Theories for charge-driven nematicity in kagome metals

Starting from a low-energy continuum model for the band dispersion of the $2 \times 2$ charge-ordered phase of the kagome metals $A$V$_3$Sb$_5$ ($A=$ K, Rb, Cs), we show that nematicity can develop in this state driven either by three inequivalent $1 \times 4$ charge fluctuations preemptive of a $1 \times 4$ charge order (CO), or by an actual zero momentum $d$-wave charge Pomeranchuk instability (PI). We perform a Kohn-Luttinger analysis in the particle-hole sector, which allows us to establish a criterion for the development of an attractive nematic channel near the onset of the $1 \times 4$ CO and near the $d$-wave charge PI, respectively. We derive an effective charge-fermion model for the $d$-wave PI with a nematic susceptibility given via a random phase approximation (RPA) summation. By contrast, for the finite momentum CO, the RPA scheme breaks down and needs to be improved upon by including Aslamazov-Larkin contributions to the nematic pairing vertex. We then move to the derivation of the Ginzburg-Landau potentials for the $1 \times 4$ CO and for the $d$-wave PI, and we obtain the corresponding analytical expression for the nematic susceptibility at the nematic transition temperature T $ \sim \text{T}_\text{nem}$ in both cases. Our work establishes a relation between the nematicity observed in some of the iron-based superconductors, where the nematic phase might be driven by spin fluctuations, and the vanadium-based kagome metals, where charge fluctuations likely induce nematicity. The two microscopic mechanisms we propose for the stabilization of the nematic state in $A$V$_3$Sb$_5$ are distinguishable by diffusive scattering experiments, meaning that it is possible to gauge which of the two theories, if any, is the most likely to describe this phase. Both mechanisms might also be relevant for the recently discovered titanium-based family $A$Ti$_3$Sb$_5$.