$b \to c \ell \nu$ anomalies in light of extended scalar sectors

TL;DR

This paper investigates anomalies in semileptonic B meson decays indicating potential new physics, analyzing scalar sector extensions and their compatibility with experimental constraints, and assessing specific models like Georgi-Machacek and Leptoquark scenarios.

Contribution

It explores the impact of scalar current operators on B decay anomalies, constrains models with experimental data, and evaluates the viability of Georgi-Machacek and Leptoquark models.

Findings

Constraints on scalar interactions are consistent with flavor data if BaBar results are excluded.

Including BaBar results disfavors the Georgi-Machacek model by over 3σ.

Leptoquark scenario remains viable within 68% confidence level with limited parameter space.

Abstract

Considering the recent experimental results on exclusive semileptonic $B$ meson decays showing sizable departure from their Standard Model prediction of lepton flavor universality and keeping ongoing and proposed non-standard Higgs searches in mind, we explore the charged current flavor observables ($\mathcal{R}_{D^{(*)}}$, $\mathcal{R}_{J/\psi}$), among other $b\to c\ell \nu$ transitions, in the presence of a relevant scalar current effective new physics operator. We use $B_c$ lifetime and predicted bounds on the branching fraction of $B_c \to \tau \nu$ decay as constraints. We show the allowed parameter space in terms of the real and imaginary parts of the corresponding Wilson coefficients for such interactions. Under the light of obtained results, we study the prospect of two benchmark models, rendering the Wilson coefficients real (Georgi-Machacek (GM)) and complex (Leptoquark (LQ)) respectively. We show that constraints from $b\to c\ell \nu$ on GM parameters are consistent with other flavor constraints on the model, if we drop the \Babar~results. Including those disfavors the model by more than $3\sigma$. On the other hand, one benchmark LQ scenario, which gives rise to a single scalar current effective interaction, is still allowed within $68\%$ confidence level, albeit with a shrunk parameter space.