efficiency factor to final selected events to get the final state cross section June 07 Les Houches meeting 16

Simultaneous Top Reconstruction Minimize sqrt((mrec(t->b. W) – mtop)2 + (mrec(t->bbb. W) – mtop)2) signal Cross W ln | (b)| section & < 2. 5 pb W qq PT(b)> 20 Ge. V (no. of events) Atleast 6 jets 76100 14265 8. 675 17803 4 b-jets matche d to bquarks 6579 Reco W(l+MET) & W(2 jets) Top H 1 (3 b-jets + W) 150< mt < 200 (2 b-jets) 65<mjj<95 (b-jet +W) 150 < mt <200 2634 1793 892 879 1. 832 fb (260000) ttbb 3. 285 29037 15563 13124 6672 2242 1074 58 51 0. 104 fb (100000) June 07 35 < mh 1 < 65 Les Houches meeting 17

Backups June 07 Les Houches meeting 18

Separation of two b-jets used to reconstruct H 1 June 07 Les Houches meeting 19

June 07 Les Houches meeting 20