Difference between revisions of "Density Functional Theory"
From William Parker Wiki
(→Parameters to Converge) |
(Added quote from Mattsson et al.) |
||
| Line 14: | Line 14: | ||
References: | References: | ||
| − | + | # [https://iopscience.iop.org/article/10.1088/0965-0393/13/1/R01/pdf Ann E Mattsson et al 2005 Modelling Simul. Mater. Sci. Eng. 13 R1] | |
| − | + | # [https://www.quantum-espresso.org/Doc/INPUT_PW.html Quantum ESPRESSO Input File Description] | |
| + | |||
| + | |||
| + | <h2>Quote from Reference 1</h2> | ||
| + | |||
| + | (a) Factors affecting precision: | ||
| + | # PPs: If used, identify them. Any deviation from standard, published PPs should be | ||
| + | described in sufficient detail for the work to be reproducible. | ||
| + | # k points: Report the sampling that was used and which convergence tests were | ||
| + | performed. | ||
| + | # Basis sets/kinetic energy cutoff: Basis sets should be identified and, if non-standard, | ||
| + | the reason for using them given. If feasible, a calculation should be repeated with | ||
| + | another appropriate basis set and the result reported. If plane waves are used the | ||
| + | cutoff should be given along with the convergence it affords. | ||
| + | # Trajectory length and time step in AIMD: Motivate why they are appropriate. | ||
| + | # Equilibration in AIMD: How are the initial configurations prepared? | ||
| + | # Fictitious electron mass in CPMD: Report and motivate the choosen mass. | ||
| + | |||
| + | (b) Factors affecting accuracy: | ||
| + | # Functional: First and foremost, which functional was used? It is a good practice to | ||
| + | repeat some of the key calculations using a second functional and report the result. | ||
| + | This provides an estimate of the accuracy as well as information highly important | ||
| + | for further development of functionals. Even negative results are valuable—which | ||
| + | functional not to use for a specific system. | ||
| + | # System size: Is the property under study converged with respect to the size of the | ||
| + | supercell or cluster? | ||
| + | # Relaxation: Report on whether only volume or full relaxation was used. Justify the | ||
| + | reliability of any results obtained for an unrelaxed system. | ||
| + | # Boundary conditions: The exact treatment of the boundary conditions should be | ||
| + | described for a simulation where the system is not a simple bulk crystal. | ||
Revision as of 16:34, 16 September 2022
Parameters to Converge
| Parameter | QE Input File Name |
|---|---|
| Basis function cutoff G2/2 < Ecutoff | ecutwfc |
| K-point mesh | first three numbers under K_POINTS |
References:
- Ann E Mattsson et al 2005 Modelling Simul. Mater. Sci. Eng. 13 R1
- Quantum ESPRESSO Input File Description
Quote from Reference 1
(a) Factors affecting precision:
- PPs: If used, identify them. Any deviation from standard, published PPs should be
described in sufficient detail for the work to be reproducible.
- k points: Report the sampling that was used and which convergence tests were
performed.
- Basis sets/kinetic energy cutoff: Basis sets should be identified and, if non-standard,
the reason for using them given. If feasible, a calculation should be repeated with another appropriate basis set and the result reported. If plane waves are used the cutoff should be given along with the convergence it affords.
- Trajectory length and time step in AIMD: Motivate why they are appropriate.
- Equilibration in AIMD: How are the initial configurations prepared?
- Fictitious electron mass in CPMD: Report and motivate the choosen mass.
(b) Factors affecting accuracy:
- Functional: First and foremost, which functional was used? It is a good practice to
repeat some of the key calculations using a second functional and report the result. This provides an estimate of the accuracy as well as information highly important for further development of functionals. Even negative results are valuable—which functional not to use for a specific system.
- System size: Is the property under study converged with respect to the size of the
supercell or cluster?
- Relaxation: Report on whether only volume or full relaxation was used. Justify the
reliability of any results obtained for an unrelaxed system.
- Boundary conditions: The exact treatment of the boundary conditions should be
described for a simulation where the system is not a simple bulk crystal.
