Cosmological structure formation in complex scalar field dark matter versus real ultralight axions: a comparative study using CLASS

TL;DR

This study compares complex scalar field dark matter and real ultralight axions in cosmology, analyzing their effects on structure formation, CMB, and matter power spectra using a modified CLASS code, revealing key differences and constraints.

Contribution

It introduces a modified CLASS code to incorporate complex SFDM physics, including the early stiff phase, and compares its cosmological predictions with ULAs and FDM models.

Findings

SFDM with large cores is disfavored by small-scale structure constraints.

Differences in power spectra are due to the phase kinetic energy and Jeans mass evolution.

Complex SFDM and ULAs show distinct features in structure formation and power spectra.

Abstract

(abridged) We continue the study of $\Lambda$SFDM cosmologies, which differ from $\Lambda$CDM in that CDM is replaced by scalar field dark matter (SFDM) by calculating the evolution of the background Universe, as well as linear perturbations, focusing on scalar modes. We consider models with complex scalar field with a repulsive, quartic self-interaction (SI), and models without SI, referred to as fuzzy dark matter (FDM). To this end, we modify the Boltzmann code CLASS, to incorporate the physics of complex SFDM which has as one of its characteristics that its equation of state is maximally stiff in the very early Universe, dominating then over all the other cosmic components, even over radiation. We calculate CMB and matter power spectra as well as unconditional Press-Schechter halo mass functions for various models, expanding previous literature that were limited either to the background, or to a semi-analytical approach to SFDM density perturbations neglecting the early stiff phase. Comparing our results of each, SFDM and FDM, with real-field ultralight axion models (ULAs) without SI, we characterize the differences between the respective background evolution and linear structure growth. Our calculations confirm previous results of recent literature, implying that SFDM models with $\gtrsim$ kpc-size halo cores are disfavored, questioning their ability to explain the small-scale problems on dwarf-galactic scales. Also we find that the kinetic energy due to the phase of the complex field leads to marked differences between SFDM/FDM versus ULAs. The mild falloff in the SFDM power spectrum toward high k is similar to that of CDM but based on different effects, namely the rapidly shrinking Jeans mass for SFDM as opposed to the Meszaros effect for CDM. In addition, we find that the sharp cutoff in the ULA power spectrum is also followed by a mild falloff, albeit at very small power.