Large-scale Stability and Astronomical Constraints for Coupled Dark-Energy Models
Weiqiang Yang, Supriya Pan, John D. Barrow
TL;DR
This study addresses large-scale perturbation instabilities in interacting dark-energy models by introducing a factor $1+w_x$ into the background energy transfer $Q$, enabling constraints across the full dark-energy equation of state space. The authors formulate a generalized IDE framework and specialize to three concrete models (IDE1–IDE3), deriving the corresponding perturbation equations and a doom-factor stability criterion. Using Planck 2015 CMB, BAO, SNIa (JLA), RSD, weak lensing, cosmic chronometers, and an $H_0$ prior, they perform MCMC analyses and find that current data mildly favor a small nonzero coupling, with $w_x$ near $-1$ and allowing phantom crossing; the standard non-interacting LCDM scenario remains allowed within 68.3% CL. The results indicate that IDEs can ease the $H_0$ tension and remain perturbatively stable, though all three models remain close to LCDM in their background evolution and produce only small deviations in observables. This work thus provides a robust, data-driven pathway to test IDE models across the full $w_x$ landscape, with implications for the nature of dark energy and its possible interaction with dark matter.
Abstract
We study large-scale inhomogeneous perturbations and instabilities of interacting dark energy (IDE) models. Past analysis of large-scale perturbative instabilities, has shown that we can only test IDE models with observational data when its parameter ranges are either $w_{x}\geq -1$ and $ξ\geq 0,$ or $w_{x}\leq -1~$ and $~ξ\leq 0$, where $w_{x}$ is the dark energy equation of state (EoS), and $ξ$ is a coupling parameter governing the strength and direction of the energy transfer. We show that by adding a factor $(1+w_{x})$ to the background energy transfer, the whole parameter space can be tested against all the data and thus, the instabilities in such interaction models can be removed. We test three classes of interaction model using the latest astronomical data from different sources. Precise constraints are found. Our analysis shows that a very small but non-zero deviation from pure $Λ$-cosmology is suggested by the observational data while the no-interaction scenario can be recovered at the 68.3% confidence-level. In particular, for three IDE models, identified as IDE 1, IDE 2, and IDE 3, the 68.3% CL constraints on the interaction coupling strengths are, $ξ= 0.0360_{-0.0360}^{+0.0091}$ (IDE 1), $ξ= 0.0433_{-0.0433}^{+0.0062}$ (IDE 2), $ξ= 0.1064_{-0.1064}^{+0.0437}$ (IDE 3). In addition, we find that the dark energy EoS tends towards the phantom region taking the 68.3% CL constraints, $w_x= -1.0230_{-0.0257}^{+0.0329}$ (IDE 1), $w_x= -1.0247_{-0.0302}^{+0.0289}$ (IDE 2), and $w_x= -1.0275_{-0.0318}^{+0.0228}$ (IDE 3). However, the possibility of $w_{x}>-1$ is also not rejected by the astronomical data used here. Moreover, we find in all IDE models that, as the value of Hubble constant decreases, the behavior of the dark energy EoS shifts from phantom to quintessence type with its EoS very close to that a simple cosmological constant at the present time.