SIMBAD references

We use Milky Way-like chemodynamical simulations with a new treatment for dust destruction and growth to investigate how these two processes affect the properties of the interstellar medium in galaxies. We focus on the role of two specific parameters, namely f_des (a new parameter that determines the fraction of dust destroyed in a single gas particle vicinity of a supernova) and C_s (the probability that a metal atom or ion sticks to the dust grain after colliding, i.e. the sticking coefficient), in regulating the amount and distribution of dust, cold gas and metals in galaxies. We find that simulated galaxies with low f_des and/or high C_s values not only produce more dust, but they also have a shallower correlation between the dust surface density and the total gas surface density, and a steeper correlation between the dust-to-gas ratio and the metallicity. Only for values of f_des between 0.01 and 0.02, and of C_s between 0.5 and 1 do our simulations produce an average slope of the dust-to-gas ratio versus metallicity relationship that is consistent with observations. f_des values correspond to a total fraction of dust destroyed by a single supernova ranging between 0.42 and 0.44. Finally, we compare predictions of several simulations (with different star formation recipes, gas fractions, central metallicities, and metallicity gradients) with the spatially resolved M101 galaxy, and conclude that metallicity is the primary driver of the spatial distribution of dust, while the dust-to-gas ratio controls the cold gas distribution, as it regulates the atomc-to-molecular hydrogen conversion rate.