This shows you the differences between two versions of the page.
Both sides previous revision Previous revision Next revision | Previous revision | ||
creating_inputs [2017/02/10 15:23] krejcio |
creating_inputs [2021/09/30 13:06] (current) krejcio Updated FHI-aims output; no need for parallel recompilation any more |
||
---|---|---|---|
Line 1: | Line 1: | ||
+ | //Note:// If you work with the **flexible-tip**=**PP-AFM** adjust the geometry into the necessary box for the force-field calculations. Don't forget to put the top-most atoms above ''0.0'', but lower part of the box in the //z// direction. | ||
+ | |||
At the moment the PP-STM code can read inputs from following DFT codes: | At the moment the PP-STM code can read inputs from following DFT codes: | ||
Line 84: | Line 86: | ||
- | == FHI-AIMS == | + | == FHI-aims == |
[[http://aimsclub.fhi-berlin.mpg.de/]] | [[http://aimsclub.fhi-berlin.mpg.de/]] | ||
- | Works for PBC calculations, just add: | + | Works for PBC calculations serial or parallel (mpi.scalapack) version. Ideally with the final geometry run a single point calculation with the following outputs added to your __control.in__: |
output eigenvectors | output eigenvectors | ||
output band 0 0 0 0.5 0.5 0.0 2 G K | output band 0 0 0 0.5 0.5 0.0 2 G K | ||
- | into __control.in__. | + | KS_method lapack_fast # for mpi-version (mpi.scalapack): The error message with HDF5 is not true in the aims output; use this keywords instead of recompilation # |
+ | See the [[https://github.com/ondrejkrejci/PPSTM/blob/master/tests/CuPc/control.in|CuPc example]] | ||
In the case of hybrid functionals (B3LYP, PBE0, HSE) add also: | In the case of hybrid functionals (B3LYP, PBE0, HSE) add also: | ||
- | exx_band_structure_version 1 ## for hybrid functionals only ## | + | exx_band_structure_version 1 ## for hybrid functionals only; both options 1 and 2 are working (1 takes more memory) ## |
into your __control.in__ file. | into your __control.in__ file. | ||
The calculations produces: __KS_eigenvectors.band_1.kpt_1.out__ for spin-restricted calculations or __KS_eigenvectors_dn.band_1.kpt_1.out__ & __KS_eigenvectors_up.band_1.kpt_1.out__ in the case of spin-polarized calculations. | The calculations produces: __KS_eigenvectors.band_1.kpt_1.out__ for spin-restricted calculations or __KS_eigenvectors_dn.band_1.kpt_1.out__ & __KS_eigenvectors_up.band_1.kpt_1.out__ in the case of spin-polarized calculations. | ||
+ | You will also need __geometry.in__ in the same directory. All of these files are supposed to be easily read by the __PPSTM_simple.py__ script or within __GUI.py__. | ||
+ | |||
+ | Note: Under normal conditions, please use PBC calculations, Just adjust the __geometry.in__ with a suitable ''lattice_vector''s (for small molecules ''30.0 30.0 30.0'' box is enough). There is a possibility to run strict cluster calculations, but additional parsing through Mathematica scripts (and some additional outputs) have to be used for creating PP-STM inputs. | ||
+ | |||
+ | == CP2K == | ||
+ | [[http://www.cp2k.org/]] | ||
+ | |||
+ | into your input file __YOUR_INPUT_FILE.inp__ into FORCE_EVAL → DFT → PRINT section add: | ||
- | Note: For cluster calculations Mathematica scripts have to be used for creating PP-STM inputs. | + | &MO ON |
+ | EIGVECS | ||
+ | CARTESIAN | ||
+ | FILENAME cartesian-mos | ||
+ | &EACH | ||
+ | QS_SCF 0 | ||
+ | &END EACH | ||
+ | &END MO | ||
+ | Examples of input (and output) files can be found in the code in __examples/__ __4N-coronene/__ __CuPc/__ and __TOAT/__ . | ||
+ | Note: CP2K reading procedure was mainly written by Ole Schütt, those days working at EMPA, Switzerland. |