We make use of Spitzer imaging between 4 and 16 μm and near-infrared data at 2.2 μm to investigate the nature and distribution of the mid-infrared emission in a sample of early-type galaxies (ETGs) in the Virgo cluster. These data allow us to conclude, with some confidence, that the emission at 16 μm in passive ETGs is stellar in origin, consistent with previous work concluding that the excess mid-infrared emission comes from the dusty envelopes around evolved asymptotic giant branch (AGB) stars. There is little evidence for the mid-infrared emission of an unresolved central component, as might arise in the presence of a dusty torus associated with a low-luminosity active galactic nucleus.
We none the less find that the 16-μm emission is more centrally peaked than the near-infrared emission, implying a radial stellar population gradient. By comparing with independent evidence from studies at optical wavelengths, we conclude that a metallicity that falls with increasing radius is the principal driver of the observed gradient.
We also plot the mid-infrared colour–magnitude diagram and combine with similar work on the Coma cluster to define the colour–magnitude relation for absolute K-band magnitudes from −26 to −19. Because a correlation between mass and age would produce a relation with a gradient in the opposite sense to that observed, we conclude that the relation reflects the fact that passive ETGs of lower mass also have a lower average metallicity. The colour–magnitude relation is thus driven by metallicity effects.
In contrast to what is found in Coma, we do not find any objects with anomalously bright 16-μm emission relative to the colour–magnitude relation. Although there is little overlap in the mass ranges probed in the two clusters, this may suggest that observable ‘rejuvenation’ episodes are limited to intermediate-mass objects.