What Drives Cluster Cool-Core Transformations? A Population Level Analysis of TNG-Cluster

TL;DR

This work analyzes 352 massive galaxy clusters from the TNG-Cluster simulation to quantify how often and why cluster cores transform between cool-core (CC) and non-cool-core (NCC) states, and whether mergers or AGN feedback drive these transitions. It introduces an automatic, entropy-derivative-based method to identify core transformations from the central entropy history, and correlates these events with merger histories and SMBH-driven feedback episodes. The study finds that CC→NCC transformations are common (about 478 events) and typically occur on timescales of roughly 1–2 Gyr, with many transformations linked to recent mergers, especially major ones, though a substantial fraction occur without a direct merger trigger. It also shows that prolonged, high-duty-cycle AGN activity can heat and disrupt cores, suggesting AGN feedback—potentially triggered by mergers—plays a major role in driving NCC states, while mergers alone can displace core gas. Overall, cluster cores exhibit diverse thermodynamic histories, with six archetypes illustrating the spectrum from stable CC to persistent NCC states and temporary phases, highlighting the intricate, coupled evolution of ICM thermodynamics, mergers, and AGN feedback in shaping cluster cores.

Abstract

In this study, we examine the frequency and physical drivers of transformations from cool-core (CC) to non-cool-core (NCC) clusters, and vice versa, in a sample of 352 massive galaxy clusters (M_vir = 10^14-15.3 M_sun) from the TNG-Cluster magnetohydrodynamical cosmological simulation of galaxies. By identifying transformations based on the evolution of central entropy and focusing on z<2.5, we find that clusters frequently undergo such events, depending on their assembly and supermassive black hole histories. On average, clusters experience 2 to 3 transformations. Transformations can occur in both directions and can be temporary, but those to higher entropy cores, i.e. in the direction from CC to NCC states, are the vast majority. CC phases are shorter than NCC phases, and thus overall the TNG-Cluster population forms with low-entropy cores and moves towards NCC states with time. We study the role that mergers play in driving transformations, and find that mergers within ~1Gyr prior to a transformation toward higher (but not lower) entropy cores occur statistically more often than in a random control sample. Most importantly, we find examples of mergers associated with CC disruption regardless of their mass ratio or angular momentum. However, past merger activity is not a good predictor for z=0 CC status, at least based on core entropy, even though clusters undergoing more mergers eventually have the highest core entropy values at z=0. We consider the interplay between AGN feedback and evolving cluster core thermodynamics. We find that core transformations are accompanied by an increase in AGN activity, whereby frequent and repeated (kinetic) energy injections from the central SMBHs can produce a collective, long-term impact on central entropy, ultimately heating cluster cores. Whether such fast-paced periods of AGN activity are triggered by mergers is plausible, but not necessary.

Figures (24)

• Figure 1: Representative time evolution of the central cooling state of a massive galaxy cluster. In particular, the central entropy of the most massive $z=0$ halo in TNG-Cluster is given by the circular symbols (color indicating relaxedness, in terms of the ratio of kinetic to thermal energy within $r_{\rm 500c}$). Correspondingly, the light gray line shows the smoothed time evolution of $K_0$, while the dark gray curve shows its time derivative. Local extrema of the derivative are indicated by crosses, with larger orange crosses marking significant core transformation events (see text). The blue markers indicate the timescales or duration of the two core transformations (see text for a definition). This halo undergoes a core transformation between $z \sim 0.3-0.5$ wherein the core evolves towards a more cool-core (CC) state. In this case the transformation is temporary and lasts $\sim 2-3$ Gyr, after which the central entropy returns to its previous level. This type of behavior is one of several archetypes (see text for discussion). The pink stars mark the start of the three most massive merger events in the history of this cluster. The size of the markers indicate the mass ratio of the merging event. We mark the formation time of the cluster with the silver vertical, dashed line. Note that throughout the paper, time evolves from right to left in all plots that depict a time evolution.
• Figure 2: Correlation among halo mass ${\rm M}_{\rm 500c}$ at $z=0$, formation redshift and cool-coreness in TNG-Cluster, across cosmic epochs. The formation redshift throughout this paper is defined as the redshift at which a cluster reaches a mass of $10^{14} \rm{M}_\odot$. The upper panels show the correlation of formation redshift and the halo mass at $z=0$. Symbols are color-coded by the central entropy of each halo at $z=0$ (left plot) and at the formation redshift $z_{\rm form}$ (right panel), with the color bar indicating the classification into non-cool-core (NCC), weak cool-core (WCC), and strong cool-core (SCC) clusters. Histograms of the distribution of formation redshifts and cluster masses are shown on the top and right panels, respectively. The lower plots present entropy profiles at $z=0$ (left) and at $z_{\rm form}$ (right) colored by core status. More massive halos form, i.e. reach the status of cluster, on average earlier than less massive halos. For a given halo mass, later forming halos are biased towards NCC states at $z=0$. In addition, clusters form with low central entropy, with SCCs and WCCs having similarly shaped profiles exhibiting a prominent decline towards the core. According to TNG-Cluster, clusters are predominately born as CCs.
• Figure 3: Time evolution of the central entropy for all halos of TNG-Cluster after the formation of each cluster. Each line represents one of the 352 clusters as it evolves and grows in mass through time. Lines are colored by the core status at each redshift. The overall cluster population is dominated by SCCs and WCCs at high redshift and gains progressively more NCC towards low redshift. Amid the overall evolution towards the NCC regime, there is a huge diversity in the central entropy evolution from cluster to cluster.
• Figure 4: Rate of core changes as a function of lookback time for three different mass bins according to TNG-Cluster. We define the rate as the number of transformations in a mass and time bin divided by the length of the time interval and number of halos in that bin (details in the text). Orange (blue) lines indicate the rate of changes to higher (lower) core entropy. The bands for each line mark the methodological uncertainty, that is, the variation of the result we obtain when redoing the analysis with a larger and lower threshold of the entropy time derivative. The solid lines show the result for our fiducial choice of threshold. We find no trend with mass and only very weak redshift trends for both type of transformations, but the rate for changes to lower core entropy is overall smaller.
• Figure 5: Correlation between the lookback time of a transformation and cluster mass at $z=0$ according to TNG-Cluster. Transformations to higher core entropy are shown in orange, while transformations to lower core entropy are shown in dark blue. Marginalized histograms of each axis quantity are shown with the top and right subpanels, respectively. The gray dashed line indicates the earliest formation times for our halos, i.e. the earliest possible transformation times. The solid grey line depicts the average formation times (i.e. the fit to the results presented in Fig. \ref{['fig:M500VsZform']}). The gold (light blue) lines and bands show the median and 16 to 84 percentile ranges for changes to higher (lower) core entropy. The average lookback time of transformations increases with halo mass, in part because more massive clusters form earlier and may therefore experience more and earlier transformations. Changes to higher core entropy happen at all times, while changes to lower core entropy occur only at later times, potentially reflecting different physical drivers.