Core-collapse supernova enrichment in the core of the Virgo Cluster

Using a deep (574 ks) Chandra observation of M87, the dominant galaxy of the nearby Virgo Cluster, we present the best measurements to date of the radial distribution of metals in the central intracluster medium (ICM). Our measurements, made in 36 independent annuli with $\sim$250,000 counts each, extend out to a radius r$\sim$40 kpc and show that the abundance profiles of Fe, Si, S, Ar, Ca, Ne, Mg, and Ni are all centrally peaked. Interestingly, the abundance profiles of Si and S - which are measured robustly and to high precision - are even more centrally peaked than Fe, while the Si/S ratio is relatively flat. These measurements challenge the standard picture of chemical enrichment in galaxy clusters, wherein type Ia supernovae (SN Ia) from an evolved stellar population are thought to dominate the central enrichment. The observed abundance patterns are most likely due to one or more of the following processes: continuing enrichment by winds of a stellar population pre-enriched by SNCC products; intermittent formation of massive stars in the central cooling core; early enrichment of the low entropy gas. We also discuss other processes that might have contributed to the observed radial profiles, such as a stellar initial mass function that changes with radius; changes in the pre-enrichment of core-collapse supernova progenitors; and a diversity in the elemental yields of SN Ia. Although systematic uncertainties prevent us from measuring the O abundance robustly, indications are that it is about 2 times lower than predicted by the enrichment models.

💡 Research Summary

This paper presents a detailed study of the metal distribution in the central intracluster medium (ICM) of the Virgo Cluster, using an exceptionally deep (574 ks) Chandra observation of its dominant galaxy, M87. By dividing the inner 40 kpc of the ICM into 36 concentric annuli (each containing roughly 250,000 net counts), the authors obtain high‑precision abundance measurements for a suite of elements: Fe, Si, S, Ar, Ca, Ne, Mg, and Ni. The data reduction follows the procedures described in earlier papers of the series, employing CIAO 4.3 with CALDB 4.4.1, careful background subtraction using blank‑sky fields, and the exclusion of the bright X‑ray arms and the innermost multiphase core to minimise multi‑temperature biases. Spectra are fitted in the 0.6–7.0 keV band with both the APEC (AtomDB v2.0.1) and MEKAL plasma codes, using a single‑temperature absorbed thermal model (NH = 1.93 × 10²⁰ cm⁻²). Free parameters include temperature, normalization, and the abundances of O, Ne, Mg, Si, S, Ar, Ca, Fe, and Ni; the fits are performed with the extended C‑statistic and uncertainties are derived from Markov Chain Monte Carlo (MCMC) posterior distributions.

The resulting radial abundance profiles show that all measured elements are centrally peaked. Fe reaches > 1.2 Solar in the core, while Si and S peak at even higher values (~1.5 Solar). Beyond ~25 kpc the abundances decline steadily to ~0.6 Solar. Notably, the Si/Fe and S/Fe ratios are not flat: they peak at ~1.3 Solar near the nucleus and decrease linearly to ~1.0 Solar at 35 kpc (gradient ≈ ‑7 × 10⁻³ Solar kpc⁻¹), a trend that is statistically significant at the 5–11 σ level. In contrast, the Si/S ratio remains essentially constant (1.00 ± 0.01 Solar) across the whole radial range, indicating that the relative production of Si and S does not vary with radius. The Ne/Fe and Ni/Fe ratios exhibit larger systematic differences between the two plasma codes (e.g., Ne/Fe ≈ 2 Solar with APEC versus 0.8 Solar with MEKAL), but both show a decreasing trend outward. Oxygen could not be robustly measured because of ACIS low‑energy calibration issues and background uncertainties; the authors note that the O abundance appears roughly a factor of two lower than predicted by standard enrichment models.

Systematic uncertainties are carefully examined. Updating the atomic database from AtomDB v1.3.2 to v2.0.1 reduces the Fe abundance by ~20 % while affecting other elements by ≤ 10 %. Multi‑temperature simulations indicate that residual temperature structure could bias Si/Fe, S/Fe, and Si/S ratios by less than 10 %. The authors also discuss potential biases from the effective area calibration around the Si edge, but conclude that these are minor compared with the statistical precision achieved.

The observed central enrichment pattern challenges the conventional picture in which Type Ia supernovae (SN Ia) from an old stellar population dominate the metal budget in cool‑core clusters, while core‑collapse supernovae (SN CC) products (O, Ne, Mg) are uniformly mixed. The fact that Si and S—elements produced in comparable amounts by both SN Ia and SN CC—are more centrally concentrated than Fe, and that Si/Fe and S/Fe decline outward, suggests a substantial contribution from SN CC or SN Ia with atypical yields in the core. The authors propose several plausible mechanisms: (1) continuous enrichment by winds from a stellar population that was pre‑enriched by SN CC products; (2) intermittent formation of massive stars within the cooling core, leading to recent SN CC events; (3) early enrichment of low‑entropy gas that later sank to the cluster centre. Additional possibilities include a radially varying initial mass function (IMF), changes in the pre‑enrichment of SN CC progenitors, or a diversity of SN Ia explosion channels that alter the Si/Fe yield ratio.

In summary, the deep Chandra data reveal that the metal distribution in M87’s core cannot be explained by a simple SN Ia‑dominated model. Instead, a more complex enrichment history involving both SN Ia and SN CC, possibly modulated by ongoing star formation, stellar winds, or early‑epoch processes, is required. The authors emphasize that future high‑resolution X‑ray spectroscopy (e.g., XRISM, Athena) will be essential to measure low‑energy lines (especially O) with sufficient accuracy, thereby refining constraints on the relative contributions of different supernova types and on the physical processes shaping the chemical evolution of galaxy‑cluster cores.