Abstract
We discuss a search for natural Supersymmetry in events with at least one bottom-quark jet using razor variables in proton-proton collisions at 8 TeV with the CMS detector. The event distribution in the plane defined by the razor variables $R^2$ and $M_R$ is studied, searching for a peaking signal on top of a smoothly falling standard model background. The data are consistent with the expected background, modeled by a template function. The 95% C.L. exclusion limit on the masses of the gluino and lightest supersymmetric particle in a benchmark simplified model is presented. For a lightest supersymmetric particle mass of 100 GeV, the pair production of gluinos in a multi-bottom final state is excluded for gluino masses up to 1375 GeV.
Introduction: The Hierarchy Problem and Natural SUSY
While the Standard Model of particle physics is incredibly successful, it leaves several questions unanswered, most notably the "hierarchy problem"—the question of why the Higgs boson is so much lighter than the Planck scale. Supersymmetry (SUSY) provides a solution by proposing that every known particle has a heavier "superpartner."
In "natural" SUSY models, the particles responsible for stabilizing the Higgs mass, such as the top and bottom squarks and the gluino, should be relatively light and within reach of the Large Hadron Collider (LHC). This study specifically targets scenarios where gluinos are produced in pairs and decay into final states containing bottom quarks and invisible Lightest Supersymmetric Particles (LSPs), which are candidates for dark matter.
Experimental Strategy: Razor Variables and B-Tagging
To isolate a potential SUSY signal from the overwhelming Standard Model background, the CMS team utilized two primary strategies:
- Razor Variables ($M_R$ and $R^2$): Instead of relying solely on missing transverse energy, these variables characterize the event's mass scale ($M_R$) and the balance of momentum ($R^2$). Standard Model processes typically create a smoothly falling distribution in this 2D plane, while SUSY signals would appear as a localized enhancement or "bump."
- B-Tagging and Categorization: Since natural SUSY models often involve bottom squarks, the researchers required at least one jet to be identified as a bottom quark jet ("b-jet"). Events were further sorted into "boxes" based on jet and b-tag multiplicity to optimize sensitivity to different decay chains.
Rather than relying purely on simulations, the background was modeled using a data-driven approach. A mathematical template function was fitted to the data in "sideband" regions where no signal was expected, ensuring a robust estimation of the Standard Model noise.
Results: Constraints on Gluino Masses
The observed event distributions in the $M_R$ and $R^2$ plane were found to be entirely consistent with the Standard Model background predictions. No significant evidence for new physics was observed. Consequently, the researchers established 95% Confidence Level (C.L.) upper limits on the production cross-section of gluinos.
In the benchmark simplified model where gluinos decay to two bottom quarks and an LSP, gluino masses up to 1375 GeV are excluded for an LSP mass of 100 GeV. This result significantly constrains the available parameter space for natural Supersymmetry.
Conclusion
This search demonstrates the power of razor variables in broad-spectrum searches for new physics. While no discovery was made in the 8 TeV dataset, the results place stringent limits on the mass of the gluino, a key component of natural SUSY. Future analyses using higher-energy data from the LHC will continue to probe these models, either discovering these elusive particles or further refining our understanding of the limits of naturalness in particle physics.
Summary updated