Arneson, Ryan2018-03-142018-03-142017-12https://hdl.handle.net/11299/194633University of Minnesota Ph.D. dissertation. December 2017. Major: Astrophysics. Advisor: Robert Gehrz. 1 computer file (PDF); viii, 93 pages.Luminous blue variables (LBVs) represent a post-main sequence phase in which massive stars (Mi ≥ 20 M , Langer et al., 1994) lose a considerable amount of mass via giant eruptions and minor outbursts creating circumstellar nebula. RV Tauri and yellow semi-regular (SRd) variables are two classes of post-main sequence stars with a typical mass of ~0.7 M (Tuchman et al., 1993) that are thought to be the immediate precursors of planetary nebulae (PNe) and have been termed "proto-planetary nebulae" (PPNe). These classes of stars are responsible for creating unique circumstellar environments where various dust species grow, undergo processing and dissipate into the interstellar medium (ISM). This thesis studies the dust mineralogy and mass-loss history of both high mass (LBVs) and low mass (RV Tauri and SRd variables) post-main sequence stars using SOFIA/FORCAST 5 - 40 µm mid-infrared imaging and spectroscopy. In particular, I present SOFIA/FORCAST 5-40 µm mid-infrared imaging of the dust surrounding the LBV candidates MN 90 and HD 168625 to quantify the mineral abundances of the circumstellar dust and constrain the evolutionary state of these objects. The images between 7.7-37.1 µm of HD 168625 compliment previously obtained mid-IR imaging. The dust color temperature and optical depth maps of HD 168625 show evidence for the limb-brightened peaks of an equatorial torus. Our image at 37.1 µm of MN 90 shows a limb-brightened, spherical dust shell. A least-squares t to the spectral energy distribution of MN 90 yields a dust temperature of 69 ± 3 K, with the peak of the emission at 42 µm. Similarly for HD 168625, I estimate a dust temperature of 160 ± 7 K, with the peak of the emission at 18 µm. Radiative transfer calculations of HD 168625 using 2-Dust estimate that mass-loss occurred 2.6 x 10^3 years ago at a rate of 1.3x10^-7 M yr^-1 to create a dust torus/shell with a dust mass of 2.4x10^3 M. Similarly, I estimate for MN 90 that mass-loss occurred 6x10^3 years ago at a rate of 8.3x10^-6 M yr^-1 to create a dust shell with a dust mass of 2.5x10^-2 M . The 2-Dust calculations suggest the presence of oxygen rich silicates around both stars. I further present a SOFIA/FORCAST 5 - 40 µm mid-infrared grism spectroscopic survey that examines the mineralogy of the circumstellar dust in a sample of RV Tauri and SRd variable stars. A mineralogical model for each star indicates the presence of both carbon rich and oxygen rich dust species-contrary to simple dredge-up models-with a majority of the dust in the form of amorphous carbon and graphite. The oxygen rich dust is primarily in the form of amorphous silicates. For most of the systems, the analysis suggests that the grains are relatively large and have undergone significant processing, supporting the hypothesis that the dust is conned to a Keplerian disk and that the heavily processed, central regions of the disk are viewed from a nearly face-on orientation. These results help to determine the physical properties of post-main sequence circumstellar environments and to constrain models of post-main sequence mass loss and planetary nebula formation.enThesis or Dissertation