Dear Ray,
There are several things that you can improve in the input.
The SVZ basis is rather poor. The cutoff 300 is low.
You are using PADE functional (why?) with PBE pseudo potentials.
The mixing factor for the diagonalizatin scheme is too large.
The diagonalization with mixing of the density matrix might be not the best choice for your system.
Regards
Marcella