comp-chem/cp2k
1
0
Fork 0
mirror of https://github.com/cp2k/cp2k.git synced 2026-07-21 06:25:15 -04:00
Code Issues Projects Releases Packages Wiki Activity Actions

docs: add Quickstep getting-started material (#5100)

Browse source 
Co-authored-by: Thomas D. Kuehne <tkuehne@cp2k.org>
This commit is contained in:
Dynamics of Condensed Matter 2026-04-26 12:02:00 +02:00 • committed by GitHub
parent e887d2c5ab
commit c92e8381f7
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
28 changed files with 830 additions and 178 deletions
Download patch file Download diff file Expand all files Collapse all files

137
docs/getting-started/first-calculation.md
View file

@ -1,10 +1,139 @@
# Run first Calculation
# Run a First Calculation
Unfortunately no one has gotten around to writing this page yet :-(
This page walks through a small single-point energy calculation for a water molecule. It uses the
{term}`Quickstep` module, the Gaussian and plane wave ({term}`GPW`) method, a molecular Gaussian
basis set, and Goedecker-Teter-Hutter ({term}`GTH`) pseudopotentials. The example is intentionally
small enough to run in a few seconds while still showing the parts of a typical CP2K input file that
matter for larger calculations.
In the meantime, the following links might be helpful:
## Input File
- <https://www.cp2k.org/howto:static_calculation>
Save the following input as `h2o.inp`. The same file is also available as [](h2o.inp).
```text
&GLOBAL
PROJECT h2o
RUN_TYPE ENERGY
&END GLOBAL
&FORCE_EVAL
METHOD Quickstep
&DFT
BASIS_SET_FILE_NAME BASIS_MOLOPT
POTENTIAL_FILE_NAME GTH_POTENTIALS
&MGRID
CUTOFF 400
REL_CUTOFF 50
&END MGRID
&POISSON
PERIODIC NONE
PSOLVER MT
&END POISSON
&SCF
EPS_SCF 1.0E-6
MAX_SCF 50
&END SCF
&XC
&XC_FUNCTIONAL PBE
&END XC_FUNCTIONAL
&END XC
&END DFT
&SUBSYS
&CELL
ABC 10.0 10.0 10.0
PERIODIC NONE
&END CELL
&COORD
O 5.0000 5.0000 5.0000
H 5.7586 5.0000 5.5043
H 4.2414 5.0000 5.5043
&END COORD
&KIND O
BASIS_SET DZVP-MOLOPT-GTH
POTENTIAL GTH-PBE-q6
&END KIND
&KIND H
BASIS_SET DZVP-MOLOPT-GTH
POTENTIAL GTH-PBE-q1
&END KIND
&END SUBSYS
&END FORCE_EVAL
```
## Running CP2K
Run the calculation with one of the installed CP2K binaries:
```bash
OMP_NUM_THREADS=1 cp2k.psmp -i h2o.inp -o h2o.out
```
The executable name depends on how CP2K was built:
| executable | meaning |
| ----------- | --------------------------- |
| `cp2k.psmp` | MPI + OpenMP parallel build |
| `cp2k.pdbg` | MPI + OpenMP debug build |
| `cp2k.ssmp` | serial/OpenMP build |
| `cp2k.sdbg` | serial/OpenMP debug build |
For MPI-parallel runs, launch CP2K through the MPI launcher used on your system, for example:
```bash
mpirun -np 2 -x OMP_NUM_THREADS=1 cp2k.psmp -i h2o.inp -o h2o.out
```
`cp2k.psmp` supports both MPI and OpenMP. Setting `OMP_NUM_THREADS=1` keeps this first example in a
simple MPI-only layout. To see the output on screen while also saving it, replace `-o h2o.out` with
`| tee h2o.out`.
## What the Input Does
[RUN_TYPE](#CP2K_INPUT.GLOBAL.RUN_TYPE) is set to `ENERGY`, so CP2K evaluates the electronic ground
state energy without moving the atoms.
[METHOD](#CP2K_INPUT.FORCE_EVAL.METHOD) selects `Quickstep`, CP2K's electronic-structure module for
Gaussian-based density functional theory and related methods.
`BASIS_SET_FILE_NAME` and `POTENTIAL_FILE_NAME` tell CP2K where to find the basis-set and
pseudopotential libraries. The matching [KIND](#CP2K_INPUT.FORCE_EVAL.SUBSYS.KIND) sections then
choose one Gaussian basis set and one GTH pseudopotential for each element.
[CUTOFF](#CP2K_INPUT.FORCE_EVAL.DFT.MGRID.CUTOFF) and
[REL_CUTOFF](#CP2K_INPUT.FORCE_EVAL.DFT.MGRID.REL_CUTOFF) control the real-space integration grids
used by the GPW method. They are not a replacement for increasing the Gaussian basis quality; for
accurate work the basis set and grid parameters should be converged together.
The [POISSON](#CP2K_INPUT.FORCE_EVAL.DFT.POISSON) section and the
[CELL](#CP2K_INPUT.FORCE_EVAL.SUBSYS.CELL) section both use `PERIODIC NONE`, which is appropriate
for this isolated molecule in a large non-periodic box.
## Checking the Result
At the end of `h2o.out`, CP2K prints the total energy in Hartree:
```text
ENERGY| Total FORCE_EVAL ( QS ) energy [hartree]
```
You should also see a line stating that the self-consistent field ({term}`SCF`) cycle converged. If
the SCF cycle does not converge, increase `MAX_SCF`, improve the initial guess, or use a more robust
SCF setup.
The timing table printed at the end of every CP2K run is useful for a first performance check. For
larger calculations, compare timings between MPI/OpenMP layouts and watch whether most of the time
is spent in grid operations, sparse matrix operations, diagonalization, or communication.
## Next Steps
- Converge [CUTOFF](#CP2K_INPUT.FORCE_EVAL.DFT.MGRID.CUTOFF) and
[REL_CUTOFF](#CP2K_INPUT.FORCE_EVAL.DFT.MGRID.REL_CUTOFF): [](../methods/dft/cutoff)
- Learn the idea behind GPW: [](../methods/dft/gpw)
- Learn about basis sets and pseudopotentials: [](../methods/dft/basis_sets),
[](../methods/dft/pseudopotentials)
- Build or install CP2K: [](build-from-source), [](build-with-spack), [](distributions)
- Explore more complete examples: <https://github.com/cp2k/cp2k-examples>
- Read the practical CP2K overview paper: [](#Iannuzzi2026)
```{youtube} qMR-NAaUheg
---

47
docs/getting-started/h2o.inp Normal file
View file

@ -0,0 +1,47 @@
&GLOBAL
PROJECT h2o
RUN_TYPE ENERGY
&END GLOBAL
&FORCE_EVAL
METHOD Quickstep
&DFT
BASIS_SET_FILE_NAME BASIS_MOLOPT
POTENTIAL_FILE_NAME GTH_POTENTIALS
&MGRID
CUTOFF 400
REL_CUTOFF 50
&END MGRID
&POISSON
PERIODIC NONE
PSOLVER MT
&END POISSON
&SCF
EPS_SCF 1.0E-6
MAX_SCF 50
&END SCF
&XC
&XC_FUNCTIONAL PBE
&END XC_FUNCTIONAL
&END XC
&END DFT
&SUBSYS
&CELL
ABC 10.0 10.0 10.0
PERIODIC NONE
&END CELL
&COORD
O 5.0000 5.0000 5.0000
H 5.7586 5.0000 5.5043
H 4.2414 5.0000 5.5043
&END COORD
&KIND O
BASIS_SET DZVP-MOLOPT-GTH
POTENTIAL GTH-PBE-q6
&END KIND
&KIND H
BASIS_SET DZVP-MOLOPT-GTH
POTENTIAL GTH-PBE-q1
&END KIND
&END SUBSYS
&END FORCE_EVAL

97
docs/methods/dft/basis_sets.md
View file

@ -1,10 +1,103 @@
# Basis Sets
Unfortunately no one has gotten around to writing this page yet :-(
In CP2K's {term}`Quickstep` module, the Kohn-Sham orbitals are expanded in atom-centered Gaussian
basis functions. This is different from pure plane-wave codes: increasing the real-space grid
[CUTOFF](#CP2K_INPUT.FORCE_EVAL.DFT.MGRID.CUTOFF) improves the auxiliary plane-wave representation
of densities and potentials, but it does not by itself reach the complete basis set limit. For
systematic convergence, the Gaussian basis quality and the grid parameters have to be considered
together.
In the meantime, the following links might be helpful:
Basis sets are selected for each atomic [KIND](#CP2K_INPUT.FORCE_EVAL.SUBSYS.KIND). The files that
contain the basis definitions are listed in
[BASIS_SET_FILE_NAME](#CP2K_INPUT.FORCE_EVAL.DFT.BASIS_SET_FILE_NAME). CP2K searches these files in
the current directory and in the configured CP2K data directory.
```text
&FORCE_EVAL
&DFT
BASIS_SET_FILE_NAME BASIS_MOLOPT
&END DFT
&SUBSYS
&KIND O
BASIS_SET DZVP-MOLOPT-GTH
&END KIND
&KIND H
BASIS_SET DZVP-MOLOPT-GTH
&END KIND
&END SUBSYS
&END FORCE_EVAL
```
## Basis Set Names
Many CP2K basis sets encode their purpose in the name:
- `SZV`, `DZVP`, `TZVP`, `TZV2P`, and `QZVPP` indicate increasing basis quality.
- `MOLOPT` basis sets are molecularly optimized Gaussian basis sets commonly used with GPW.
- `SR` indicates short-range MOLOPT variants. They are less diffuse and often more efficient for
large condensed-phase systems when the target property does not require diffuse functions.
- `GTH` basis sets are intended for GTH pseudopotential calculations.
- `q1`, `q4`, `q6`, and similar suffixes indicate the number of valence electrons represented by the
matching pseudopotential.
- `ae` basis sets are all-electron basis sets, commonly used with the GAPW method and
`POTENTIAL ALL`.
For production calculations, use basis sets and pseudopotentials that were designed to work
together. For example, `DZVP-MOLOPT-GTH` for oxygen is normally paired with `GTH-PBE-q6` in a PBE
calculation, while UZH protocol basis sets from `BASIS_MOLOPT_UZH` are paired with corresponding
entries from `POTENTIAL_UZH`. For new GPW production inputs, these UZH protocol pairs are the
preferred starting point where available; the older MOLOPT/GTH libraries remain useful for
compatibility and comparison with established inputs. The same `BASIS_MOLOPT_UZH` file also contains
all-electron MOLOPT basis sets for GAPW simulations, such as `SVP-MOLOPT-GGA-ae`,
`TZVPP-MOLOPT-GGA-ae`, and `QZVPP-MOLOPT-GGA-ae`, to be used with `POTENTIAL ALL`.
## Basis Set Roles
The keyword [BASIS_SET](#CP2K_INPUT.FORCE_EVAL.SUBSYS.KIND.BASIS_SET) can carry an optional basis
type. Without an explicit type, CP2K uses the primary orbital basis:
```text
&KIND O
BASIS_SET ORB DZVP-MOLOPT-GTH
&END KIND
```
This is equivalent to the more common short form:
```text
&KIND O
BASIS_SET DZVP-MOLOPT-GTH
&END KIND
```
Other basis roles are used by specific methods, for example:
- `AUX_FIT` for the auxiliary density matrix method.
- `RI_AUX` for resolution-of-the-identity correlation methods.
- `LRI` for local resolution-of-the-identity approaches.
The method-specific documentation usually states which auxiliary basis is required.
## Convergence and Practical Choices
Start with a basis set that is appropriate for the target accuracy, then converge the grid
parameters. For many routine condensed-phase GPW calculations, double-zeta or triple-zeta MOLOPT
basis sets are common starting points. For new setups, first check whether a matching UZH protocol
basis and pseudopotential pair is available. Accurate energy differences, weak interactions,
response properties, and post-Hartree-Fock methods may require larger or more specialized basis
sets.
Diffuse basis functions can improve accuracy for molecular anions, excited states, polarizabilities,
and weak interactions, but they also increase the cost and may make the overlap matrix more
ill-conditioned. In periodic calculations, diffuse functions can also increase the number of
periodic images that have to be considered.
For a simple tested example, see [](../../getting-started/first-calculation).
## See Also
- <https://en.wikipedia.org/wiki/Gaussian_orbital>
- <https://www.cp2k.org/basis_sets>
- <https://www.cp2k.org/tools:cp2k-basis>
- [](#VandeVondele2007)
- [](#Iannuzzi2026)

77
docs/methods/dft/gapw.md
View file

@ -1,11 +1,82 @@
# Gaussian Augmented Plane Waves
Unfortunately no one has gotten around to writing this page yet :-(
The Gaussian augmented plane wave ({term}`GAPW`) method extends GPW so that all-electron
calculations and calculations with very small-core pseudopotentials become practical in CP2K. The
central idea is to keep the smooth part of the density on the regular GPW grids while treating the
rapidly varying density close to the nuclei with atom-centered contributions.
In the meantime, the following links might be helpful:
GAPW is useful when the core electron density matters, for example in all-electron calculations,
core-level spectroscopy, magnetic properties, and some small-core pseudopotential setups. For
standard valence-only pseudopotential DFT calculations, GPW is usually simpler and faster.
## Activating GAPW
GAPW is activated in the [QS](#CP2K_INPUT.FORCE_EVAL.DFT.QS) section:
```text
&FORCE_EVAL
METHOD Quickstep
&DFT
&QS
METHOD GAPW
&END QS
&END DFT
&END FORCE_EVAL
```
All-electron GAPW calculations also require all-electron basis sets and `POTENTIAL ALL` for the
corresponding atomic kinds:
```text
&KIND O
BASIS_SET SVP-MOLOPT-GGA-ae
POTENTIAL ALL
LEBEDEV_GRID 110
RADIAL_GRID 80
&END KIND
```
A complete tested water example is available as [](gapw_h2o.inp). It is intentionally small and is
meant as a starting point rather than as a production-quality benchmark.
## Accuracy Parameters
Several GAPW-specific tolerances control the split between soft grid-based and hard atom-centered
contributions:
- [EPSFIT](#CP2K_INPUT.FORCE_EVAL.DFT.QS.EPSFIT) controls how Gaussian exponents are split into the
hard and soft parts. Lowering it includes harder functions in the soft density and usually
requires a larger [CUTOFF](#CP2K_INPUT.FORCE_EVAL.DFT.MGRID.CUTOFF).
- [EPSRHO0](#CP2K_INPUT.FORCE_EVAL.DFT.QS.EPSRHO0) controls the range used for the hard compensation
density contribution.
- [EPSSVD](#CP2K_INPUT.FORCE_EVAL.DFT.QS.EPSSVD) controls the singular value decomposition tolerance
used for projector matrices.
The atom-centered integration grid is controlled per kind with
[LEBEDEV_GRID](#CP2K_INPUT.FORCE_EVAL.SUBSYS.KIND.LEBEDEV_GRID) and
[RADIAL_GRID](#CP2K_INPUT.FORCE_EVAL.SUBSYS.KIND.RADIAL_GRID). Increasing these values can improve
the electron count and the accuracy of properties that depend on the near-core density, but it also
increases cost.
## Practical Guidance
When setting up a GAPW calculation:
- Use an all-electron basis set for `POTENTIAL ALL`, or a basis set designed for the chosen
small-core pseudopotential.
- Inspect the electron count printed by CP2K after SCF convergence. It is a useful diagnostic for
the quality of the hard/soft density split.
- Tighten `EPSFIT`, `EPSRHO0`, `EPSSVD`, and the atomic grids only as much as needed for the target
property.
- Increase the density [CUTOFF](#CP2K_INPUT.FORCE_EVAL.DFT.MGRID.CUTOFF) when harder Gaussian
exponents are included in the soft density.
- Prefer GPW when the calculation does not need all-electron or near-core accuracy.
## See Also
- [](#Lippert1999)
- [](#VandeVondele2006)
- [](#Krack2000)
- [](#Iannuzzi2026)
```{youtube} L0hKLjvjIFU
---

56
docs/methods/dft/gapw_h2o.inp Normal file
View file

@ -0,0 +1,56 @@
&GLOBAL
PROJECT gapw_h2o
RUN_TYPE ENERGY
&END GLOBAL
&FORCE_EVAL
METHOD Quickstep
&DFT
BASIS_SET_FILE_NAME BASIS_MOLOPT_UZH
&MGRID
CUTOFF 800
REL_CUTOFF 80
&END MGRID
&POISSON
PERIODIC NONE
PSOLVER MT
&END POISSON
&QS
EPSFIT 1.0E-6
EPSRHO0 1.0E-8
EPSSVD 1.0E-10
METHOD GAPW
&END QS
&SCF
EPS_SCF 1.0E-6
MAX_SCF 50
&END SCF
&XC
&XC_FUNCTIONAL PBE
&END XC_FUNCTIONAL
&END XC
&END DFT
&SUBSYS
&CELL
ABC 10.0 10.0 10.0
PERIODIC NONE
&END CELL
&COORD
O 5.0000 5.0000 5.0000
H 5.7586 5.0000 5.5043
H 4.2414 5.0000 5.5043
&END COORD
&KIND O
BASIS_SET SVP-MOLOPT-GGA-ae
LEBEDEV_GRID 110
POTENTIAL ALL
RADIAL_GRID 80
&END KIND
&KIND H
BASIS_SET SVP-MOLOPT-GGA-ae
LEBEDEV_GRID 110
POTENTIAL ALL
RADIAL_GRID 80
&END KIND
&END SUBSYS
&END FORCE_EVAL

3
docs/methods/dft/gpw.md
View file

@ -36,7 +36,8 @@ e -- "Integrate" --> f
- <https://www.cp2k.org/gpw>
- <https://www.cp2k.org/quickstep>
- [](#VandeVondele2005)
- [](#Lippert1997)
- [](#K%C3%BChne2020)
```{youtube} v2vnZbhNEpw
---

87
docs/methods/dft/hartree-fock/admm.md
View file

@ -1,11 +1,94 @@
# HFX with ADMM
Unfortunately no one has gotten around to writing this page yet :-(
The auxiliary density matrix method ({term}`ADMM`) reduces the cost of Hartree-Fock exchange in
hybrid DFT calculations by projecting the density matrix from the primary orbital basis onto a
smaller auxiliary basis. CP2K evaluates exact exchange in the auxiliary basis and adds a correction
term for the difference between the primary and auxiliary exchange descriptions.
In the meantime, the following links might be helpful:
ADMM is most useful when exact exchange is the bottleneck, especially with larger or more diffuse
Gaussian basis sets. It is commonly used for hybrid DFT, and the ADMM2 variant is also supported by
several post-SCF methods that reuse exact-exchange machinery.
## Basic Setup
An ADMM calculation needs three pieces of input:
- a hybrid functional or another setup that evaluates Hartree-Fock exchange,
- an auxiliary basis set for each atomic kind, specified with `BASIS_SET AUX_FIT`,
- an [AUXILIARY_DENSITY_MATRIX_METHOD](#CP2K_INPUT.FORCE_EVAL.DFT.AUXILIARY_DENSITY_MATRIX_METHOD)
section that selects the ADMM variant and correction functional.
For example:
```text
&DFT
BASIS_SET_FILE_NAME BASIS_MOLOPT_UZH
BASIS_SET_FILE_NAME BASIS_ADMM_UZH
POTENTIAL_FILE_NAME POTENTIAL_UZH
&AUXILIARY_DENSITY_MATRIX_METHOD
ADMM_TYPE ADMMS
EXCH_CORRECTION_FUNC PBEX
&END AUXILIARY_DENSITY_MATRIX_METHOD
&XC
&XC_FUNCTIONAL PBE
&END XC_FUNCTIONAL
&HF
FRACTION 0.25
&END HF
&END XC
&END DFT
&SUBSYS
&KIND O
BASIS_SET ccGRB-D-q6
BASIS_SET AUX_FIT admm-dz-q6
POTENTIAL GTH-HYB-q6
&END KIND
&END SUBSYS
```
## Choosing the Auxiliary Basis
The auxiliary basis should be chosen for the primary basis family and for the intended accuracy. For
MOLOPT-style calculations, the `BASIS_ADMM_MOLOPT` family provides compact auxiliary bases. The
newer UZH basis-set collection includes `BASIS_ADMM_UZH` and related basis files for correlation
consistent setups. All-electron calculations can use all-electron auxiliary basis sets when
available.
The auxiliary basis is part of the approximation. A too small auxiliary basis can make the exchange
correction large and reduce accuracy; a too large one gives back less speedup. For production work,
test at least one larger auxiliary basis or compare against a smaller reference system without ADMM.
## Choosing the ADMM Variant
[ADMM_TYPE](#CP2K_INPUT.FORCE_EVAL.DFT.AUXILIARY_DENSITY_MATRIX_METHOD.ADMM_TYPE) is a shortcut that
sets the projection, purification, and scaling options consistently. `ADMM1` and `ADMM2` are the
original variants, while `ADMMS`, `ADMMP`, and `ADMMQ` use additional models introduced later.
`ADMM2` is often the most broadly supported variant for workflows beyond ground-state hybrid DFT.
The
[EXCH_CORRECTION_FUNC](#CP2K_INPUT.FORCE_EVAL.DFT.AUXILIARY_DENSITY_MATRIX_METHOD.EXCH_CORRECTION_FUNC)
keyword selects the exchange functional used for the ADMM correction. It should be chosen
consistently with the exchange part of the main exchange-correlation setup; `PBEX` is a common
choice for PBE-based hybrid calculations.
## Practical Checks
When using ADMM:
- keep the same primary basis and potential convergence checks that would be used without ADMM,
- check the sensitivity to the auxiliary basis size,
- compare total energies, forces, or target properties against a non-ADMM reference for a small
representative system,
- remember that ADMM accelerates the exchange calculation but does not replace convergence of the
primary Gaussian basis, real-space grid, or SCF thresholds.
## See Also
- [](#Guidon2009)
- [](#Guidon2010)
- [](#Merlot2014)
- [](#Iannuzzi2026)
```{youtube} snG4fbpI0_g
---

18
docs/methods/dft/index.md
View file

@ -18,9 +18,23 @@ pseudopotentials
cutoff
```
```{youtube} kYxOWYWxYcQ
Density functional theory in CP2K is primarily provided by the Quickstep module. Most production
calculations use the Gaussian and plane waves (GPW) method with Gaussian basis sets,
pseudopotentials, and real-space grids for densities and potentials. The Gaussian augmented plane
waves (GAPW) method extends the same framework to all-electron and more core-sensitive calculations.
For new inputs, first choose a consistent basis-set and potential pair, then converge the MGRID
cutoffs and the SCF settings for the target property. The pages in this section collect the main
Quickstep ingredients: GPW/GAPW, hybrid functionals and ADMM, local RI, constraints, k-points, basis
sets, pseudopotentials, and grid convergence.
## References
- [](#K%C3%BChne2020)
- [](#Iannuzzi2026)
```{youtube} 3Cw4h3MrZ8k
---
url_parameters: ?start=216
align: center
privacy_mode:
---

77
docs/methods/dft/pseudopotentials.md
View file

@ -1,8 +1,80 @@
# Pseudopotentials
Unfortunately no one has gotten around to writing this page yet :-(
Most GPW calculations in CP2K use norm-conserving Goedecker-Teter-Hutter ({term}`GTH`)
pseudopotentials. A pseudopotential removes chemically inactive core electrons from the explicit
electronic problem and represents their effect on the valence electrons through an effective
potential. This reduces the number of electrons and avoids the very hard core density that would
otherwise require extremely fine grids.
In the meantime, the following links might be helpful:
Pseudopotential files are selected in
[POTENTIAL_FILE_NAME](#CP2K_INPUT.FORCE_EVAL.DFT.POTENTIAL_FILE_NAME), and the actual potential is
selected for each atomic [KIND](#CP2K_INPUT.FORCE_EVAL.SUBSYS.KIND) with
[POTENTIAL](#CP2K_INPUT.FORCE_EVAL.SUBSYS.KIND.POTENTIAL):
```text
&FORCE_EVAL
&DFT
POTENTIAL_FILE_NAME GTH_POTENTIALS
&END DFT
&SUBSYS
&KIND O
POTENTIAL GTH-PBE-q6
&END KIND
&KIND H
POTENTIAL GTH-PBE-q1
&END KIND
&END SUBSYS
&END FORCE_EVAL
```
The suffix `q6` in `GTH-PBE-q6`, for example, means that six valence electrons are treated
explicitly. The chosen basis set should match this valence configuration; for oxygen, a common
matching basis is `DZVP-MOLOPT-GTH`.
## Choosing a Pseudopotential
Use a pseudopotential generated for the exchange-correlation functional family used in the
calculation. For example, `GTH-PBE-q6` is a natural choice for PBE calculations with oxygen. Mixing
functional families can be acceptable for exploratory work in some cases, but it is not a systematic
route to high accuracy.
The CP2K data directory contains several pseudopotential libraries:
- `GTH_POTENTIALS` contains widely used GTH potentials for common GPW calculations.
- `POTENTIAL_UZH` contains the UZH protocol GTH potentials designed to be used with matching UZH
basis sets.
- `NLCC_POTENTIALS` and `GTH_SOC_POTENTIALS` contain more specialized potentials.
- `ECP_POTENTIALS` contains effective core potentials for Gaussian integral based calculations.
For new GPW production inputs, prefer a matching UZH protocol pair from `POTENTIAL_UZH` and
`BASIS_MOLOPT_UZH` when it is available for the element and functional family. The older
`GTH_POTENTIALS` library remains important for reproducing established calculations and for cases
where a matching UZH setup is not available.
For all-electron calculations, use `POTENTIAL ALL` together with an all-electron basis set and the
GAPW method:
```text
&KIND O
BASIS_SET SVP-MOLOPT-GGA-ae
POTENTIAL ALL
&END KIND
```
## Consistency Checks
Useful checks when setting up a calculation are:
- The basis set and pseudopotential should be available in the files named in the `DFT` section.
- The pseudopotential valence charge should match the basis set suffix where such a suffix is used.
- The exchange-correlation functional should be consistent with the pseudopotential family.
- For heavy elements, decide whether a large-core, medium-core, small-core, or all-electron
description is appropriate for the property of interest.
For a tested minimal GPW input using GTH pseudopotentials, see
[](../../getting-started/first-calculation).
## See Also
- <https://en.wikipedia.org/wiki/Pseudopotential>
- <https://cp2k.org/static/potentials/>
@ -10,3 +82,4 @@ In the meantime, the following links might be helpful:
- [](#Goedecker1996)
- [](#Hartwigsen1998)
- [](#Krack2005)
- [](#Iannuzzi2026)

22
docs/methods/post_hartree_fock/index.md
View file

@ -11,17 +11,21 @@ rpa
low-scaling
```
```{youtube} paM4lPL_-aI
---
url_parameters: ?start=115
align: center
privacy_mode:
---
```
Post-Hartree-Fock methods in CP2K add wavefunction-based correlation, quasiparticle, or response
corrections on top of a converged reference calculation. The reference is usually Hartree-Fock, a
hybrid functional, or semilocal DFT, depending on the target method and property.
```{youtube} 1vUuethWhbs
Start with the [preliminaries](preliminaries) page when setting up MP2, RPA, or related methods. It
explains the common choices for primary and RI basis sets, pseudopotentials, reference orbitals,
memory layout, and GPW-based integral grids. The method-specific pages then cover canonical and
RI-MP2, RPA and SOS-MP2, and the low-scaling implementations for larger systems.
These methods are considerably more expensive than semilocal DFT. For production work, converge the
reference calculation first, then check basis-set size, auxiliary basis or auto-generated RI basis
settings, quadrature parameters, and memory distribution.
```{youtube} wyux20qVlck
---
url_parameters: ?start=10
align: center
privacy_mode:
---

39
docs/methods/qm_mm/gromacs.md
View file

@ -1,10 +1,41 @@
# QM/MM with Gromacs
# QM/MM with GROMACS
Unfortunately no one has gotten around to writing this page yet :-(
GROMACS can use CP2K as a quantum-mechanical engine for QM/MM simulations. In such workflows,
GROMACS handles the classical molecular mechanics model, topology, constraints, and sampling
machinery, while CP2K evaluates the electronic structure of the selected QM region and returns
energies and forces.
In the meantime, the following links might be helpful:
This setup is useful when the chemically active part of a system must be described with
electronic-structure methods, but the surrounding solvent, biomolecule, or material environment is
too large for a full QM treatment. Typical applications include reactions in enzymes, solvated
molecular complexes, and embedded active sites.
- [Installation guide](https://manual.gromacs.org/current/install-guide/index.html#building-with-cp2k-qm-mm-support)
## Setup Outline
1. Build GROMACS with CP2K QM/MM support and make sure it can find the CP2K executable and data
directory.
1. Prepare and equilibrate the classical system with ordinary GROMACS tools.
1. Select the QM atoms and provide a CP2K input fragment for the quantum region.
1. Choose the coupling model, charge treatment, and boundary handling consistently with the target
system.
1. Start from short test trajectories and inspect both the GROMACS and CP2K output before running
production simulations.
The CP2K part follows the same Quickstep setup principles as a standalone calculation: choose
compatible basis sets and pseudopotentials, converge the real-space grid, and use SCF settings that
are robust for the geometry changes expected during the trajectory.
```{youtube} zSt8KQ2Hf3c
---
align: center
privacy_mode:
---
```
## External Resources
- [Installation guide](https://manual.gromacs.org/current/install-guide/) (see the CP2K QM/MM build
instructions)
- [Best practices guide](https://docs.bioexcel.eu/qmmm_bpg/en/main/index.html)
- [Gromacs manual](https://manual.gromacs.org/current/reference-manual/special/qmmm.html)
- [Tutorial](https://github.com/bioexcel/gromacs-2022-cp2k-tutorial)

30
src/common/bibliography.F
View file

@ -81,7 +81,7 @@ MODULE bibliography
Clabaut2021, Ren2011, Ren2013, Cohen2000, Rogers2002, Filippetti2000, &
Limpanuparb2011, Martin2003, Yin2017, Goerigk2017, &
Wilhelm2016a, Wilhelm2016b, Wilhelm2017, Wilhelm2018, Wilhelm2021, Lass2018, &
cp2kqs2020, Behler2007, Behler2011, Schran2020a, Schran2020b, &
cp2kqs2020, Iannuzzi2026, Behler2007, Behler2011, Schran2020a, Schran2020b, &
Rycroft2009, Thomas2015, Brehm2018, Brehm2020, Shigeta2001, Heinecke2016, &
Brehm2021, Bussy2021a, Bussy2021b, Ditler2021, Ditler2022, Mattiat2019, &
Mattiat2022, Belleflamme2023, Knizia2013, Musaelian2023, Eriksen2020, &
@ -835,7 +835,7 @@ CONTAINS
title="Efficient and accurate Car-Parrinello-like approach to "// &
"Born-Oppenheimer molecular dynamics", &
source="Phys. Rev. Lett.", volume="98", pages="066401", &
year=2007, doi="10.1103/PhysRevLett.98.066401")
year=2007, doi="10.1103/PhysRevLett.98.066401", citation_key="Kühne2007")
CALL add_reference(key=Rengaraj2020, &
authors=s2a("V. Rengaraj", "M. Lass", "C. Plessl", "T. D. Kuhne"), &
@ -1520,7 +1520,7 @@ CONTAINS
year=2021)
CALL add_reference(key=Richters2018, &
authors=s2a("D. Richters", "M. Lass", "A. Walther", "C. Plessl", "T. D. Kuehne"), &
authors=s2a("D. Richters", "M. Lass", "A. Walther", "C. Plessl", "T. D. Kühne"), &
title="A General Algorithm to Calculate the Inverse Principal p-th Root of "// &
"Symmetric Positive Definite Matrices", &
source="Commun. Comput. Phys.", volume="25", pages="564-585", &
@ -1623,27 +1623,39 @@ CONTAINS
year=2021, doi="10.1021/acs.jctc.0c01282")
CALL add_reference(key=Lass2018, &
authors=s2a("M. Lass", "S. Mohr", "H. Wiebeler", "T. D. Kuehne", "C. Plessl"), &
authors=s2a("M. Lass", "S. Mohr", "H. Wiebeler", "T. D. Kühne", "C. Plessl"), &
title="A Massively Parallel Algorithm for the Approximate Calculation of "// &
"Inverse P-Th Roots of Large Sparse Matrices", &
source="Proceedings of the Platform for Advanced Scientific Computing (PASC) Conference", &
year=2018, doi="10.1145/3218176.3218231")
CALL add_reference(key=cp2kqs2020, &
authors=s2a("T. D. Kuehne", "M. Iannuzzi", "M. Del Ben", "V. V. Rybkin", &
authors=s2a("T. D. Kühne", "M. Iannuzzi", "M. Del Ben", "V. V. Rybkin", &
"P. Seewald", "F. Stein", "T. Laino", "R. Z. Khaliullin", &
"O. Schuett", "F. Schiffmann", "D. Golze", "J. Wilhelm", &
"O. Schütt", "F. Schiffmann", "D. Golze", "J. Wilhelm", &
"S. Chulkov", "M. H. Bani-Hashemian", "V. Weber", &
"U. Borstnik", "M. Taillefumier", "A. S. Jakobovits", &
"A. Lazzaro", "H. Pabst", "T. Mueller", "R. Schade", "M. Guidon", &
"A. Lazzaro", "H. Pabst", "T. Müller", "R. Schade", "M. Guidon", &
"S. Andermatt", "N. Holmberg", "G. K. Schenter", "A. Hehn", &
"A. Bussy", "F. Belleflamme", "G. Tabacchi", "A. Gloess", &
"A. Bussy", "F. Belleflamme", "G. Tabacchi", "A. Glöß", &
"M. Lass", "I. Bethune", "C. J. Mundy", "C. Plessl", &
"M. Watkins", "J. VandeVondele", "M. Krack", "J. Hutter"), &
title="CP2K: An electronic structure and molecular dynamics software package - Quickstep: "// &
"Efficient and accurate electronic structure calculations", &
source="J. Chem. Phys.", volume="152", pages="194103", &
year=2020, doi="10.1063/5.0007045")
year=2020, doi="10.1063/5.0007045", citation_key="Kühne2020")
CALL add_reference(key=Iannuzzi2026, &
authors=s2a("M. Iannuzzi", "J. Wilhelm", "F. Stein", "A. Bussy", &
"H. Elgabarty", "D. Golze", "A. Hehn", "M. Graml", &
"S. Marek", "B. Sertcan Gökmen", "C. Schran", "H. Forbert", &
"R. Z. Khaliullin", "A. Kozhevnikov", "M. Taillefumier", &
"R. Meli", "V. V. Rybkin", "M. Brehm", "R. Schade", "O. Schütt", &
"J. V. Pototschnig", "H. Mirhosseini", "A. Knüpfer", "D. Marx", &
"M. Krack", "J. Hutter", "T. D. Kühne"), &
title="The CP2K Program Package Made Simple", &
source="J. Phys. Chem. B", volume="130", pages="1237-1310", &
year=2026, doi="10.1021/acs.jpcb.5c05851")
CALL add_reference(key=Rycroft2009, &
authors=s2a("C. H. Rycroft"), &

63
src/common/reference_manager.F
View file

@ -121,22 +121,23 @@ CONTAINS
!> \param pages ...
!> \param year ...
!> \param doi ...
!> \param citation_key ...
!> \par History
!> 08.2007 created [Joost VandeVondele]
!> 07.2024 complete rewrite [Ole Schuett]
!> \note
!> - see bibliography.F for it use.
! **************************************************************************************************
SUBROUTINE add_reference(key, authors, title, source, volume, pages, year, doi)
SUBROUTINE add_reference(key, authors, title, source, volume, pages, year, doi, citation_key)
INTEGER, INTENT(OUT) :: key
CHARACTER(LEN=*), DIMENSION(:), INTENT(IN) :: authors
CHARACTER(LEN=*), INTENT(IN) :: title, source
CHARACTER(LEN=*), INTENT(IN), OPTIONAL :: volume, pages
INTEGER, INTENT(IN) :: year
CHARACTER(LEN=*), INTENT(IN), OPTIONAL :: doi
CHARACTER(LEN=*), INTENT(IN), OPTIONAL :: doi, citation_key
CHARACTER :: tmp
CHARACTER(LEN=default_string_length) :: author, citation_key, key_a, key_b
CHARACTER(LEN=default_string_length) :: author, citation_key_, key_a, key_b
INTEGER :: i, ires, match, mylen, periodloc
IF (nbib + 1 > max_reference) CPABORT("increase max_reference")
@ -168,29 +169,35 @@ CONTAINS
thebib(key)%ref%doi = TRIM(doi)
END IF
! construct a citation_key
author = authors(1)
periodloc = INDEX(author, '.', back=.TRUE.)
IF (periodloc > 0) author = author(periodloc + 1:)
CPASSERT(LEN_TRIM(author) > 0)
WRITE (citation_key, '(A,I4)') TRIM(author), year
IF (PRESENT(citation_key)) THEN
citation_key_ = citation_key
CPASSERT(LEN_TRIM(citation_key_) > 4)
ELSE
! construct a citation_key
author = authors(1)
periodloc = INDEX(author, '.', back=.TRUE.)
IF (periodloc > 0) author = author(periodloc + 1:)
CPASSERT(LEN_TRIM(author) > 0)
WRITE (citation_key_, '(A,I4)') TRIM(author), year
! avoid special characters in names, just remove them
mylen = LEN_TRIM(citation_key)
ires = 0
DO I = 1, mylen
IF (INDEX("0123456789thequickbrownfoxjumpsoverthelazydogTHEQUICKBROWNFOXJUMPSOVERTHELAZYDOG", citation_key(i:i)) /= 0) THEN
ires = ires + 1
tmp = citation_key(i:i)
citation_key(ires:ires) = tmp
END IF
END DO
citation_key(ires + 1:) = ""
CPASSERT(LEN_TRIM(citation_key) > 4) ! At least one character of the author should be left.
! avoid special characters in names, just remove them
mylen = LEN_TRIM(citation_key_)
ires = 0
DO I = 1, mylen
IF (INDEX("0123456789thequickbrownfoxjumpsoverthelazydogTHEQUICKBROWNFOXJUMPSOVERTHELAZYDOG", &
citation_key_(i:i)) /= 0) THEN
ires = ires + 1
tmp = citation_key_(i:i)
citation_key_(ires:ires) = tmp
END IF
END DO
citation_key_(ires + 1:) = ""
CPASSERT(LEN_TRIM(citation_key_) > 4) ! At least one character of the author should be left.
END IF
! avoid duplicates, search through the list for matches (case-insensitive)
mylen = LEN_TRIM(citation_key)
key_a = citation_key(1:mylen)
mylen = LEN_TRIM(citation_key_)
key_a = citation_key_(1:mylen)
CALL uppercase(key_a)
match = 0
DO I = 1, nbib - 1
@ -198,10 +205,16 @@ CONTAINS
CALL uppercase(key_b)
IF (key_a == key_b) match = match + 1
END DO
IF (match > 0) citation_key = citation_key(1:mylen)//CHAR(ICHAR('a') + match)
IF (match > 0) THEN
IF (PRESENT(citation_key)) THEN
CPABORT("explicit citation key already exists")
ELSE
citation_key_ = citation_key_(1:mylen)//CHAR(ICHAR('a') + match)
END IF
END IF
! finally store it
thebib(key)%ref%citation_key = citation_key
thebib(key)%ref%citation_key = citation_key_
END SUBROUTINE add_reference

54
src/input_cp2k_dft.F
View file

@ -126,12 +126,14 @@ CONTAINS
CPASSERT(.NOT. ASSOCIATED(section))
CALL section_create(section, __LOCATION__, name="DFT", &
description="Parameter needed by LCAO DFT programs", &
description="Controls electronic-structure settings for Quickstep and related "// &
"Gaussian-basis DFT methods.", &
n_keywords=3, n_subsections=4, repeats=.FALSE.)
NULLIFY (keyword)
CALL keyword_create(keyword, __LOCATION__, name="BASIS_SET_FILE_NAME", &
description="Name of the basis set file, may include a path", &
description="Name of a basis-set library file, optionally including a path. "// &
"This keyword can be repeated to search several basis-set files.", &
usage="BASIS_SET_FILE_NAME <FILENAME>", &
type_of_var=lchar_t, repeats=.TRUE., &
default_lc_val="BASIS_SET", n_var=1)
@ -139,7 +141,8 @@ CONTAINS
CALL keyword_release(keyword)
CALL keyword_create(keyword, __LOCATION__, name="POTENTIAL_FILE_NAME", &
description="Name of the pseudo potential file, may include a path", &
description="Name of the pseudopotential library file, optionally including a path. "// &
"The potential selected for each kind is set with KIND%POTENTIAL.", &
usage="POTENTIAL_FILE_NAME <FILENAME>", &
default_lc_val="POTENTIAL")
CALL section_add_keyword(section, keyword)
@ -319,7 +322,8 @@ CONTAINS
CALL keyword_create(keyword, __LOCATION__, &
name="SORT_BASIS", &
description="Sort basis sets according to a certain criterion. ", &
description="Sorts basis functions according to a selected criterion. "// &
"Sorting by exponent can improve data locality for selected exact-exchange and RI workflows.", &
enum_c_vals=s2a("DEFAULT", "EXP"), &
enum_i_vals=[basis_sort_default, basis_sort_zet], &
enum_desc=s2a("don't sort", "sort w.r.t. exponent"), &
@ -672,15 +676,16 @@ CONTAINS
NULLIFY (keyword)
CPASSERT(.NOT. ASSOCIATED(section))
CALL section_create(section, __LOCATION__, name="AUXILIARY_DENSITY_MATRIX_METHOD", &
description="Parameters needed for the ADMM method.", &
description="Controls the auxiliary density matrix method (ADMM), which evaluates "// &
"Hartree-Fock exchange on a smaller auxiliary basis and adds an exchange correction.", &
n_keywords=1, n_subsections=1, repeats=.FALSE., &
citations=[Guidon2010])
CALL keyword_create( &
keyword, __LOCATION__, &
name="ADMM_TYPE", &
description="Type of ADMM (sort name) as refered in literature. "// &
"This sets values for METHOD, ADMM_PURIFICATION_METHOD, and EXCH_SCALING_MODEL", &
description="Named ADMM variant from the literature. This shortcut sets METHOD, "// &
"ADMM_PURIFICATION_METHOD, and EXCH_SCALING_MODEL consistently for the selected variant.", &
enum_c_vals=s2a("NONE", "ADMM1", "ADMM2", "ADMMS", "ADMMP", "ADMMQ"), &
enum_desc=s2a("No short name is used, use specific definitions (default)", &
"ADMM1 method from Guidon2010", &
@ -745,8 +750,8 @@ CONTAINS
CALL keyword_create( &
keyword, __LOCATION__, &
name="EXCH_CORRECTION_FUNC", &
description="Exchange functional which is used for the ADMM correction. "// &
"LibXC implementations require linking with LibXC", &
description="Exchange functional used for the ADMM correction. It should be chosen consistently "// &
"with the exchange functional in the main XC setup. LibXC implementations require linking with LibXC.", &
enum_c_vals=s2a("DEFAULT", "PBEX", "NONE", "OPTX", "BECKE88X", &
"PBEX_LIBXC", "BECKE88X_LIBXC", "OPTX_LIBXC", "DEFAULT_LIBXC", "LDA_X_LIBXC"), &
enum_i_vals=[do_admm_aux_exch_func_default, do_admm_aux_exch_func_pbex, &
@ -1230,26 +1235,32 @@ CONTAINS
CPASSERT(.NOT. ASSOCIATED(section))
CALL section_create(section, __LOCATION__, name="mgrid", &
description="multigrid information", &
description="Controls the multigrid used by GPW/GAPW to represent densities, "// &
"potentials, and Gaussian products on real-space grids.", &
n_keywords=5, n_subsections=1, repeats=.FALSE.)
NULLIFY (keyword)
CALL keyword_create(keyword, __LOCATION__, name="NGRIDS", &
description="The number of multigrids to use", &
description="Number of multigrid levels. Smooth Gaussian products can be mapped to "// &
"coarser levels, while sharper products require finer levels.", &
usage="ngrids 1", default_i_val=4)
CALL section_add_keyword(section, keyword)
CALL keyword_release(keyword)
CALL keyword_create(keyword, __LOCATION__, name="cutoff", &
description="The cutoff of the finest grid level. Default value for "// &
"SE or DFTB calculation is 1.0 [Ry].", &
usage="cutoff 300", default_r_val=cp_unit_to_cp2k(value=280.0_dp, &
unit_str="Ry"), n_var=1, unit_str="Ry")
description= &
"Plane-wave cutoff of the finest real-space grid level. "// &
"Increasing this value improves the grid representation, but it is "// &
"not a substitute for converging the Gaussian basis set. "// &
"Default value for SE or DFTB calculation is 1.0 [Ry].", &
usage="cutoff 300", &
default_r_val=cp_unit_to_cp2k(value=280.0_dp, unit_str="Ry"), &
n_var=1, unit_str="Ry")
CALL section_add_keyword(section, keyword)
CALL keyword_release(keyword)
CALL keyword_create(keyword, __LOCATION__, name="progression_factor", &
description="Factor used to find the cutoff of the multigrids that"// &
" where not given explicitly", &
description="Factor used to derive the cutoff of coarser multigrid levels when "// &
"they are not given explicitly.", &
usage="progression_factor <integer>", default_r_val=3._dp)
CALL section_add_keyword(section, keyword)
CALL keyword_release(keyword)
@ -1271,11 +1282,10 @@ CONTAINS
CALL keyword_create(keyword, __LOCATION__, name="REL_CUTOFF", &
variants=["RELATIVE_CUTOFF"], &
description="Determines the grid at which a Gaussian is mapped,"// &
" giving the cutoff used for a gaussian with alpha=1."// &
" A value 50+-10Ry might be required for highly accurate results,"// &
" Or for simulations with a variable cell."// &
" Versions prior to 2.3 used a default of 30Ry.", &
description="Controls to which multigrid level a Gaussian product is mapped. "// &
"It is the reference cutoff for a Gaussian with exponent alpha=1. Larger values "// &
"keep more Gaussian products on finer grids and can be important for accurate "// &
"energies, forces, stress tensors, and variable-cell simulations.", &
usage="RELATIVE_CUTOFF real", default_r_val=20.0_dp, &
unit_str="Ry")
CALL section_add_keyword(section, keyword)

7
src/input_cp2k_force_eval.F
View file

@ -80,7 +80,8 @@ CONTAINS
NULLIFY (subsection)
NULLIFY (keyword)
CALL keyword_create(keyword, __LOCATION__, name="METHOD", &
description="Which method should be used to compute forces", &
description="Selects the method used by this FORCE_EVAL section to compute energies, "// &
"forces, and related properties.", &
usage="METHOD <STRING>", &
enum_c_vals=s2a("QS", &
"SIRIUS", &
@ -97,11 +98,11 @@ CONTAINS
"Molecular Mechanics", &
"Hybrid quantum classical", &
"Empirical Interatomic Potential", &
"Electronic structure methods (DFT, ...)", &
"Electronic structure methods in the Quickstep module, including GPW and GAPW DFT.", &
"Neural Network Potentials", &
"Use a combination of two of the above", &
"Perform an embedded calculation", &
"Recieve forces from i–PI client"), &
"Receive forces from an i-PI client"), &
enum_i_vals=[do_qs, do_sirius, do_fist, do_qmmm, do_eip, do_qs, do_nnp, do_mixed, do_embed, do_ipi], &
default_i_val=do_qs)
CALL section_add_keyword(section, keyword)

4
src/input_cp2k_global.F
View file

@ -390,8 +390,8 @@ CONTAINS
CALL keyword_create( &
keyword, __LOCATION__, name="RUN_TYPE", &
description="Type of run that you want to perform Geometry "// &
"optimization, md, montecarlo,...", &
description="Selects the top-level task CP2K should run, such as an energy, "// &
"energy-and-force, molecular dynamics, geometry optimization, or response calculation.", &
usage="RUN_TYPE MD", &
default_i_val=energy_force_run, &
citations=[Ceriotti2014, Schonherr2014], &

33
src/input_cp2k_hfx.F
View file

@ -60,14 +60,15 @@ CONTAINS
CPASSERT(.NOT. ASSOCIATED(section))
CALL section_create(section, __LOCATION__, name="HF", &
description="Sets up the Hartree-Fock parameters if requested ", &
description="Controls Hartree-Fock exchange for hybrid DFT, Hartree-Fock, "// &
"and related post-Hartree-Fock workflows.", &
n_keywords=5, n_subsections=2, repeats=.TRUE., &
citations=[Guidon2008, Guidon2009])
NULLIFY (keyword, print_key, subsection)
CALL keyword_create(keyword, __LOCATION__, name="FRACTION", &
description="The fraction of Hartree-Fock to add to the total energy. "// &
description="Fraction of Hartree-Fock exchange to add to the total energy. "// &
"1.0 implies standard Hartree-Fock if used with XC_FUNCTIONAL NONE. "// &
"NOTE: In a mixed potential calculation this should be set to 1.0, otherwise "// &
"all parts are multiplied with this factor. ", &
@ -213,7 +214,8 @@ CONTAINS
CPASSERT(.NOT. ASSOCIATED(section))
CALL section_create(section, __LOCATION__, name="INTERACTION_POTENTIAL", &
description="Sets up interaction potential if requested ", &
description="Defines the Coulomb, range-separated, mixed, or truncated interaction "// &
"operator used for Hartree-Fock exchange.", &
n_keywords=1, n_subsections=0, repeats=.FALSE., &
citations=[guidon2008, guidon2009])
@ -221,8 +223,8 @@ CONTAINS
CALL keyword_create( &
keyword, __LOCATION__, &
name="POTENTIAL_TYPE", &
description="Which interaction potential should be used "// &
"(Coulomb, longrange or shortrange).", &
description="Selects the interaction potential used for Hartree-Fock exchange. "// &
"Periodic hybrid calculations commonly use a short-range, truncated, or mixed potential.", &
usage="POTENTIAL_TYPE SHORTRANGE", &
enum_c_vals=s2a("COULOMB", "SHORTRANGE", "LONGRANGE", "MIX_CL", "GAUSSIAN", &
"MIX_LG", "IDENTITY", "TRUNCATED", "MIX_CL_TRUNC"), &
@ -275,9 +277,10 @@ CONTAINS
CALL keyword_release(keyword)
CALL keyword_create(keyword, __LOCATION__, name="CUTOFF_RADIUS", &
description="Determines cutoff radius (in Angstroms) for the truncated $\frac{1}{r}$ "// &
"potential or the shortrange $\frac{\mathrm{erfc}(\omega \cdot r)}{r}$ potential. Default value "// &
"for shortrange potential when this keyword is omitted is solved from "// &
description="Cutoff radius for the truncated $\frac{1}{r}$ potential or the short-range "// &
"$\frac{\mathrm{erfc}(\omega \cdot r)}{r}$ potential. For truncated Coulomb in a "// &
"periodic cell, choose a radius compatible with the cell dimensions. The default value "// &
"for short-range potentials when this keyword is omitted is solved from "// &
"$\frac{\mathrm{erfc}(\omega \cdot r)}{r} = \epsilon_{\mathrm{schwarz}}$ "// &
"by Newton-Raphson method, with $\epsilon_{\mathrm{schwarz}}$ set by SCREENING/EPS_SCHWARZ", &
usage="CUTOFF_RADIUS 10.0", type_of_var=real_t, & ! default_r_val=10.0_dp,&
@ -311,7 +314,7 @@ CONTAINS
CPASSERT(.NOT. ASSOCIATED(section))
CALL section_create(section, __LOCATION__, name="SCREENING", &
description="Sets up screening parameters if requested ", &
description="Controls screening thresholds for Hartree-Fock exchange integrals.", &
n_keywords=1, n_subsections=0, repeats=.FALSE., &
citations=[guidon2008, guidon2009])
@ -319,9 +322,8 @@ CONTAINS
CALL keyword_create( &
keyword, __LOCATION__, &
name="EPS_SCHWARZ", &
description="Screens the near field part of the electronic repulsion "// &
"integrals using the Schwarz inequality for the given "// &
"threshold.", &
description="Schwarz inequality threshold for screening near-field electronic repulsion integrals. "// &
"Tighter values reduce screening error but increase cost.", &
usage="EPS_SCHWARZ 1.0E-6", &
default_r_val=1.0E-10_dp)
CALL section_add_keyword(section, keyword)
@ -331,10 +333,9 @@ CONTAINS
CALL keyword_create( &
keyword, __LOCATION__, &
name="EPS_SCHWARZ_FORCES", &
description="Screens the near field part of the electronic repulsion "// &
"integrals using the Schwarz inequality for the given "// &
"threshold. This will be approximately the accuracy of the forces, "// &
"and should normally be similar to EPS_SCF. Default value is 100*EPS_SCHWARZ.", &
description="Schwarz threshold used for force-related electronic repulsion integrals. "// &
"This is approximately the force accuracy and should normally be similar to EPS_SCF. "// &
"Default value is 100*EPS_SCHWARZ.", &
usage="EPS_SCHWARZ_FORCES 1.0E-5", &
default_r_val=1.0E-6_dp)
CALL section_add_keyword(section, keyword)

15
src/input_cp2k_kpoints.F
View file

@ -66,12 +66,12 @@ CONTAINS
CPASSERT(.NOT. ASSOCIATED(section))
CALL section_create(section, __LOCATION__, name="KPOINTS", &
description="Sets up the kpoints.", &
description="Controls Brillouin-zone sampling with k-points.", &
n_keywords=1, n_subsections=0, repeats=.FALSE.)
NULLIFY (keyword)
CALL keyword_create(keyword, __LOCATION__, name="SCHEME", &
description="Kpoint scheme to be used. Available options are:"//newline// &
description="K-point generation scheme. Available options are:"//newline// &
"- `NONE`"//newline// &
"- `GAMMA`"//newline// &
"- `MONKHORST-PACK`"//newline// &
@ -96,7 +96,7 @@ CONTAINS
CALL keyword_create(keyword, __LOCATION__, name="UNITS", &
description="Special k-points are defined either in units"// &
" of reciprocal lattice vectors or in Cartesian coordinates in uints of 2Pi/len."// &
" of reciprocal lattice vectors or in Cartesian coordinates in units of 2Pi/len."// &
" B_VECTOR: in multiples of the reciprocal lattice vectors (b)."// &
" CART_ANGSTROM: In units of 2*Pi/Angstrom."// &
" CART_BOHR: In units of 2*Pi/Bohr.", &
@ -112,7 +112,7 @@ CONTAINS
CALL keyword_release(keyword)
CALL keyword_create(keyword, __LOCATION__, name="FULL_GRID", &
description="Use full non-reduced kpoint grid.", &
description="Use the full, non-symmetry-reduced k-point grid.", &
usage="FULL_GRID <LOGICAL>", &
default_l_val=.FALSE., lone_keyword_l_val=.TRUE.)
CALL section_add_keyword(section, keyword)
@ -133,7 +133,7 @@ CONTAINS
CALL keyword_release(keyword)
CALL keyword_create(keyword, __LOCATION__, name="PARALLEL_GROUP_SIZE", &
description="Number of processors to be used for a single kpoint."// &
description="Number of MPI processes to be used for a single k-point."// &
" This number must divide the total number of processes."// &
" The number of groups must divide the total number of kpoints."// &
" Value=-1 (smallest possible number of processes per group, satisfying the constraints)."// &
@ -145,7 +145,8 @@ CONTAINS
CALL keyword_release(keyword)
CALL keyword_create(keyword, __LOCATION__, name="WAVEFUNCTIONS", &
description="Use real/complex wavefunctions if possible.", &
description="Select whether real or complex wavefunctions should be used "// &
"when allowed by the k-point set.", &
usage="WAVEFUNCTIONS REAL", &
default_i_val=use_complex_wfn, &
enum_c_vals=s2a("REAL", "COMPLEX"), &
@ -196,7 +197,7 @@ CONTAINS
!
CALL keyword_create(keyword, __LOCATION__, name="UNITS", &
description="Special k-points are defined either in units"// &
" of reciprocal lattice vectors or in Cartesian coordinates in uints of 2Pi/len."// &
" of reciprocal lattice vectors or in Cartesian coordinates in units of 2Pi/len."// &
" B_VECTOR: in multiples of the reciprocal lattice vectors (b)."// &
" CART_ANGSTROM: In units of 2*Pi/Angstrom."// &
" CART_BOHR: In units of 2*Pi/Bohr.", &

31
src/input_cp2k_mp2.F
View file

@ -82,8 +82,8 @@ CONTAINS
CPASSERT(.NOT. ASSOCIATED(section))
CALL section_create(section, __LOCATION__, name="WF_CORRELATION", &
description="Sets up the wavefunction-based correlation methods as MP2, "// &
"RI-MP2, RI-SOS-MP2, RI-RPA and GW (inside RI-RPA). ", &
description="Controls wavefunction-based correlation methods such as MP2, "// &
"RI-MP2, RI-SOS-MP2, RI-RPA, and GW inside RI-RPA.", &
n_keywords=4, n_subsections=7, repeats=.TRUE., &
citations=[DelBen2012, DelBen2013, DelBen2015, DelBen2015b, Rybkin2016, &
Wilhelm2016a, Wilhelm2016b, Wilhelm2017, Wilhelm2018, Stein2022, &
@ -94,7 +94,7 @@ CONTAINS
CALL keyword_create( &
keyword, __LOCATION__, &
name="MEMORY", &
description="Maximum allowed total memory usage during MP2 methods [MiB].", &
description="Maximum allowed total memory usage during MP2 and related WF_CORRELATION methods [MiB].", &
usage="MEMORY 1500 ", &
default_r_val=1.024E+3_dp)
CALL section_add_keyword(section, keyword)
@ -228,13 +228,13 @@ CONTAINS
keyword, __LOCATION__, &
name="METHOD", &
citations=[DelBen2012, DelBen2013], &
description="Method that is used to compute the MP2 energy.", &
description="Selects the implementation used to compute the canonical MP2 energy.", &
usage="METHOD MP2_GPW", &
enum_c_vals=s2a("NONE", "DIRECT_CANONICAL", "MP2_GPW"), &
enum_i_vals=[mp2_method_none, mp2_method_direct, mp2_method_gpw], &
enum_desc=s2a("Skip MP2 calculation.", &
"Use the direct mp2 canonical approach.", &
"Use the GPW approach to MP2."), &
"Use the direct canonical MP2 approach.", &
"Use the GPW approach to MP2 integrals."), &
default_i_val=mp2_method_direct)
CALL section_add_keyword(section, keyword)
CALL keyword_release(keyword)
@ -269,7 +269,7 @@ CONTAINS
NULLIFY (keyword)
CALL keyword_create(keyword, __LOCATION__, name="_SECTION_PARAMETERS_", &
description="Putting the &RI_MP2 section activates RI-MP2 calculation.", &
description="Activates an RI-MP2 calculation.", &
usage="&RI_MP2 .TRUE.", &
default_l_val=.FALSE., lone_keyword_l_val=.TRUE.)
CALL section_add_keyword(section, keyword)
@ -512,13 +512,13 @@ CONTAINS
CPASSERT(.NOT. ASSOCIATED(section))
CALL section_create(section, __LOCATION__, name="RI_RPA", &
description="Parameters influencing RI-RPA and GW.", &
description="Controls RI-RPA and GW calculations.", &
n_keywords=8, n_subsections=4, repeats=.FALSE., &
citations=[DelBen2013, DelBen2015])
NULLIFY (keyword, subsection)
CALL keyword_create(keyword, __LOCATION__, name="_SECTION_PARAMETERS_", &
description="Putting the &RI_RPA section activates RI-RPA calculation.", &
description="Activates an RI-RPA calculation.", &
usage="&RI_RPA .TRUE.", &
default_l_val=.FALSE., lone_keyword_l_val=.TRUE.)
CALL section_add_keyword(section, keyword)
@ -543,7 +543,7 @@ CONTAINS
CALL keyword_create(keyword, __LOCATION__, name="QUADRATURE_POINTS", &
variants=["RPA_NUM_QUAD_POINTS"], &
description="Number of quadrature points for the numerical integration in the RI-RPA method.", &
description="Number of quadrature points for the RI-RPA numerical integration.", &
usage="QUADRATURE_POINTS 60", &
default_i_val=40)
CALL section_add_keyword(section, keyword)
@ -575,8 +575,8 @@ CONTAINS
keyword, __LOCATION__, &
name="MINIMAX_QUADRATURE", &
variants=["MINIMAX"], &
description="Use the Minimax quadrature scheme for performing the numerical integration. "// &
"Maximum number of quadrature point limited to 20.", &
description="Use the Minimax quadrature scheme for the numerical integration. "// &
"It usually needs fewer quadrature points than Clenshaw-Curtis but is limited to 20 points.", &
usage="MINIMAX_QUADRATURE", &
default_l_val=.FALSE., &
lone_keyword_l_val=.TRUE.)
@ -597,11 +597,10 @@ CONTAINS
CALL keyword_create( &
keyword, __LOCATION__, &
name="ADMM", &
description="Decide whether to perform ADMM in the exact exchange calc. for RPA and/or GW. "// &
description="Decide whether to use ADMM in the exact-exchange calculation for RPA and/or GW. "// &
"The ADMM XC correction is governed by the AUXILIARY_DENSITY_MATRIX_METHOD section in &DFT. "// &
"In most cases, the Hartree-Fock exchange is not too expensive and there is no need for ADMM, "// &
"ADMM can however provide significant speedup and memory savings in case of diffuse basis sets. "// &
"If it is a GW bandgap calculations, RI_SIGMA_X can also be used. ", &
"ADMM can provide significant speedup and memory savings, especially with diffuse basis sets. "// &
"For GW band-gap calculations, RI_SIGMA_X can also be used. ", &
usage="ADMM", &
default_l_val=.FALSE., &
lone_keyword_l_val=.TRUE.)

9
src/input_cp2k_poisson.F
View file

@ -108,7 +108,7 @@ CONTAINS
CPASSERT(.NOT. ASSOCIATED(section))
CALL section_create(section, __LOCATION__, name="poisson", &
description="Sets up the poisson resolutor.", &
description="Controls the Poisson solver and electrostatic boundary conditions used by DFT.", &
n_keywords=1, n_subsections=0, repeats=.FALSE.)
NULLIFY (keyword, subsection)
@ -137,10 +137,9 @@ CONTAINS
CALL keyword_release(keyword)
CALL keyword_create(keyword, __LOCATION__, name="PERIODIC", &
description="Specify the directions in which PBC apply. Important notice,"// &
" this only applies to the electrostatics."// &
" See the CELL section to specify the periodicity used for e.g. the pair lists."// &
" Typically the settings should be the same.", &
description="Specifies the directions in which periodic boundary conditions apply to electrostatics. "// &
"See the CELL section for the periodicity used by geometry and pair lists; "// &
"the settings are usually the same.", &
usage="PERIODIC (x|y|z|xy|xz|yz|xyz|none)", &
enum_c_vals=s2a("x", "y", "z", "xy", "xz", "yz", "xyz", "none"), &
enum_i_vals=[use_perd_x, use_perd_y, use_perd_z, &

9
src/input_cp2k_properties_dft.F
View file

@ -1533,9 +1533,8 @@ CONTAINS
CPASSERT(.NOT. ASSOCIATED(section))
CALL section_create(section, __LOCATION__, name="TDDFPT", &
description="Parameters needed to set up the Time-Dependent "// &
"Density Functional Perturbation Theory. "// &
"Current implementation works for hybrid functionals. ", &
description="Controls time-dependent density functional perturbation theory "// &
"(TDDFPT) calculations for electronic excitations and related properties.", &
n_keywords=14, n_subsections=4, repeats=.FALSE., &
citations=[Iannuzzi2005, Hanasaki2025, Hernandez2025])
@ -1543,7 +1542,7 @@ CONTAINS
CALL keyword_create(keyword, __LOCATION__, &
name="_SECTION_PARAMETERS_", &
description="Controls the activation of the TDDFPT procedure", &
description="Activates the TDDFPT procedure.", &
default_l_val=.FALSE., &
lone_keyword_l_val=.TRUE.)
CALL section_add_keyword(section, keyword)
@ -1582,7 +1581,7 @@ CONTAINS
CALL keyword_create(keyword, __LOCATION__, name="NPROC_STATE", &
description="Number of MPI processes to be used per excited state. "// &
"Default is to use all processors (0).", &
"Default is to use all MPI processes (0).", &
n_var=1, type_of_var=integer_t, &
default_i_val=0)
CALL section_add_keyword(section, keyword)

9
src/input_cp2k_qmmm.F
View file

@ -95,22 +95,23 @@ CONTAINS
CPASSERT(.NOT. ASSOCIATED(section))
CALL section_create(section, __LOCATION__, name="qmmm", &
description="Input for QM/MM calculations.", &
description="Controls QM/MM calculations, including mechanical, electrostatic, "// &
"Gaussian-expanded, and periodic embedding options.", &
n_keywords=6, n_subsections=3, repeats=.FALSE., &
citations=[Laino2005, Laino2006])
NULLIFY (keyword, subsection)
CALL keyword_create(keyword, __LOCATION__, name="E_COUPL", &
variants=s2a("QMMM_COUPLING", "ECOUPL"), &
description="Specifies the type of the QM - MM electrostatic coupling.", &
description="Selects the QM-MM coupling model used for the electrostatic interaction.", &
usage="E_COUPL GAUSS", &
enum_c_vals=s2a("NONE", "COULOMB", "GAUSS", "S-WAVE", "POINT_CHARGE"), &
enum_i_vals=[do_qmmm_none, do_qmmm_coulomb, do_qmmm_gauss, do_qmmm_swave, do_qmmm_pcharge], &
enum_desc=s2a("Mechanical coupling (i.e. classical point charge based)", &
"Using analytical 1/r potential (Coulomb) - not available for GPW/GAPW", &
"Using fast gaussian expansion of the electrostatic potential (Erf(r/rc)/r) "// &
"Using fast Gaussian expansion of the electrostatic potential (Erf(r/rc)/r) "// &
"- not available for DFTB.", &
"Using fast gaussian expansion of the s-wave electrostatic potential", &
"Using fast Gaussian expansion of the s-wave electrostatic potential", &
"Using quantum mechanics derived point charges interacting with MM charges"), &
default_i_val=do_qmmm_none)
CALL section_add_keyword(section, keyword)

10
src/input_cp2k_qs.F
View file

@ -169,7 +169,9 @@ CONTAINS
CALL keyword_create(keyword, __LOCATION__, name="EPSFIT", &
variants=["EPS_FIT"], &
description="GAPW: precision to give the extension of a hard gaussian ", &
description="GAPW: tolerance controlling the split of Gaussian basis functions "// &
"into hard atom-centered and soft grid-expanded parts. Smaller values include "// &
"harder Gaussians in the soft density and can require a larger MGRID cutoff.", &
usage="EPSFIT real", default_r_val=1.0E-4_dp)
CALL section_add_keyword(section, keyword)
CALL keyword_release(keyword)
@ -183,14 +185,16 @@ CONTAINS
CALL keyword_create(keyword, __LOCATION__, name="EPSSVD", &
variants=["EPS_SVD"], &
description="GAPW: tolerance used in the singular value decomposition of the projector matrix", &
description="GAPW: tolerance used in the singular value decomposition of the "// &
"projector matrix. Smaller values can improve numerical accuracy at increased cost.", &
usage="EPS_SVD real", default_r_val=1.0E-8_dp)
CALL section_add_keyword(section, keyword)
CALL keyword_release(keyword)
CALL keyword_create(keyword, __LOCATION__, name="EPSRHO0", &
variants=s2a("EPSVRHO0", "EPS_VRHO0"), &
description="GAPW : precision to determine the range of V(rho0-rho0soft)", &
description="GAPW: tolerance used to determine the range of the "// &
"V(rho0-rho0_soft) compensation contribution.", &
usage="EPSRHO0 real", default_r_val=1.0E-6_dp)
CALL section_add_keyword(section, keyword)
CALL keyword_release(keyword)

3
src/input_cp2k_resp.F
View file

@ -61,7 +61,8 @@ CONTAINS
CPASSERT(.NOT. ASSOCIATED(section))
CALL section_create(section, __LOCATION__, name="RESP", &
description="Requests a RESP fit of charges. When using a periodic "// &
description="Requests a restrained electrostatic potential (RESP) fit of atomic charges. "// &
"When using a periodic "// &
"Poisson solver and a periodic cell, the periodic RESP routines are "// &
"used. If the Hartree potential matches with the one of an isolated "// &
"system (i.e. isolated Poisson solver and big, nonperiodic cells), "// &

24
src/input_cp2k_scf.F
View file

@ -139,7 +139,7 @@ CONTAINS
CALL keyword_release(keyword)
CALL keyword_create(keyword, __LOCATION__, name="MAX_SCF", &
description="Maximum number of SCF iteration to be performed for one optimization", &
description="Maximum number of inner SCF iterations for one electronic optimization.", &
usage="MAX_SCF 200", default_i_val=50)
CALL section_add_keyword(section, keyword)
CALL keyword_release(keyword)
@ -167,7 +167,7 @@ CONTAINS
CALL keyword_release(keyword)
CALL keyword_create(keyword, __LOCATION__, name="EPS_SCF", &
description="Target accuracy for the SCF convergence.", &
description="Target convergence threshold for the inner SCF cycle.", &
usage="EPS_SCF 1.e-6", default_r_val=1.e-5_dp)
CALL section_add_keyword(section, keyword)
CALL keyword_release(keyword)
@ -206,7 +206,7 @@ CONTAINS
CALL keyword_create( &
keyword, __LOCATION__, name="SCF_GUESS", &
description="Change the initial guess for the wavefunction.", &
description="Selects how the initial wavefunction or density matrix is generated.", &
usage="SCF_GUESS RESTART", default_i_val=atomic_guess, &
enum_c_vals=s2a("ATOMIC", "RESTART", "RANDOM", "CORE", &
"HISTORY_RESTART", "MOPAC", "EHT", "SPARSE", "NONE"), &
@ -237,8 +237,9 @@ CONTAINS
CALL keyword_release(keyword)
CALL keyword_create(keyword, __LOCATION__, name="ADDED_MOS", &
description="Number of additional MOS added for each spin. Use -1 to add all available. "// &
"alpha/beta spin can be specified independently (if spin-polarized calculation requested).", &
description="Number of additional molecular orbitals added for each spin channel. "// &
"This is commonly needed for smearing, excited-state, or post-Hartree-Fock calculations. "// &
"Use -1 to add all available orbitals.", &
usage="ADDED_MOS", default_i_val=0, n_var=-1)
CALL section_add_keyword(section, keyword)
CALL keyword_release(keyword)
@ -434,13 +435,14 @@ CONTAINS
CPASSERT(.NOT. ASSOCIATED(section))
CALL section_create(section, __LOCATION__, name="OUTER_SCF", &
description="parameters controlling the outer SCF loop", &
description="Controls an outer SCF loop, often used to stabilize difficult OT convergence, "// &
"constraints, or other variables wrapped around the inner SCF cycle.", &
n_keywords=13, n_subsections=1, repeats=.FALSE.)
NULLIFY (keyword)
CALL keyword_create(keyword, __LOCATION__, name="_SECTION_PARAMETERS_", &
description="controls the activation of the outer SCF loop", &
description="Activates the outer SCF loop.", &
usage="&OUTER_SCF ON", default_l_val=.FALSE., lone_keyword_l_val=.TRUE.)
CALL section_add_keyword(section, keyword)
CALL keyword_release(keyword)
@ -525,7 +527,7 @@ CONTAINS
CALL keyword_release(keyword)
CALL keyword_create(keyword, __LOCATION__, name="MAX_SCF", &
description="The maximum number of outer loops ", &
description="Maximum number of outer SCF loops.", &
usage="MAX_SCF 20", default_i_val=50)
CALL section_add_keyword(section, keyword)
CALL keyword_release(keyword)
@ -1263,7 +1265,7 @@ CONTAINS
CALL section_create(section, __LOCATION__, &
name="SMEAR", &
description="Define the smearing of the MO occupation numbers", &
description="Controls smearing of MO occupation numbers for systems with small or zero gaps.", &
n_keywords=6, &
n_subsections=0, &
repeats=.FALSE.)
@ -1281,7 +1283,7 @@ CONTAINS
CALL keyword_create(keyword, __LOCATION__, &
name="METHOD", &
description="Smearing method to be applied", &
description="Selects the smearing method to apply.", &
usage="METHOD Fermi_Dirac", &
default_i_val=smear_gaussian, &
enum_c_vals=s2a("FERMI_DIRAC", "ENERGY_WINDOW", "LIST", "GAUSSIAN", &
@ -1323,7 +1325,7 @@ CONTAINS
CALL keyword_create(keyword, __LOCATION__, &
name="ELECTRONIC_TEMPERATURE", &
variants=s2a("ELEC_TEMP", "TELEC"), &
description="Electronic temperature in the case of Fermi-Dirac smearing", &
description="Electronic temperature used for Fermi-Dirac smearing.", &
repeats=.FALSE., &
n_var=1, &
type_of_var=real_t, &

24
src/input_cp2k_subsys.F
View file

@ -1059,7 +1059,8 @@ CONTAINS
CPASSERT(.NOT. ASSOCIATED(section))
CALL section_create(section, __LOCATION__, name="KIND", &
description="The description of the kind of the atoms (mostly for QM)", &
description="Defines settings shared by atoms of the same kind, such as basis sets, "// &
"pseudopotentials, all-electron treatment, and atom-centered grids.", &
n_keywords=20, n_subsections=1, repeats=.TRUE.)
NULLIFY (keyword)
@ -1071,8 +1072,8 @@ CONTAINS
CALL keyword_release(keyword)
CALL keyword_create(keyword, __LOCATION__, name="BASIS_SET", &
description="The primary Gaussian basis set (NONE implies no basis used, meaningful with GHOST). "// &
"Defaults are set for TYPE {ORB} and FORM {GTO}. Possible values for TYPE are "// &
description="Selects a Gaussian basis set for this kind. The default type is ORB and the default "// &
"form is GTO; NONE implies no basis and is meaningful for ghost atoms. Possible values for TYPE are "// &
"{ORB, AUX, MIN, RI_AUX, LRI, ...}. Possible values for "// &
"FORM are {GTO, STO}. Where STO results in a GTO expansion of a Slater type basis. "// &
"If a value for FORM is given, also TYPE has to be set explicitly.", &
@ -1185,8 +1186,13 @@ CONTAINS
CALL keyword_create(keyword, __LOCATION__, name="POTENTIAL", &
variants=["POT"], &
description="The type (ECP, ALL, GTH, UPS) and name of the pseudopotential for the defined kind.", &
usage="POTENTIAL [type] <POTENTIAL-NAME>", type_of_var=char_t, default_c_vals=[" ", " "], &
description= &
"The type (ECP, ALL, GTH, UPS) and name of the "// &
"pseudopotential for the defined kind. Use GTH potentials "// &
"for most GPW calculations, ECP for Gaussian-integral effective core "// &
"potentials, and ALL for all-electron calculations.", &
usage="POTENTIAL [type] <POTENTIAL-NAME>", type_of_var=char_t, &
default_c_vals=[" ", " "], &
citations=[Goedecker1996, Hartwigsen1998, Krack2005], n_var=-1)
CALL section_add_keyword(section, keyword)
CALL keyword_release(keyword)
@ -1251,15 +1257,15 @@ CONTAINS
CALL keyword_release(keyword)
CALL keyword_create(keyword, __LOCATION__, name="LEBEDEV_GRID", &
description="The number of points for the angular part of "// &
"the local grid (GAPW)", &
description="GAPW: size of the angular Lebedev grid used for "// &
"atom-centered integrations for this kind.", &
usage="LEBEDEV_GRID 40", default_i_val=50)
CALL section_add_keyword(section, keyword)
CALL keyword_release(keyword)
CALL keyword_create(keyword, __LOCATION__, name="RADIAL_GRID", &
description="The number of points for the radial part of "// &
"the local grid (GAPW)", &
description="GAPW: number of radial grid points used for atom-centered "// &
"integrations for this kind.", &
usage="RADIAL_GRID 70", default_i_val=50)
CALL section_add_keyword(section, keyword)
CALL keyword_release(keyword)

15
src/input_cp2k_xas.F
View file

@ -71,9 +71,8 @@ CONTAINS
CPASSERT(.NOT. ASSOCIATED(section))
CALL section_create(section, __LOCATION__, name="xas", &
description="Sets the method of choice to calculate core-level excitation spectra. "// &
"The occupied states from which we calculate the "// &
"excitation should be specified. "// &
description="Controls transition-potential and delta-SCF calculations of core-level excitation spectra. "// &
"The occupied states from which the excitations are calculated should be specified. "// &
"Localization of the orbitals may be useful.", &
n_keywords=10, n_subsections=1, repeats=.FALSE., &
citations=[Iannuzzi2007])
@ -381,7 +380,7 @@ CONTAINS
CPASSERT(.NOT. ASSOCIATED(section))
CALL section_create(section, __LOCATION__, name="XAS_TDP", &
description="XAS simulations using linear-response TDDFT. Excitation from "// &
"specified core orbitals are considered one at a time. In case of high "// &
"specified core orbitals is considered one at a time. In case of high "// &
"symmetry structures, donor core orbitals should be localized.", &
n_keywords=19, n_subsections=4, repeats=.FALSE.)
@ -423,7 +422,7 @@ CONTAINS
CALL keyword_create(keyword, __LOCATION__, name="EXCITATIONS", &
variants=["EXCITATION"], &
description="Specify the type of excitation to consider. In case of a "// &
"resctricted closed-shell ground state calculation, "// &
"restricted closed-shell ground state calculation, "// &
"RCS_SINGLET or/and RCS_TRIPLET can be chosen. In case of a "// &
"open-shell ground state calculation (either UKS or ROKS), "// &
"standard spin conserving excitation (OS_SPIN_CONS) or/and "// &
@ -442,7 +441,7 @@ CONTAINS
CALL keyword_create(keyword, __LOCATION__, name="EPS_PGF_XAS", &
variants=s2a("EPS_PGF", "EPS_PGF_XAS_TDP"), &
description="The threshold used to determine the spacial extent of all "// &
description="The threshold used to determine the spatial extent of all "// &
"primitive Gaussian functions used for the construction "// &
"of neighbor lists in the XAS_TDP method. "// &
"By default, takes the value of QS%EPS_PGF_ORB. Useful if "// &
@ -865,7 +864,7 @@ CONTAINS
description="A threshold to determine which primitive 3-center integrals "// &
"are kept for contraction, as the latter operation can be "// &
"expensive (especially for large basis sets ). "// &
"If |(ab|c)| < EPS_SCREENNING, it is discarded.", &
"If |(ab|c)| < EPS_SCREENING, it is discarded.", &
default_r_val=1.0E-8_dp, &
repeats=.FALSE.)
CALL section_add_keyword(subsubsection, keyword)
@ -957,7 +956,7 @@ CONTAINS
"XAS TDP calculations", repeats=.FALSE.)
CALL cp_print_key_section_create(print_key, __LOCATION__, name="SPECTRUM", &
description="Controles the dumping of the XAS TDP spectrum in output files", &
description="Controls the dumping of the XAS TDP spectrum in output files", &
print_level=low_print_level, filename="", common_iter_levels=3)
CALL section_add_subsection(subsection, print_key)
CALL section_release(print_key)

8
src/start/cp2k_runs.F
View file

@ -8,8 +8,9 @@
! **************************************************************************************************
MODULE cp2k_runs
USE atom, ONLY: atom_code
USE bibliography, ONLY: Hutter2014,&
cite_reference
USE bibliography, ONLY: Iannuzzi2026,&
cite_reference,&
cp2kqs2020
USE bsse, ONLY: do_bsse_calculation
USE cell_opt, ONLY: cp_cell_opt
USE cp2k_debug, ONLY: cp2k_debug_energy_and_forces
@ -205,7 +206,8 @@ CONTAINS
NULLIFY (globenv, force_env)
CALL cite_reference(Hutter2014)
CALL cite_reference(cp2kqs2020)
CALL cite_reference(Iannuzzi2026)
! Parse the input
input_file => read_input(input_declaration, input_file_name, &