exercises:2014_ethz_mmm:tio2_gap
Differences
This shows you the differences between two versions of the page.
Both sides previous revisionPrevious revisionNext revision | Previous revisionNext revisionBoth sides next revision | ||
exercise:tio2_gap [2014/04/10 15:20] – [TiO$_2$ Band Gap as a function of %hfx] oschuett | exercise:tio2_gap [2014/04/10 15:34] – oschuett | ||
---|---|---|---|
Line 1: | Line 1: | ||
====== TiO$_2$ Band Gap as a function of %hfx ====== | ====== TiO$_2$ Band Gap as a function of %hfx ====== | ||
- | One problem with standard DFT is that correlation effects can lead to errors in evaluating certain system properties, such as the band gap of semiconductors. More information can be found here: \\ | + | One problem with standard DFT is that correlation effects can lead to errors in evaluating certain system properties, such as the band gap of semiconductors. More information can be found here: |
- | A common approach to correct for electronic self-interaction is the //ad hoc// mixing of a fraction of the exact Hartree-Fock exchange. \\ | + | <note todo> |
+ | LINK? | ||
+ | </note> | ||
- | The goal of this exercise is to identify the needed amount of exact Hartree-Fock exchange (%hfx) to correctly reproduce the anatase TiO$_2$ experimental band gap (3.2 eV). \\ | + | A common approach to correct for electronic self-interaction is the //ad hoc// mixing of a fraction of the exact Hartree-Fock exchange. |
- | + | ||
- | To do so, you will need to run at least 4 single point calculations on bulk TiO$_2$ with varying amount of exact exchange. | + | |
Report the result as follows: | Report the result as follows: | ||
^%hfx^E$_{gap}$^ | ^%hfx^E$_{gap}$^ | ||
Line 12: | Line 13: | ||
| 10 | ... | | | 10 | ... | | ||
| .. | ... | | | .. | ... | | ||
- | Then plot E$_{gap}$ | + | Then plot E$_{gap}$ |
- | By interpolating the data on the plot, the needed amount of hfx to correctly reproduce the experimental band gap can be determined. | + | |
- | <note tip> Hybrid calculations can be fairly expensive, because they scale with $\mathcal{O}(N^4)$. Therefore, you should run these jobs on 16 nodes.</ | + | <note tip> |
- | <note tip> | + | Hybrid calculations can be fairly expensive, because they scale with $\mathcal{O}(N^4)$. Therefore, you should run these jobs on 16 nodes. To further |
+ | </ | ||
===== Questions ===== | ===== Questions ===== | ||
Line 23: | Line 24: | ||
===== Required files ===== | ===== Required files ===== | ||
- | * file with truncation parameters: | + | ==== Parameters for Truncated Coulomb Potential ==== |
- | | + | |
+ | | ||
+ | ==== Restart | ||
+ | This is useful to speed up the calculation | ||
+ | {{: | ||
< | < | ||
Line 32: | Line 37: | ||
</ | </ | ||
- | * input file | + | ==== Input File ==== |
<code - anatase_25hfx.inp> | <code - anatase_25hfx.inp> | ||
Line 43: | Line 48: | ||
METHOD Quickstep | METHOD Quickstep | ||
&DFT | &DFT | ||
- | ! external data-files for basis-set, pseudo-potentials and wavefunction | + | ! external data-files for basis-set, pseudo-potentials and wave-function |
| | ||
| | ||
| | ||
- | & | + | & |
CUTOFF 400 ! for this system the default value is too small | CUTOFF 400 ! for this system the default value is too small | ||
&END MGRID ! and can lead to non-physical results | &END MGRID ! and can lead to non-physical results | ||
Line 60: | Line 65: | ||
&END | &END | ||
| | ||
- | & | + | & |
EPS_SCF 1.0E-6 | EPS_SCF 1.0E-6 | ||
- | SCF_GUESS RESTART | + | SCF_GUESS RESTART |
MAX_SCF 40 | MAX_SCF 40 | ||
& | & | ||
Line 87: | Line 92: | ||
&HF | &HF | ||
FRACTION 0.25 ! this is the hfx section. The amount of hfx must be consistent with above | FRACTION 0.25 ! this is the hfx section. The amount of hfx must be consistent with above | ||
- | & | + | & |
EPS_SCHWARZ 1.0E-6 | EPS_SCHWARZ 1.0E-6 | ||
- | SCREEN_ON_INITIAL_P TRUE ! having an external | + | SCREEN_ON_INITIAL_P TRUE ! having an external |
& | & | ||
&MEMORY | &MEMORY | ||
Line 112: | Line 117: | ||
COORD_FILE_NAME tio.cif | COORD_FILE_NAME tio.cif | ||
&END | &END | ||
- | &KIND O ! external basis and pseudopotentials | + | &KIND O ! external basis and pseudo-potentials |
BASIS_SET cpFIT3 | BASIS_SET cpFIT3 | ||
POTENTIAL GTH-PBE-q6 | POTENTIAL GTH-PBE-q6 | ||
Line 124: | Line 129: | ||
</ | </ | ||
- | + | ==== Basis Set ==== | |
- | * basis set | + | |
<code - BASIS_TiO > | <code - BASIS_TiO > | ||
Line 172: | Line 176: | ||
</ | </ | ||
- | * potential | + | ==== Pseudo-Potential ==== |
<code - POTENTIALS_TiO> | <code - POTENTIALS_TiO> | ||
# | # | ||
Line 194: | Line 198: | ||
</ | </ | ||
- | * anatase *.cif file (cristallographic | + | ==== Anatse Crystal Structure ==== |
+ | |||
+ | '' | ||
<code - tio.cif > | <code - tio.cif > |
exercises/2014_ethz_mmm/tio2_gap.txt · Last modified: 2020/08/21 10:15 by 127.0.0.1