Creation of Viscous Dark Energy by the Hubble Flow: Comparison with SNe Ia Master Sample Binned Data

TL;DR

This work extends ΛCDM by incorporating dark energy production from the expanding universe and bulk-viscous dissipation, introducing two extra parameters linked to matter creation and viscosity. The authors impose a consistency constraint by matching the present-day deceleration parameter to its ΛCDM value and define an effective running Hubble constant $\\mathcal{H}_{0}(z)$ to compare predictions with the Master SNe Ia binned data via MCMC across 20 redshift bins. They find that the intrinsic equation of state of the created dark energy is phantom ($w<-1$), while the case with bulk viscosity alone yields a quintessence-like intrinsic EoS but phantom behavior in $w_{eff}(z)$; overall, no strong statistical preference emerges among the dynamical models, though a power-law running Hubble model is favored by BIC. The results underscore the utility of the running Hubble constant as a diagnostic for late-time dark energy evolution and suggest phantom-like dynamics may be preferred when confronted with binned SNe Ia data, in line with some DESI indications.

Abstract

We study a cosmological model featuring evolutionary dark energy, according to the idea that the creation of its constituents arises from the gravitational field of the expanding universe, whose non-equilibrium physics is described by a non-zero bulk viscosity coefficient. This physical scenario calls for two additional parameters with respect to the ΛCDM model, one of which is the equation of state parameter of the created dark energy. The model is constrained by the requirement that its deceleration parameter coincides with the one predicted by the ΛCDM model. Then, we construct the effective running Hubble constant, a theoretical function that corresponds to the ratio of the Hubble parameter in our model to the ΛCDM expansion rate. The model's theoretical predictions for the effective running Hubble constant are compared with the binned data of the Supernovae Ia Master Sample. The comparison is performed by a MCMC procedure for each bin, with three parameters left free to vary, while the particle creation rate is taken from a grid of values, each of which is fixed in the given MCMC run. The most important result emerging from this analysis is that the created dark energy constituent corresponds to an equation of state parameter with phantom character. Only if particle creation is removed do the dark energy constituents acquire a quintessence character. No matter the intrinsic nature of the constituents, their effective z-dependent equation of state parameter is, both with and without considering particle creation, entirely of phantom nature across the considered redshift range.

Figures (5)

• Figure 1: Fit results for the parameter $w$ in the dynamical dark energy model M$[\Gamma,\xi]$ tested with the Master Sample, while varying $H_\Gamma$ over a grid of fixed values. In red, the value of $H_\Gamma$ selected by identifying the minimum of the $\chi^2$ among the different fits.
• Figure 2: Check on the behaviour of the term $-H_\xi E$ appearing in Eq. \ref{['eq:omegap']}. In the studied redshift range, corresponding to the redshift range covered by the Master Sample data, this term stays smaller than unity and the perturbative character of the bulk viscosity is not violated.
• Figure 3: Fit results for the dynamical dark energy models M$[\Gamma,\xi]$, M$[\Gamma]$, and M$[\xi]$ , following Eq. \ref{['ref:eq:Hrunning']}. The parameters used for the plot are listed in Tab. \ref{['tab:results']}. The best fit for power-law model PL has also been included for comparison.
• Figure 4: Effective parameter of state for the different theoretical models, using the fitted parameters in Tab. \ref{['tab:results']} to plot the curves in Eq. \ref{['eq:weff']}, Eq. \ref{['eq:weffmatter']}, Eq. \ref{['eq:weffbulk']}.
• Figure 5: Posterior distribution constructed via the MCMC procedure, for the fitted parameters of the dynamical dark energy model M$[\Gamma,\xi]$. Mean values and standard deviations are displayed in Tab. \ref{['tab:results']}. The dark blue areas correspond to a 68% confidence interval, the light blue areas to a 95% confidence interval.