Institute for Advanced Study Astrophysics Seminar

Abstract: Numerical simulations of galaxies and the interstellar medium are now reaching resolutions where small-scale physical processes can be resolved directly. This raises a natural question: which pieces of microphysics must be modeled explicitly in order to capture the behavior of astrophysical systems? In this talk I will argue that dust provides a particularly natural place to begin. Dust grains interact simultaneously with radiation, thermal balance, chemical networks, and gas dynamics, processes that are critical to the evolution of astrophysical environments. Yet in most simulations dust is assumed to remain perfectly coupled to the gas, reducing it to a passive source of opacity and erasing its dynamical role.

Using a combination of simulations and analytical arguments, I will present work that relaxes this assumption and follows dust as a dynamically distinct component. I consider several environments, including AGN-driven winds, star-forming molecular clouds, and magnetized dusty plasmas. In each case, allowing dust to move relative to the gas produces qualitative changes that are absent when dust and gas are assumed to move together. These include modifications to the structure of radiation-driven outflows, systematic effects on the composition and efficiency of star formation, and new pathways for seeding magnetic fields in the early Universe.

Taken together, these results illustrate how grain-scale dust physics propagates upward to shape the large-scale structure and evolution of astrophysical systems.