Exponential quintessence with momentum coupling to dark matter

TL;DR

This work investigates a dynamical dark energy model where quintessence with an exponential potential is coupled to dark matter via pure momentum transfer. Using DESI DR2 BAO along with Planck CMB and DES Y5 SN data, the authors show that allowing a momentum coupling permits string-theory motivated values λ ≥ √2, particularly favoring a negative coupling β when λ is fixed, which can suppress late-time growth and alleviate S8-related tensions. They derive 95% upper limits on the sum of neutrino masses, with ∑mν < 0.06 eV when λ is fixed and ∑mν < 0.16 eV when λ varies, highlighting degeneracies between λ, β, and neutrino mass. The results motivate further exploration of interacting dark energy models that respect swampland bounds and remain testable with current and upcoming cosmological data, including improved nonlinear modelling for Stage IV surveys.

Abstract

We present updated constraints on an interacting dark energy - dark matter model with pure momentum transfer, where dark energy is in the form of a quintessence scalar field with an exponential potential. We run a suite of MCMC analyses using the DESI DR2 BAO measurements, in combination with CMB data from Planck and supernovae data from DESY5. In contrast to the standard case of uncoupled quintessence, we find that values for the potential's slope parameter $λ\geq \sqrt{2}$, which are conjectured by string theory scenarios, are not excluded. If $λ$ is fixed to such a value, we find that the data favour the negative coupling branch of the model, which is the branch exhibiting late-time growth suppression. We also derive 95% upper limits on the sum of the neutrino masses, finding $\sum m_ν< 0.06$ eV ($\sum m_ν< 0.16$ eV) when $λ$ is fixed (varied). Our results motivate further studies on dynamical dark energy models that obey string theory bounds and can be constrained with cosmological observations.

Figures (6)

• Figure 1: One dimensional posterior distributions of the parameters $\{\beta,\lambda\}$ together with the contours containing 68% and 95% of the posterior probability for the momentum coupling model (solid green lines) vs uncoupled quintessence (dashed black line), where the latter corresponds to fixing the coupling parameter $\beta=0$. We notice that the coupled case allows $\lambda \geq \sqrt{2}$, which is conjectured by string theory scenarios.
• Figure 2: Dark energy equation of state for the coupled quintessence model with $(\beta,\lambda)=(-0.8,1)$ from \ref{['tab:coupled-uncoupled']}, compared to the CPL parametrisation with $(w_0,w_a)=(-0.75,-0.86)$ from \ref{['eq:DESIbestfit']}.
• Figure 3: One dimensional posterior distributions of the parameters $\{\beta,\lambda\}$, and the derived parameters $\{H_0, \sigma_8\}$, for the momentum coupling model with varying (solid green lines) vs fixed $\lambda=1.5$ (dashed-dotted blue lines), and uncoupled quintessence (dashed black lines). Note that in all cases shown here the total neutrino mass is fixed, $M_\nu=0.06$.
• Figure 4: One dimensional posterior distributions of the parameters $\{\beta, \lambda, H_0,\sigma_8, M_\nu\}$ together with the contours containing 68% and 95% of the posterior probability for the momentum coupling model when allowing the total neutrino mass to vary. We see that allowing $\lambda$ to vary (solid green lines) significantly relaxes the neutrino bounds compared to the case where $\lambda=1.5$ (dashed blue lines).
• Figure 5: One dimensional posterior distributions of the parameters $\{H_0,\Omega_{\rm m}, M_\nu\}$ together with the contours containing 68% and 95% of the posterior probability for the momentum coupling model (solid green lines) and uncoupled quintessence (dashed black lines), as well as the $w_0w_a$ and constant $w$ parametrisations (dotted dashed orange lines and dotted magenta lines, respectively). Neither quintessence models are able to reproduce the positive peak in the $M_\nu$ posterior of the $w_0w_a$ parametrisation, but they allow for much more relaxed $M_\nu$ bounds than $w$CDM. Note that for the MCMC runs shown here we have used the CMB and BAO datasets only (without SN).