## Determination of parameters for dielectric tensor calculation document.SUBSCRIPTION_OPTIONS = { "thing": "thread", "subscribed": false, "url": "subscribe", "icon": { "css": "fa fa-envelope-o" } };

Elk Users
Ao Teng
2013-07-24
2013-07-25
• Ao Teng - 2013-07-24

Hello,

I'm a little bit confused by some parameters in the dielectric response calculation.Could anybody give me some hint how the following parameters are determined for metal:

1.scissor
2.dos
3.vkloff

Thank you!

Regards

Ao

• Markus - 2013-07-25

Hi Ao,

for a metal, scissor has no meaning: it shifts the unoccupied states up (or down) if you have a gap. This is to correct the so-called "band gap error".

dos (or wplot as it is called now) determines how many energy points you want in your plot and which range you want to have. See the manual for details. Just generate a plot and see if it meets your needs in terms of resolution and range. Remember that energies are in Hartree.

vkloff may be useful in order to save some calculation time by breaking the symmetry of the k-point mesh. This may be relevant, because some parts of the code do not exploit the symmetries. Thus, it is better to have a, say, 10x10x10 mesh with a shift instead of a, say, 20x20x20 mesh without a shift at a similar computational cost.

When you are interested in metals, do not forget to switch on the intraband transitions (the Drude term).

Markus

• Ao Teng - 2013-07-25

Thanks, Markus;)

Ao

On Jul 25, 2013, at 9:30, "Markus" markusmeinert@users.sf.net<mailto:markusmeinert@users.sf.net> wrote:

Hi Ao,

for a metal, scissor has no meaning: it shifts the unoccupied states up (or down) if you have a gap. This is to correct the so-called "band gap error".

dos (or wplot as it is called now) determines how many energy points you want in your plot and which range you want to have. See the manual for details. Just generate a plot and see if it meets your needs in terms of resolution and range. Remember that energies are in Hartree.

vkloff may be useful in order to save some calculation time by breaking the symmetry of the k-point mesh. This may be relevant, because some parts of the code do not exploit the symmetries. Thus, it is better to have a, say, 10x10x10 mesh with a shift instead of a, say, 20x20x20 mesh without a shift at a similar computational cost.

When you are interested in metals, do not forget to switch on the intraband transitions (the Drude term).

Markus

Determination of parameters for dielectric tensor calculationhttps://sourceforge.net/p/elk/discussion/897820/thread/d48b4fb9/?limit=25#7f87

Sent from sourceforge.nethttp://sourceforge.net because you indicated interest in https://sourceforge.net/p/elk/discussion/897820/

To unsubscribe from further messages, please visit https://sourceforge.net/auth/subscriptions/