Infrared (IR) excesses around K-type red giants (RGs) have previously been discovered using IRAS data, and past studies have suggested a link between RGs with overabundant Li and IR excesses, implying the ejection of circumstellar shells or disks. We
revisit the question of IR excesses around RGs using higher spatial resolution IR data, primarily from WISE. Our goal was to elucidate the link between three unusual RG properties: fast rotation, enriched Li, and IR excess. We have 316 targets thought to be K giants, about 40% of which we take to be Li-rich. In 24 cases with previous detections of IR excess at low spatial resolution, we believe that source confusion is playing a role, in that either (a) the source that is bright in the optical is not responsible for the IR flux, or (b) there is more than one source responsible for the IR flux as measured in IRAS. We looked for IR excesses in the remaining sources, identifying 28 that have significant IR excesses by ~20 um (with possible excesses for 2 additional sources). There appears to be an intriguing correlation in that the largest IR excesses are all in Li-rich K giants, though very few Li-rich K giants have IR excesses (large or small). These largest IR excesses also tend to be found in the fastest rotators. There is no correlation of IR excess with the carbon isotopic ratio, 12C/13C. IR excesses by 20 um, though relatively rare, are at least twice as common among our sample of Li-rich K giants. If dust shell production is a common by-product of Li enrichment mechanisms, these observations suggest that the IR excess stage is very short-lived, which is supported by theoretical calculations. Conversely, the Li-enrichment mechanism may only occasionally produce dust, and an additional parameter (e.g., rotation) may control whether or not a shell is ejected.
A Li-rich red giant star (2M19411367+4003382) recently discovered in the direction of NGC 6819 belongs to the rare subset of Li-rich stars that have not yet evolved to the luminosity bump, an evolutionary stage where models predict Li can be replenis
hed. The currently favored model to explain Li enhancement in first-ascent red giants like 2M19411367+4003382 requires deep mixing into the stellar interior. Testing this model requires a measurement of 12C/13C, which is possible to obtain from APOGEE spectra. However, the Li-rich star also has abnormal asteroseismic properties that call into question its membership in the cluster, even though its radial velocity and location on color-magnitude diagrams are consistent with membership. To address these puzzles, we have measured a wide array of abundances in the Li-rich star and three comparison stars using spectra taken as part of the APOGEE survey to determine the degree of stellar mixing, address the question of membership, and measure the surface gravity. We confirm that the Li-rich star is a red giant with the same overall chemistry as the other cluster giants. However, its log g is significantly lower, consistent with the asteroseismology results and suggestive of a very low mass if the star is indeed a cluster member. Regardless of the cluster membership, the 12C/13C and C/N ratios of the Li-rich star are consistent with standard first dredge-up, indicating that Li dilution has already occurred, and inconsistent with internal Li enrichment scenarios that require deep mixing.
