QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: ams_nanolab on September 1, 2016, 12:59

Title: Cutoff energy and suprecell size
Post by: ams_nanolab on September 1, 2016, 12:59

We know a bigger suprecell has higher total energy but from an simulation accuracy point of view how to estimate the correct cutoff energy to simulate large suprecells? What is the correlation between the cut-off energy used in calculations and the size of the suprecell.

Title: Re: Cutoff energy and suprecell size
Post by: Ulrik G. Vej-Hansen on September 1, 2016, 13:57

Are you referring to the density mesh cutoff? If so, then there is no correlation. The density mesh cutoff describes the density of the grid, which should be independent of the cell size. Of course, if you want to converge your total energy in absolute terms, then you might need to use a denser grid for a larger supercell, but that is only relevant for some specific classes of problems.

Title: Re: Cutoff energy and suprecell size
Post by: ams_nanolab on September 1, 2016, 14:22

Yes it was the mesch cutoff energy, thanks.

Title: Re: Cutoff energy and suprecell size
Post by: Jess Wellendorff on September 2, 2016, 07:28

So, in short: You should use the same density mesh cutoff energy for both small and large supercells.