SciPost Submission Page
Search for heavy scalar or pseudoscalar states in $\mathrm{t \bar{t}}$ events at CMS
by Laurids Jeppe
Submission summary
| Authors (as registered SciPost users): | Laurids Jeppe |
| Submission information | |
|---|---|
| Preprint Link: | https://arxiv.org/abs/2411.18414v2 (pdf) |
| Date submitted: | April 24, 2025, 2:59 p.m. |
| Submitted by: | Jeppe, Laurids |
| Submitted to: | SciPost Physics Proceedings |
| Proceedings issue: | The 17th International Workshop on Top Quark Physics (TOP2024) |
| Ontological classification | |
|---|---|
| Academic field: | Physics |
| Specialties: |
|
| Approach: | Experimental |
Abstract
A search for scalar or pseudoscalar states decaying to a top quark-antiquark pair ($\mathrm{t \bar{t}}$), using $138\,\mathrm{fb}^{-1}$ of pp collision data taken at $\sqrt{s} = 13\,\mathrm{TeV}$ using the CMS detector, is presented. Events with one or two leptons are analyzed using the invariant $\mathrm{t \bar{t}}$ mass ($m_{\mathrm{t \bar{t}}}$) as well as angular and spin correlation observables. An excess in the data is observed for low values of $m_{\mathrm{t \bar{t}}}$, preferring a pseudoscalar over a scalar hypothesis. It is interpreted in terms of a generic model of (pseudo)scalar boson production, as well as a simplified model of a $\mathrm{t \bar{t}}$ bound state ($\eta_{\mathrm{t}}$), yielding good agreement with the data. Moreover, limits on the couplings of additional (pseudo)scalar bosons to top quarks are set.
Author comments upon resubmission
Thank you for your report and your recommendation to publish the proceedings. Please find below my answers to your comments.
- The couplings of the toy model are arbitrary and irrelevant since they only determine the cross section, which is the free-floating POI in the fit. The width is set to 7 GeV in the PAS presented at Saint-Malo, which I added to the text.
- The resolution is ~15% at the ttbar threshold, which I added to the text.
- This method is described in https://arxiv.org/abs/1310.3263, which I’ve added as a reference. It roughly works as follows: It is assumed that the lost or merged jet is one of the light jets from the hadronic W decay. A proxy for the hadronic top quark is defined as the vectorial sum of one b jet and one light jet, with a likelihood criterion used for the assignment. The four-momentum of this proxy is then rescaled as a function of its invariant mass, such that the peak of the reconstructed mtt distribution in simulation coincides with the generator-level peak. The same function is used for the rescaling in simulation and data. Since the procedure is somewhat complicated, and the lepton+3jets channel the least sensitive, I would prefer not to explain it in detail in the text.
- I’ve added a short explanation as well as two references.
List of changes
- Added the width of 7 GeV for the EtaT toy model considered.
- Added the mtt resolution of ~15%.
- Added a reference to https://arxiv.org/abs/1310.3263 regarding the energy correction in the lepton+3jets channels.
- Added a description of the spin correlation observables chel and chan: "Three-dimensional templates are constructed from $\mtt$ as well as two spin correlation observables $\chel$ and $\chan$. They are defined in the helicity basis for the top and antitop spins, corresponding to boosting the lepton four-momenta into the rest frames of their parent top quarks. The observable $\chel$ is the scalar product of the lepton directions of flight in the helicity basis, while $\chan$ is defined as a similar scalar product containing a negative sign in the top quark direction of flight. "
- Added two references for chel and chan and the helicity basis: https://arxiv.org/abs/2401.08751 and https://arxiv.org/abs/1508.05271.
- Fixed several spelling mistakes.