Searching for Top Squarks at the LHC in Fully Hadronic Final State

Searching for Top Squarks at the LHC in Fully Hadronic Final State

Bhaskar Dutta    Teruki Kamon    Nikolay Kolev    Kuver Sinha    Kechen Wang  Department of Physics, Texas A&M University, College Station, TX 77843-4242, USA
 Department of Physics, Kyungpook National University, Daegu 702-701, South Korea
 Department of Physics, University of Regina, SK, S4S 0A2, Canada
Abstract

We pursue a scenario where the lighter top squark (stop) mass is accessible for the Large Hadron Collider (LHC) in the near future, while gluinos and first two generation squarks are too heavy. At TeV, we investigate the identification of stops which decay predominantly into a top quark and the stable lightest supersymmetric particle. We use a simple kinematical variable, , to reconstruct two top quarks which are pair-produced from the stops, in the fully hadronic channel. The dominant Standard Model (SM) background for this signal stems from plus jets, with one top quark decaying into , where the lepton is undetected and the produces missing transverse momentum. The lepton identification efficiency is thus crucial in order to estimate the background correctly. We identify kinematical variables to reduce the SM background. We find that it is possible to achieve signal and background cross-section at similar levels for stop masses around GeV for a mass of GeV.

MIFPA-12-25

I Introduction

Experiments at the Large Hadron Collider (LHC) have recently reported preliminary evidence for a Higgs-like particle, with mass in the region of GeV ATLAS:2012ae (); Chatrchyan:2012tx ().

In the Standard Model (SM), higher loop corrections to the Higgs mass are quadratically divergent; to avoid fine-tuning, new physics should appear around a scale of TeV and cut off the divergences. The problem is most severe in the case of the one-loop correction from the top sector, since other contributions to the Higgs mass are suppressed by gauge or smaller Yukawa couplings. Thus, reducing fine-tuning in the SM leads minimally to the conclusion that there should a partner for the top quark in the sub-TeV regime which is responsible for the cancellations.

The most widely studied mechanism for canceling the divergences is supersymmetry, and in particular the dangerous top quark loops are canceled by the scalar superpartner of the top quark, called a stop ().While the LHC has already started to put constraints on the first two generation squarks-gluino plane, the paucity of strong constraints on stop masses is due to the small production cross-section of stop pairs and a huge background from top quark production. The current constraints are: (a) for a directly produced stop going to top quark plus next to lightest neutralino, with neutralino decaying to gravitino plus , ATLAS excludes stop masses upto GeV using fb Aad:2012cz () (b) at the Tevatron, the stop mass constraint is about GeV Abazov:2008rc (); TevatronConstraintCDF ().

In this paper, we will probe a technique for stop searches in the following decay mode

(1)

where the is the lightest neutralino, which we will take to be the lightest supersymmetric particle (LSP). In -parity conserving models, the LSP is the main source of missing energy in the event. We will always be speaking about the lightest top quark superpartner , which we will hereafter call . We will not make any assumptions about the spectrum, except that the above decay mode is kinematically allowed and dominant.

The main challenge in such searches is the fact that the LHC is a top quark factory and distinguishing top quarks produced from stop decay, as opposed top quarks produced directly, can be very challenging. There are several established techniques of probing the system or identifying top quarks:

  , the Razor, and : These techniques rely on the identification of two hemispheres to maximize sensitivity in searches for a pair of heavy colored objects and their cascade decays. These are inclusive searches that do not rely on the reconstruction of top quark(s); they have larger signal acceptance, but larger background hemispheres1 (); hemispheres2 (); hemispheres3 (); hemispheres4 (); hemispheres5 (); hemispheres6 ().

   Methods that reconstruct a top quark or two top quarks, followed by cuts on topology or kinematics to reduce the background. These methods have lower background, and lower acceptance of signal.

We choose to explicitly reconstruct the two-top quark system. Within this class of searches, several options are available:

   : This is the invariant mass of trijet combinations with highest vectorially summed . has been used in top quark studies at CMS Chatrchyan:2011ew () and CDF CDFtopstudy ();

   Kinematic Fits: One minimizes in dijet and trijet invariant masses, matching to and top quark masses; for example Blyweert:2012bq ();

   Top Taggers: See Plehn:2011tg () for a recent review;

  Bi-Event Subtraction Technique (BEST): This method has been used to identify top quark system in Dutta:2011gs (). Jet combinatorics is reduced by mixing events in hadronic decay chains.

For the two top quark system in Eq. 1, the highest jets are most likely to be from the top quark, if it is signal. Intuitively, should work well in such a system (note that such an assumption is untrue in a SUSY environment with multiple superpartners undergoing decays, where the highest jets are probably not from the top quark; BEST works better there).

In Plehn:2012pr (); Kaplan:2012gd (), top taggers have been used to reconstruct top quarks coming from stop decay, while in Alves:2012ft (), shape analyses of -related distributions have been used to probe the stop system when the stop-LSP mass difference is degenerate with the top quark mass.

At the LHC, it is expected that the existence of stops will be indirectly established initially using inclusive jets + single lepton + analysis, however once any excess is observed the direct evidence of the stop can be established though the existence of top quarks in the signal. Our goal in the paper is to establish the existence of two top quarks in the final states along with the missing energy in all hadronic channel. In order to make our analysis realistic we use PGS4  detector simulation pgs () and consider jets, jets and jets, with as well as single top jets backgrounds. Interestingly we noticed that jets contribution to the background is comparable to jets.

Our finding is that simple kinematical selections with the variable is an effective tool for stop searches. At TeV, we achieve background and signal cross-sections at comparable levels for stop masses around GeV.

The outline of this paper is as follows. In Section II, we outline our search strategy. In Section III, we give our results in detail. The results are also summarized in Table 5 appearing at the end of the paper. We end with our conclusions.

Ii Search Strategy

We consider the fully hadronic mode

(2)

We investigate samples with at least four non -tagged jets and at least two -tagged jets, along with large missing energy. Signal events are generated with ISAJETisajet () + PYTHIApythia (), background events are generated with ALPGENMangano:2002ea () + PYTHIA, followed by PGS4  detector simulation.

The main source of missing energy for the background are neutrinos coming from the leptonic decay of a , while for the signal the dominant source of missing energy is the neutralino. Clearly, after the missing energy cut, the most critical factor affecting the discrimination of signal over background in the fully hadronic mode is the lepton veto efficiency. Due to imperfections of the lepton veto, events with leptonic decay could be a dominant source of background.

Figure 1: [Left] schematic diagram of the signal. The stop pair gives rise to and neutralinos, which are the main source of . In the fully hadronic mode, the top quarks decay into trijet systems. “System A” is the trijet system containing the leading jet and reconstructed using , while the remaining jets are called “System B”. [Right] background after lepton veto where the lepton is undetected. The main source of here is the neutrino from decaying leptonically. The associated lepton passing the veto is termed a “lost lepton”.

Our method is to reconstruct a top quark using the trijet invariant mass , use kinematic correlations between the constituents of the two systems and to improve the reconstruction of the pair of top quarks and finally apply again to identify the second top quark. We describe these steps below, before showing our results in the next section.

  We use twice. First, combinations of three jets are made in the sample, keeping one -tagged jet and two untagged jets in each trijet combination. Next, the trijet combination with the largest vectorially summed transverse momentum is chosen. The invariant mass of this trijet combination is defined as . It approximates the mass of the hadronically-decaying top quark. Similarly, we find a leading trijet combination and invariant mass . Associated with , we also define , which is the invariant mass of the two untagged jets in the trijet combination.

Using , we identify a first top quark, which we call “System A”. This is done by calculating for the trijet and dijet combination corresponding to the leading combination and also for the leading combination , with a mean top quark mass of GeV and width of GeV, and a mean mass of GeV and a width of GeV. The combination with the lowest is then taken to represent System A. We call this combination .

We note that this analysis allows for more signal events in the identification of System A. We show the results of this analysis in Section III.3.

After the identification of System A, we classify the remaining -jet and non-tagged jets to be “System B”; thus, we would denote them as . We employ various cuts on azimuthal angles between jets and , and between and . These are motivated by the fact that for signal, the main source of missing energy is the neutralino, while for the background, the main source of missing energy is the neutrino coming from the leptonic decay of the , or from jet mismeasurement. Thus, for example, for the background, is aligned along , as is clear from the schematic diagram shown in Figure 1. For the stop decay, however, the correlation between the in the form of neutralino and the is far weaker. The results of this analysis are shown in Section III.4.

At the final stage, we employ again to identify the second top quark, System B. The result of this analysis is shown in Section III.5.

Iii Results

In this section, we describe our selection criteria and the cross sections after every stage of cuts (see Table 5).

iii.1 Background

We generate the following backgrounds with ALPGEN+ PYTHIA+ PGS4: jets with , single top jets, jets and jets with upto . The background cross-sections are given in Table 1. While Ref. Plehn:2012pr () considers background with jets in the stop analysis, we find that jets contribution is comparable to jets.


Background
Others
Cross-section
Table 1: Main sources of background. “Others” includes single top jets, jets and jets with . All cross-sections are in fb.

iii.2 Baseline Cuts: jets, Lepton veto,

Our baseline selection cuts are:

, and at least two loosely tagged -jets.

The leading jet has GeV in , and all other jets have GeV in .

Lepton veto: We reject isolated electrons and muons with GeV in . The isolation criteria are GeV with .

veto: We also reject any hadronically decaying with GeV in . We assume a identification efficiency of and a fake rate of .

GeV.

iii.3 : Tagging Top System A

In this section, we use to tag the top quark in System A, after a cut to further reduce SM background. The value of the cut is determined by maximizing the significance for each choice of mass. This is shown in Table 2.




Table 2: cut for various choices of masses. All masses are in GeV.

As described in Section II, we identify System A by using . Figure 2 shows the comparative distributions of and . We improve the top tagging by approximately in signal events in the top quark mass region.

Figure 2: Distributions of and . The inset shows the distribution after mass window cut. A gain of in signal is obtained by using . The luminosity is fb.

We next perform a mass window cut on , taking and a top quark mass window cut on , taking .

We show the distribution after mass cut in the inset of Figure 2.

We now proceed to probe the constituents of the “other top quark” in System B.

iii.4 Angular and Cuts: Kinematic Correlations between and Jets

We denote the remaining -jet and non-tagged jets as . We clean up the system with various angular and cuts, as mentioned in our search strategy in Section II. The cuts values arechosen based on Figures 3 and 4:

and , where refer to the first and second leading jets in System B, respectively.

   We choose optimal cut values for different masses (see Table 3).



Table 3: cuts for various choices of masses. All masses are in GeV.
Figure 3: Distributions of for background and signal . We cut at . Here, and denote the , leading jet, and next leading jet of System B. The luminosity is fb.
Figure 4: Distribution of for background and signal . We cut at   GeV. The luminosity is fb.

After the above cuts, we revert to the trijet in System A with similar angular cuts between missing energy and the -tagged jet as well as non-tagged jets. These angular cuts are efficient in reducing events with “lost leptons”. The cuts are chosen based on Figure 5:

   and , where refer to the first and second leading jets in System A, respectively.

Figure 5: Distributions of , , and for background and signal . We cut at and . Here, and denote the , leading jet, and next leading jet of System A. The luminosity is fb.

iii.5 Tagging Top System B

As a last step, is applied in System B, followed by a mass window cut on (). The distribution is shown in Figure 6. Our final results with are tabulated in Table 5.

We note that for the point we additionally impose . Also, for the point , the mass window cut on of System B was taken as , while the top quark mass window was taken as .

Figure 6: Distribution of in System B, after requiring . Displayed are: “other” sources of background (single top jets, jets and jets with ), total background including jets, and total background plus signal for our reference point . The luminosity is fb.

Iv Summary and Conclusion

In this paper, we have explored a search strategy for a light stop at the LHC at TeV, using variable in the fully hadronic channel. The gluino and the first two generation squarks are assumed to be too heavy to be produced significantly at the LHC.

We first performed to identify a top quark system (System A). Next, we performed again to identify the second top quark (System B), along with a series of kinematical cuts to reduce the SM backgrounds. Throughout this study, we used PGS4  detector simulation and considered jets, jets and jets, with as well as single top jets backgrounds. Interestingly we noticed that jets contribution to the background is comparable to jets.

Tables 4 and 5 are a summary of the search performance for various choices of stop and neutralino masses.



Table 4: Final significances for various choices of masses. All masses are in GeV. The luminosity is fb.

In summary, a simple kinematical reconstruction technique using the variable is an effective tool for stop searches. We find that at TeV it is possible to reduce background down to a level of signal cross-section for stop masses around GeV for a mass of GeV.

V Acknowledgements

We would like to thank Matt Buckley, Christopher Lester, Michelangelo Mangano, Tilman Plehn, Michael Spannowsky, Daniel Stolarski and Michihisa Takeuchi for very useful discussions and correspondence, and Tai Sakuma for collaboration in the early stages of this work. We would like to thank John Conway for help with PGS4 . We would also like to thank David Curtin and Will Flanagan for correspondence. This work is supported in part by the DOE grant DE-FG02-95ER40917 and by the World Class University (WCU) project through the National Research Foundation (NRF) of Korea funded by the Ministry of Education, Science & Technology (grant No. R32-2008-000-20001-0).



Signal    jets    jets    Others




Initial

Baseline Cuts (Sec. III.2)         

GeV         

System A: (Sec. III.3)         

Angular and cuts (Sec. III.4)         

System B: (Sec. III.5)         


        


Significance





Initial

Baseline Cuts (Sec. III.2)         

GeV         

System A: (Sec. III.3)         

Angular and cuts (Sec. III.4)         

System B: (Sec. III.5)         


Significance





Initial

Baseline Cuts (Sec. III.2)         

GeV         

System A: (Sec. III.3)         

Angular and cuts (Sec. III.4)         

System B: (Sec. III.5)         


Significance



Initial

Baseline Cuts (Sec. III.2)         

GeV         

System A (Sec. III.3)         

Angular and cuts (Sec. III.4)         

System B: (Sec. III.5)         


Significance



Initial

Baseline Cuts (Sec. III.2)         

GeV         

System A: (Sec. III.3)         

Angular and cuts (Sec. III.4)         

System B: (Sec. III.5)         


Significance




Initial

Baseline Cuts (Sec. III.2)         

GeV         

System A: (Sec. III.3)         

Angular and cuts (Sec. III.4)         

System B: (Sec. III.5)         



Significance




Initial

Baseline Cuts (Sec. III.2)         


System A: (Sec. III.3)         

Angular and cuts (Sec. III.4)         

System B: (Sec. III.5)         



Significance



Table 5: Summary of effective cross sections (fb) for stop pair production and the SM background events in our stop search feasibility study. Masses and momenta are in GeV. “Other” sources of background include single top jets, jets and jets with . The significance is given at fb .

References

  • (1) G. Aad et al. [ATLAS Collaboration], “Combined search for the Standard Model Higgs boson using up to 4.9 fb of collision data at TeV with the ATLAS detector at the LHC,” Phys. Lett. B 710, 49 (2012) [arXiv:1202.1408 [hep-ex]].
  • (2) S. Chatrchyan et al. [CMS Collaboration], “Combined results of searches for the standard model Higgs boson in collisions at TeV,” Phys. Lett. B 710, 26 (2012) [arXiv:1202.1488 [hep-ex]].
  • (3) G. Aad et al. [ATLAS Collaboration], “Search for scalar top quark pair production in natural gauge mediated supersymmetry models with the ATLAS detector in collisions at TeV,” arXiv:1204.6736 [hep-ex].
  • (4) V. M. Abazov et al. [D0 Collaboration], “Search for scalar top quarks in the acoplanar charm jets and missing transverse energy final state in collisions at -TeV,” Phys. Lett. B 665, 1 (2008) [arXiv:0803.2263 [hep-ex]].
  • (5) [CDF collaboration], CDF Public Note 9834.
  • (6) [CMS Collaboration], “Search for supersymmetry in all-hadronic events with ,” CMS-PAS-SUS-11-003.
  • (7) L. Randall and D. Tucker-Smith, “Dijet Searches for Supersymmetry at the LHC,” Phys. Rev. Lett. 101, 221803 (2008) [arXiv:0806.1049 [hep-ph]].
  • (8) S. Chatrchyan et al. [CMS Collaboration], “Inclusive search for squarks and gluinos in collisions at TeV,” Phys. Rev. D 85, 012004 (2012) [arXiv:1107.1279 [hep-ex]].
  • (9) C. Rogan, “Kinematical variables towards new dynamics at the LHC,” arXiv:1006.2727 [hep-ph].
  • (10) T. Aaltonen et al. [CDF Collaboration], “Top Quark Mass Measurement using mT2 in the Dilepton Channel at CDF,” Phys. Rev. D 81, 031102 (2010) [arXiv:0911.2956 [hep-ex]].
  • (11) H. Weber et al. [CMS Collaboration], “Search for supersymmetry in hadronic final states with MT2,” arXiv:1201.4659 [hep-ex].
  • (12) S. Chatrchyan et al. [CMS Collaboration], “Measurement of the Top-antitop Production Cross Section in Collisions at TeV using the Kinematic Properties of Events with Leptons and Jets,” Eur. Phys. J. C 71, 1721 (2011) [arXiv:1106.0902 [hep-ex]].
  • (13) CDF Collaboration, “A Measurement of Top Quark Mass Using MET + Jets Events with fb of CDF Data,” CDF public note 10433 (2011). Z. Ye [CDF and D0 Collaborations], “Top Quark Mass Measurements at the Tevatron,” arXiv:1107.4539 [hep-ex].
  • (14) B. Dutta, T. Kamon, N. Kolev and A. Krislock, “Bi-Event Subtraction Technique at Hadron Colliders,” Phys. Lett. B 703, 475 (2011) [arXiv:1104.2508 [hep-ph]].
  • (15) S. Blyweert [ATLAS and CMS Collaborations], “Top-quark mass measurements at the LHC,” arXiv:1205.2175 [hep-ex].
  • (16) T. Plehn and M. Spannowsky, “Top Tagging,” arXiv:1112.4441 [hep-ph].
  • (17) T. Plehn, M. Spannowsky and M. Takeuchi, “Stop searches in 2012,” arXiv:1205.2696 [hep-ph].
  • (18) D. E. Kaplan, K. Rehermann and D. Stolarski, “Searching for Direct Stop Production in Hadronic Top Data at the LHC,” arXiv:1205.5816 [hep-ph].
  • (19) D. S. M. Alves, M. R. Buckley, P. J. Fox, J. D. Lykken and C. -T. Yu, “Stops and MET: the shape of things to come,” arXiv:1205.5805 [hep-ph].
  • (20) PGS4 is a parameterized detector simulator. We use version 4 (http://www.physics.ucdavis.edu/~conway/research/software/pgs/pgs4-general.htm) in the CMS detector configuration.
  • (21) F. E. Paige, S. D. Protopopescu, H. Baer and X. Tata, “ISAJET 7.69: A Monte Carlo event generator for p p, anti-p p, and e+ e- reactions,” [hep-ph/0312045]. We use ISAJETversion 7.74.
  • (22) T. Sjostrand, S. Mrenna, and P. Skands, ”PYTHIA 6.4 Physics and Manual.” J. High Energy Phys. 05 (2006) 026. We use PYTHIA version 6.411 with TAUOLA.
  • (23) M. L. Mangano, M. Moretti, F. Piccinini, R. Pittau and A. D. Polosa, “ALPGEN, a generator for hard multiparton processes in hadronic collisions,” JHEP 0307, 001 (2003) [hep-ph/0206293].
Comments 0
Request Comment
You are adding the first comment!
How to quickly get a good reply:
  • Give credit where it’s due by listing out the positive aspects of a paper before getting into which changes should be made.
  • Be specific in your critique, and provide supporting evidence with appropriate references to substantiate general statements.
  • Your comment should inspire ideas to flow and help the author improves the paper.

The better we are at sharing our knowledge with each other, the faster we move forward.
""
The feedback must be of minimum 40 characters and the title a minimum of 5 characters
   
Add comment
Cancel
Loading ...
174764
This is a comment super asjknd jkasnjk adsnkj
Upvote
Downvote
""
The feedback must be of minumum 40 characters
The feedback must be of minumum 40 characters
Submit
Cancel

You are asking your first question!
How to quickly get a good answer:
  • Keep your question short and to the point
  • Check for grammar or spelling errors.
  • Phrase it like a question
Test
Test description