Following our previous work , we investigate through hydrodynamic simulations the destruction of newly-formed dust grains by sputtering in the reverse shocks of supernova remnants .
Using an idealized setup of a planar shock impacting a dense , spherical clump , we implant a population of Lagrangian particles into the clump to represent a distribution of dust grains in size and composition .
We vary the relative velocity between the reverse shock and ejecta clump to explore the effects of shock-heating and cloud compression .
Because supernova ejecta will be metal-enriched , we consider gas metallicities from Z / Z _ { \mathrm { \odot } } = 1 to 100 and their influence on cooling properties of the cloud and the thermal sputtering rates of embedded dust grains .
We post-process the simulation output to calculate grain sputtering for a variety of species and size distributions .
In the metallicity regime considered in this paper , the balance between increased radiative cooling and increased grain erosion depends on the impact velocity of the reverse shock .
For slow shocks ( v _ { \mathrm { shock } } \leq 3000 km s ^ { -1 } ) , the amount of dust destruction is comparable across metallicities , or in some cases is decreased with increased metallicity .
For higher shock velocities ( v _ { \mathrm { shock } } \geq 5000 km s ^ { -1 } ) , an increase in metallicity from Z / Z _ { \mathrm { \odot } } = 10 to 100 can lead to an additional 24 % destruction of the initial dust mass .
While the total dust destruction varies widely across grain species and simulation parameters , our most extreme cases result in complete destruction for some grain species and only 44 % dust mass survival for the most robust species .
These survival rates are important in understanding how early supernovae contribute to the observed dust masses in high-redshift galaxies .