Complete Two-loop Dominant Corrections to the Mass of the Lightest CP-even Higgs Boson in the Minimal Supersymmetric Standard Model
Jose Ramon Espinosa, Ren-Jie Zhang
TL;DR
This paper delivers a complete two-loop calculation of the dominant MSSM corrections to the lightest CP-even Higgs mass, $M_{h^0}$, at ${\cal O}(\alpha_t^2)$ using an effective potential framework with general stop mixing and $\tan\beta$. It derives both full analytic expressions and compact RG-improved forms, showing that large logarithms can be absorbed into a one-loop term evaluated at appropriate scales plus smaller two-loop threshold contributions. Numerically, for a SUSY scale around 1 TeV and $M_t=175\pm5$ GeV, the upper bound on $M_{h^0}$ is about $129\pm5$ GeV, with two-loop $\alpha_t^2$ effects capable of increasing the mass by up to ~5 GeV in the maximal mixing scenario. The results are validated against existing diagrammatic calculations, and the analytic approximations achieve sub-GeV precision across most of the parameter space, providing practical tools for precision MSSM Higgs phenomenology and Higgs searches.
Abstract
Using an effective potential approach, we compute two-loop radiative corrections to the MSSM lightest ${\cal CP}$-even Higgs boson mass $M_{h^0}$ to ${\cal O}(α_t^2)$ for arbitrary left-right top-squark mixing and $\tanβ$. We find that these corrections can increase $M_{h^0}$ by as much as 5 GeV; assuming a SUSY scale of 1 TeV, the upper bound on the Higgs boson mass is $M_{h^0}\approx 129\pm 5$ GeV for the top quark pole mass $175\pm 5$ GeV. We also derive an analytical approximation formula for $M_{h^0}$ which is good to a precision of $\lsim 0.5$ GeV for most of the parameter space and suitable to be further improved by including renormalization group resummation of leading and next-to-leading order logarithmic terms. Our final compact formula admits a clear physical interpretation: radiative corrections up to the two-loop level can be well approximated by a one-loop expression with parameters evaluated at the appropriate scales, plus a smaller finite two-loop threshold correction term.
