EDIFF: Difference between revisions

From VASP Wiki
mNo edit summary
 
(13 intermediate revisions by 3 users not shown)
Line 1: Line 1:
{{TAGDEF|EDIFF|[real]|<math>10^{-4}</math>}}
{{TAGDEF|EDIFF|[real]|<math>10^{-4}</math>}}


Description: {{TAG|EDIFF}} specifies the global break condition for the electronic SC-loop.
Description: {{TAG|EDIFF}} specifies the global break condition for the electronic SC-loop. {{TAG|EDIFF}} is specified in units of eV.
----
----
The relaxation of the electronic degrees of freedom will be stopped if the total (free) energy change and the band structure energy change ('change of eigenvalues') between two steps are both smaller than {{TAG|EDIFF}}. For {{TAG|EDIFF}}=0, {{TAG|NELM}} electronic SC-steps will always be performed.
The relaxation of the electronic degrees of freedom stops if the total (free) energy change and the band-structure-energy change ('change of eigenvalues') between two steps are both smaller than {{TAG|EDIFF}} (in eV). For {{TAG|EDIFF}}=0, strictly {{TAG|NELM}} electronic self-consistency steps will be performed.


'''Mind''': In most cases the convergence speed is exponential. So if you want the total energy significant to 4 figures set {{TAG|EDIFF}}=<math>10^{-4}</math>. There is no real reason to use a much smaller number.
In most cases, the convergence speed is quadratic, so often the cost for the additional iterations is small. Hence, for well converged calculations, we strongly recommend to decrease {{TAG|EDIFF}} to 1E-6. For finite difference calculations (e.g. phonons), even {{TAG|EDIFF}} {{=}} 1E-7 might be required in order to obtain precise results. On the other hand, for large systems with many atoms and/or when using meta-GGA functionals, attaining an energy convergence of 1E-8 or even 1E-7 might be difficult. So, overall {{TAG|EDIFF}}= 1E-6 is likely the best compromise.


== Related Tags and Sections ==
== Related tags and articles ==
{{TAG|EDIFFG}}
{{TAG|EDIFFG}}


== Example Calculations using this Tag ==
{{sc|EDIFF|Examples|Examples that use this tag}}
{{TAG|Alpha-AlF3}}, {{TAG|Alpha-SiO2}}, {{TAG|bandgap of Si in GW}}, {{TAG|bandstructure of SrVO3 in GW}}, {{TAG|beta-tin Si}}, {{TAG|cd Si volume relaxation}}, {{TAG|CO on Ni 111 surface}}, {{TAG|collective jumps of a Pt adatom on fcc-Pt (001): Nudged Elastic Band Calculation}}, {{TAG|Constrained MD using a canonical ensemble}}, {{TAG|Constrained MD using a microcanonical ensemble}}, {{TAG|Constraining the local magnetic moments}}, {{TAG|Determining the Magnetic Anisotropy}}, {{TAG|dielectric properties of Si}}, {{TAG|Dielectric properties of Si using BSE}}, {{TAG|dielectric properties of SiC}}, {{TAG|Equilibrium volume of Si in the RPA}}, {{TAG|Estimation of J magnetic coupling}}, {{TAG|graphite interlayer distance}}, {{TAG|graphite MBD binding energy}}, {{TAG|graphite TS binding energy}}, {{TAG|H2O vibration}}, {{TAG|Including the Spin-Orbit Coupling}}, {{TAG|liquid Si}}, {{TAG|Model BSE calculation on Si}},{{TAG|Ni 100 surface relaxation}}, {{TAG|Ni 111 surface high precision}}, {{TAG|NiO}}, {{TAG|NiO GGA+U}}, {{TAG|NiO HSE06}}, {{TAG|NiO LSDA}}, {{TAG|NiO LSDA+U}}, {{TAG|Relaxed geometry}}, {{TAG|Si HSE bandstructure}}, {{TAG|Standard relaxation}}, {{TAG|TS search using the Improved Dimer Method}}, {{TAG|TS search using the NEB Method}}, {{TAG|Vibrational Analysis of the TS}}, {{TAG|vibrational frequencies of CO on Ni 111 surface}}
 
----
----
[[The_VASP_Manual|Contents]]


[[Category:INCAR]]
[[Category:INCAR tag]][[Category:Electronic minimization]]

Latest revision as of 20:24, 15 November 2023

EDIFF = [real]
Default: EDIFF =  

Description: EDIFF specifies the global break condition for the electronic SC-loop. EDIFF is specified in units of eV.


The relaxation of the electronic degrees of freedom stops if the total (free) energy change and the band-structure-energy change ('change of eigenvalues') between two steps are both smaller than EDIFF (in eV). For EDIFF=0, strictly NELM electronic self-consistency steps will be performed.

In most cases, the convergence speed is quadratic, so often the cost for the additional iterations is small. Hence, for well converged calculations, we strongly recommend to decrease EDIFF to 1E-6. For finite difference calculations (e.g. phonons), even EDIFF = 1E-7 might be required in order to obtain precise results. On the other hand, for large systems with many atoms and/or when using meta-GGA functionals, attaining an energy convergence of 1E-8 or even 1E-7 might be difficult. So, overall EDIFF= 1E-6 is likely the best compromise.

Related tags and articles

EDIFFG

Examples that use this tag