Poster - Constraining Self-Interacting Dark Matter: Insights from Equal Mass Cluster Mergers

Transcript

1. Merging galaxy clusters are thought to be prime testbeds of non-gravitational interactions between dark matter particles. The rapidly growing sample of observed mergers have been used to generate new constraints on the dark matter self-interaction cross section by measuring the offsets between the collisionless galaxy and possibly collisional dark matter distributions. The picture that has emerged from the measure of offsets is unclear. We have run a suite of merger simulations to systematically study the development of offsets in equal mass mergers. gas, dark matter, and galaxies Offsets do not form for most mergers. Offset size depends not just on σSI , but also the number of self-interactions that occur (a function of σSI , cNFW , and impact parameter), the outcome of the self-interactions (a function of merger velocity), and merger phase. We rule out σSI ≥ 10 cm2/g, which fails to produce separated DM halos under typical merger velocities. Galaxy/BCG miscentering on 100s kpc scales in clusters after coalescence may be evidence of self-interactions. constraining self-interacting dark matter insights from equal mass cluster mergers 1st pericenter passage pre-merger after coalescence coalescence galaxies and BCG oscillate around barycenter with significant separations (not observed in CDM) few self-interactions, no offsets before coalescence Dark matter particles ejected following self-interactions form tails, which produce in offsets between the galaxy and dark matter distributions, as measured by their peaks/centroids. If sufficiently many self-interactions occur, then halos coalesce upon contact, before offsets form. DM, 1D galaxies, 1D galaxies, 2D BCG 0 200 -2000 -1000 0 1000 2000 distance from barycenter (kpc) galaxy-DM offset (kpc) modeling the mergers galaxy-DM offset (kpc) distance from barycenter (kpc) SIDM mergers and the development of offsets freefall v0 = 1000 km/s v0 = 2000 km/s halo concentration We simulate our mergers using the N- body code GADGET-2 modified to include self-interactions. We adopted a halo mass of 1015 M¤ (similar to observed mergers El Gordo and CIZA J2242.8+5301), for which central surface densities are comparable to estimates of the upper limit on σSI . SIDM halos were first evolved in isolation to allow cores to develop. Identical halos were then placed so that their virial radii were touching (4 Mpc). simulation inputs DM galaxies self-interactions: constant cross sections σSI = 0 (CDM), 1, 3, 10 cm2/g. Interactions were modeled as isotropic, hard sphere scattering events. calculating offsets Smoothed mass maps projected in 1D and 2D were generated using a kernel density estimate with 50 kpc smoothing. Peaks were identified as the point where the density was the highest. In addition, we traced a BCG particle placed at the center of each halo. In the analyses to the right, only the 1D analysis is shown. initial velocity: freefall, 1000, 2000, 3000 km/s at 4 Mpc separation (1000 km/s cosmologically favored). halo concentration: { 0.5, 1, 2 } cNFW , where cNFW = 3.3, calculated from Duffy et al. 2008.’s mass-concentration relation. Galaxies were given the same concentration. based on an equal mass 1015 M¤ merger with σSI = 3 cm2/g and v0 = 1000 km/s σSI = 10 cm2/g CDM σSI = 1 cm2/g σSI = 3 cm2/g cNFW = 3.3 cNFW = 6.6 cNFW = 1.7 CDM SIDM highly collisional gas lags behind DM and galaxies no DM-galaxy offsets DM forms tails, appears to lag behind galaxies, forming offsets varied cNFW for merger with v0 = 2000 km/s and σSI = 10 cm2/g conclusions self-interactions more expulsive (vesc ~ 2000-3000 km/s), larger offsets form ~10-100 kpc offsets. Observed equal mass mergers such as CIZA J2242.8+5301 have no significant offsets. tail formation pronounced after apocenter error bars calculated via bootstrapping vcol ~ 3800 - 4100 km/s cosmological infall velocities vcol ~ 3900 – 4200 km/s vcol ~ 4300 - 4400 km/s graduate student STACY Y. KIM kim.4905@osu.edu s. y. kim a. h. g. peter