Basis set convergence of RPA-ACFDT calculations - Revision history

Kaltakm at 12:46, 23 November 2021

2021-11-23T12:46:28Z

← Older revision		Revision as of 12:46, 23 November 2021
Line 9:		Line 9:
	<math>E_{\mathrm{c}}({\mathbf{G}})=E_{\mathrm{c}}(\infty)+\frac{A}{{\mathbf{G}}^3}</math>.		<math>E_{\mathrm{c}}({\mathbf{G}})=E_{\mathrm{c}}(\infty)+\frac{A}{{\mathbf{G}}^3}</math>.

	Furthermore, the Coulomb kernel is smoothly truncated between {{TAG\|ENCUTGWSOFT}} and {{TAG\|ENCUTGW}} using a simple cosine like window function (Hann window function). The default for {{TAG\|ENCUTGWSOFT}} is 0.8<math>\times</math>{{TAG\|ENCUTGW}} (again we do not recommend to change this default).		Furthermore, the Coulomb kernel is smoothly truncated between {{TAG\|ENCUTGWSOFT}} and {{TAG\|ENCUTGW}} using a simple cosine like window function (Hann window function).
			Alternatively, the basis set extrapolation can be performed by setting {{TAG\|LSCK}}=.TRUE., using the squeezed Coulomb kernel method.{{cite\|riemelmoser:jcp:2020}}

			The default for {{TAG\|ENCUTGWSOFT}} is 0.8<math>\times</math>{{TAG\|ENCUTGW}} (again we do not recommend to change this default).

	The integral over <math>\omega</math> is evaluated by means of a highly accurate minimax integration.{{cite\|kaltak:2014}} The number of <math>\omega</math> points is determined by the flag {{TAG\|NOMEGA}}, whereas the energy range of transitions is determined by the band gap and the energy difference between the lowest occupied and highest unoccupied one-electron orbital. VASP determines these values automatically (from vasp.5.4.1 on), and the user should only carefully converge with respect to the number of frequency points {{TAG\|NOMEGA}}. A good choice is usually {{TAG\|NOMEGA}}=12, however, for large gap systems one might obtain <math>\mu</math>eV convergence per atom already using 8 points, whereas for metals up to {{TAG\|NOMEGA}}=24 frequency points are sometimes necessary, in particular, for large unit cells.		The integral over <math>\omega</math> is evaluated by means of a highly accurate minimax integration.{{cite\|kaltak:2014}} The number of <math>\omega</math> points is determined by the flag {{TAG\|NOMEGA}}, whereas the energy range of transitions is determined by the band gap and the energy difference between the lowest occupied and highest unoccupied one-electron orbital. VASP determines these values automatically (from vasp.5.4.1 on), and the user should only carefully converge with respect to the number of frequency points {{TAG\|NOMEGA}}. A good choice is usually {{TAG\|NOMEGA}}=12, however, for large gap systems one might obtain <math>\mu</math>eV convergence per atom already using 8 points, whereas for metals up to {{TAG\|NOMEGA}}=24 frequency points are sometimes necessary, in particular, for large unit cells.

Kaltakm at 12:15, 31 July 2019

2019-07-31T12:15:10Z

← Older revision		Revision as of 12:15, 31 July 2019
Line 1:		Line 1:
	~~To reach technical convergence, a number of flags are available to control the evaluation of the ACFDT-RPA correlation energy in the fourth step.~~ The expression for the ACFDT-RPA correlation energy reads:		The expression for the ACFDT-RPA correlation energy written in terms of reciprocal lattice vectors reads:

	<math>E_{\rm c}^{\rm RPA}=\int_{0}^{\infty} \frac{\mathrm{d}\omega}{2\pi} \sum_{{\mathbf{q}}\in \mathbf{BZ} }\sum_{{\mathbf{G}}} \left\{(\mathrm{ln}[1-\tilde\chi^0({\mathbf{q}},\mathrm{i}\omega)V({\mathbf{q}})])_{{\mathbf{G,G}}} +V_{{\mathbf{G,G}}}({\mathbf{q}})\tilde\chi^0({\mathbf{q}},{\mathrm{i}}\omega) \right\} </math>.		<math>E_{\rm c}^{\rm RPA}=\int_{0}^{\infty} \frac{\mathrm{d}\omega}{2\pi} \sum_{{\mathbf{q}}\in \mathbf{BZ} }\sum_{{\mathbf{G}}} \left\{(\mathrm{ln}[1-\tilde\chi^0({\mathbf{q}},\mathrm{i}\omega)V({\mathbf{q}})])_{{\mathbf{G,G}}} +V_{{\mathbf{G,G}}}({\mathbf{q}})\tilde\chi^0({\mathbf{q}},{\mathrm{i}}\omega) \right\} </math>.

Kaltakm at 16:41, 29 July 2019

2019-07-29T16:41:12Z

← Older revision		Revision as of 16:41, 29 July 2019
Line 4:		Line 4:
	The sum over reciprocal lattice vectors has to be truncated at some <math>\mathbf{G}_{\mathrm{max}}</math>, determined by <math>\frac{\hbar^2\|{\mathbf{G}}+{\mathbf{q}}\|^2}{2\mathrm{m}_e}</math> < {{TAG\|ENCUTGW}}, which can be set in the {{TAG\|INCAR}} file. The default value is <math>\frac{2}{3}\times</math> {{TAG\|ENCUT}}, which experience has taught us not to change. For systematic convergence tests, instead increase {{TAG\|ENCUT}} and repeat steps 1 to 4, but be aware that the "maximum number of plane-waves" changes when {{TAG\|ENCUT}} is increased. Note that it is virtually impossible, to converge absolute correlation energies. Rather concentrate on relative energies (e.g. energy differences between two solids, or between a solid and the constituent atoms).		The sum over reciprocal lattice vectors has to be truncated at some <math>\mathbf{G}_{\mathrm{max}}</math>, determined by <math>\frac{\hbar^2\|{\mathbf{G}}+{\mathbf{q}}\|^2}{2\mathrm{m}_e}</math> < {{TAG\|ENCUTGW}}, which can be set in the {{TAG\|INCAR}} file. The default value is <math>\frac{2}{3}\times</math> {{TAG\|ENCUT}}, which experience has taught us not to change. For systematic convergence tests, instead increase {{TAG\|ENCUT}} and repeat steps 1 to 4, but be aware that the "maximum number of plane-waves" changes when {{TAG\|ENCUT}} is increased. Note that it is virtually impossible, to converge absolute correlation energies. Rather concentrate on relative energies (e.g. energy differences between two solids, or between a solid and the constituent atoms).

	Since correlation energies converge very slowly with respect to <math>\mathbf{G}_{\~~mathrm~~ max }</math>, VASP automatically extrapolates to the infinite basis set limit using a linear regression to the equation: {{cite\|harl:2008}}{{cite\|harl:2010}}{{cite\|klimes:2014}}		Since correlation energies converge very slowly with respect to <math>\mathbf{G}_{\rm max }</math>, VASP automatically extrapolates to the infinite basis set limit using a linear regression to the equation: {{cite\|harl:2008}}{{cite\|harl:2010}}{{cite\|klimes:2014}}

	<math>E_{\mathrm{c}}({\mathbf{G}})=E_{\mathrm{c}}(\infty)+\frac{A}{{\mathbf{G}}^3}</math>.		<math>E_{\mathrm{c}}({\mathbf{G}})=E_{\mathrm{c}}(\infty)+\frac{A}{{\mathbf{G}}^3}</math>.

Kaltakm at 16:40, 29 July 2019

2019-07-29T16:40:30Z

← Older revision		Revision as of 16:40, 29 July 2019
Line 1:		Line 1:
	To reach technical convergence, a number of flags are available to control the evaluation of the ACFDT-RPA correlation energy in the fourth step. The expression for the ACFDT-RPA correlation energy reads:		To reach technical convergence, a number of flags are available to control the evaluation of the ACFDT-RPA correlation energy in the fourth step. The expression for the ACFDT-RPA correlation energy reads:
	<math>E_{\rm c}^{\rm RPA}=\int_{0}^{\infty} \frac{\mathrm{d}\omega}{2\pi} \sum_{{\mathbf{q}}\in \mathbf{BZ} }\sum_{{\mathbf{G}}} \left\{(\mathrm{ln}[1-\chi^0({\mathbf{q}},\mathrm{i}\omega)V({\mathbf{q}})])_{{\mathbf{G,G}}} +V_{{\mathbf{G,G}}}({\mathbf{q}})\chi^0({\mathbf{q}},{\mathrm{i}}\omega) \right\} </math>.		<math>E_{\rm c}^{\rm RPA}=\int_{0}^{\infty} \frac{\mathrm{d}\omega}{2\pi} \sum_{{\mathbf{q}}\in \mathbf{BZ} }\sum_{{\mathbf{G}}} \left\{(\mathrm{ln}[1-\tilde\chi^0({\mathbf{q}},\mathrm{i}\omega)V({\mathbf{q}})])_{{\mathbf{G,G}}} +V_{{\mathbf{G,G}}}({\mathbf{q}})\tilde\chi^0({\mathbf{q}},{\mathrm{i}}\omega) \right\} </math>.

	The sum over reciprocal lattice vectors has to be truncated at some <math>\mathbf{G}_{\mathrm{max}}</math>, determined by <math>\frac{\hbar^2\|{\mathbf{G}}+{\mathbf{q}}\|^2}{2\mathrm{m}_e}</math> < {{TAG\|ENCUTGW}}, which can be set in the {{TAG\|INCAR}} file. The default value is <math>\frac{2}{3}\times</math> {{TAG\|ENCUT}}, which experience has taught us not to change. For systematic convergence tests, instead increase {{TAG\|ENCUT}} and repeat steps 1 to 4, but be aware that the "maximum number of plane-waves" changes when {{TAG\|ENCUT}} is increased. Note that it is virtually impossible, to converge absolute correlation energies. Rather concentrate on relative energies (e.g. energy differences between two solids, or between a solid and the constituent atoms).		The sum over reciprocal lattice vectors has to be truncated at some <math>\mathbf{G}_{\mathrm{max}}</math>, determined by <math>\frac{\hbar^2\|{\mathbf{G}}+{\mathbf{q}}\|^2}{2\mathrm{m}_e}</math> < {{TAG\|ENCUTGW}}, which can be set in the {{TAG\|INCAR}} file. The default value is <math>\frac{2}{3}\times</math> {{TAG\|ENCUT}}, which experience has taught us not to change. For systematic convergence tests, instead increase {{TAG\|ENCUT}} and repeat steps 1 to 4, but be aware that the "maximum number of plane-waves" changes when {{TAG\|ENCUT}} is increased. Note that it is virtually impossible, to converge absolute correlation energies. Rather concentrate on relative energies (e.g. energy differences between two solids, or between a solid and the constituent atoms).

Kaltakm at 16:40, 29 July 2019

2019-07-29T16:40:10Z

← Older revision		Revision as of 16:40, 29 July 2019
Line 1:		Line 1:
	To reach technical convergence, a number of flags are available to control the evaluation of the ACFDT-RPA correlation energy in the fourth step. The expression for the ACFDT-RPA correlation energy reads:		To reach technical convergence, a number of flags are available to control the evaluation of the ACFDT-RPA correlation energy in the fourth step. The expression for the ACFDT-RPA correlation energy reads:
	<math>E_{\rm c}^{\rm RPA}=\int_{0}^{\infty} \frac{\mathrm{d}\omega}{2\pi} \sum_{{\mathbf{q}}\in \mathbf{BZ} }\sum_{{\mathbf{G}}} \left\{(\mathrm{ln}[1-\chi^~~{KS}~~({\mathbf{q}},\mathrm{i}\omega)V({\mathbf{q}})])_{{\mathbf{G,G}}} +V_{{\mathbf{G,G}}}({\mathbf{q}})\chi^~~{KS}~~({\mathbf{q}},{\mathrm{i}}\omega) \right\} </math>.		<math>E_{\rm c}^{\rm RPA}=\int_{0}^{\infty} \frac{\mathrm{d}\omega}{2\pi} \sum_{{\mathbf{q}}\in \mathbf{BZ} }\sum_{{\mathbf{G}}} \left\{(\mathrm{ln}[1-\chi^0({\mathbf{q}},\mathrm{i}\omega)V({\mathbf{q}})])_{{\mathbf{G,G}}} +V_{{\mathbf{G,G}}}({\mathbf{q}})\chi^0({\mathbf{q}},{\mathrm{i}}\omega) \right\} </math>.

	The sum over reciprocal lattice vectors has to be truncated at some <math>\mathbf{G}_{\mathrm{max}}</math>, determined by <math>\frac{\hbar^2\|{\mathbf{G}}+{\mathbf{q}}\|^2}{2\mathrm{m}_e}</math> < {{TAG\|ENCUTGW}}, which can be set in the {{TAG\|INCAR}} file. The default value is <math>\frac{2}{3}\times</math> {{TAG\|ENCUT}}, which experience has taught us not to change. For systematic convergence tests, instead increase {{TAG\|ENCUT}} and repeat steps 1 to 4, but be aware that the "maximum number of plane-waves" changes when {{TAG\|ENCUT}} is increased. Note that it is virtually impossible, to converge absolute correlation energies. Rather concentrate on relative energies (e.g. energy differences between two solids, or between a solid and the constituent atoms).		The sum over reciprocal lattice vectors has to be truncated at some <math>\mathbf{G}_{\mathrm{max}}</math>, determined by <math>\frac{\hbar^2\|{\mathbf{G}}+{\mathbf{q}}\|^2}{2\mathrm{m}_e}</math> < {{TAG\|ENCUTGW}}, which can be set in the {{TAG\|INCAR}} file. The default value is <math>\frac{2}{3}\times</math> {{TAG\|ENCUT}}, which experience has taught us not to change. For systematic convergence tests, instead increase {{TAG\|ENCUT}} and repeat steps 1 to 4, but be aware that the "maximum number of plane-waves" changes when {{TAG\|ENCUT}} is increased. Note that it is virtually impossible, to converge absolute correlation energies. Rather concentrate on relative energies (e.g. energy differences between two solids, or between a solid and the constituent atoms).

Kaltakm at 16:23, 29 July 2019

2019-07-29T16:23:50Z

← Older revision		Revision as of 16:23, 29 July 2019
Line 1:		Line 1:
	To reach technical convergence, a number of flags are available to control the evaluation of the ACFDT-RPA correlation energy in the fourth step. The expression for the ACFDT-RPA correlation energy reads:		To reach technical convergence, a number of flags are available to control the evaluation of the ACFDT-RPA correlation energy in the fourth step. The expression for the ACFDT-RPA correlation energy reads:
			<math>E_{\rm c}^{\rm RPA}=\int_{0}^{\infty} \frac{\mathrm{d}\omega}{2\pi} \sum_{{\mathbf{q}}\in \mathbf{BZ} }\sum_{{\mathbf{G}}} \left\{(\mathrm{ln}[1-\chi^{KS}({\mathbf{q}},\mathrm{i}\omega)V({\mathbf{q}})])_{{\mathbf{G,G}}} +V_{{\mathbf{G,G}}}({\mathbf{q}})\chi^{KS}({\mathbf{q}},{\mathrm{i}}\omega) \right\} </math>.
	<math>\~~mathrm{E~~}_{\~~mathrm{c}~~}=\int_{0}^{\infty} \frac{\mathrm{d}\omega}{2\pi} \sum_{{\mathbf{q}}\in \mathbf{BZ} }\sum_{{\mathbf{G}}} \left\{(\mathrm{ln}[1-\chi^{KS}({\mathbf{q}},\mathrm{i}\omega)V({\mathbf{q}})])_{{\mathbf{G,G}}} +V_{{\mathbf{G,G}}}({\mathbf{q}})\chi^{KS}({\mathbf{q}},{\mathrm{i}}\omega) \right\} </math>.

	The sum over reciprocal lattice vectors has to be truncated at some <math>\mathbf{G}_{\mathrm{max}}</math>, determined by <math>\frac{\hbar^2\|{\mathbf{G}}+{\mathbf{q}}\|^2}{2\mathrm{m}_e}</math> < {{TAG\|ENCUTGW}}, which can be set in the {{TAG\|INCAR}} file. The default value is <math>\frac{2}{3}\times</math> {{TAG\|ENCUT}}, which experience has taught us not to change. For systematic convergence tests, instead increase {{TAG\|ENCUT}} and repeat steps 1 to 4, but be aware that the "maximum number of plane-waves" changes when {{TAG\|ENCUT}} is increased. Note that it is virtually impossible, to converge absolute correlation energies. Rather concentrate on relative energies (e.g. energy differences between two solids, or between a solid and the constituent atoms).		The sum over reciprocal lattice vectors has to be truncated at some <math>\mathbf{G}_{\mathrm{max}}</math>, determined by <math>\frac{\hbar^2\|{\mathbf{G}}+{\mathbf{q}}\|^2}{2\mathrm{m}_e}</math> < {{TAG\|ENCUTGW}}, which can be set in the {{TAG\|INCAR}} file. The default value is <math>\frac{2}{3}\times</math> {{TAG\|ENCUT}}, which experience has taught us not to change. For systematic convergence tests, instead increase {{TAG\|ENCUT}} and repeat steps 1 to 4, but be aware that the "maximum number of plane-waves" changes when {{TAG\|ENCUT}} is increased. Note that it is virtually impossible, to converge absolute correlation energies. Rather concentrate on relative energies (e.g. energy differences between two solids, or between a solid and the constituent atoms).

Kaltakm: Created page with "To reach technical convergence, a number of flags are available to control the evaluation of the ACFDT-RPA correlation energy in the fourth step. The expression for the ACFDT-..."

2019-07-29T16:09:10Z

Created page with "To reach technical convergence, a number of flags are available to control the evaluation of the ACFDT-RPA correlation energy in the fourth step. The expression for the ACFDT-..."

New page

To reach technical convergence, a number of flags are available to control the evaluation of the ACFDT-RPA correlation energy in the fourth step. The expression for the ACFDT-RPA correlation energy reads:

<math>\mathrm{E}_{\mathrm{c}}=\int_{0}^{\infty} \frac{\mathrm{d}\omega}{2\pi} \sum_{{\mathbf{q}}\in \mathbf{BZ} }\sum_{{\mathbf{G}}} \left\{(\mathrm{ln}[1-\chi^{KS}({\mathbf{q}},\mathrm{i}\omega)V({\mathbf{q}})])_{{\mathbf{G,G}}} +V_{{\mathbf{G,G}}}({\mathbf{q}})\chi^{KS}({\mathbf{q}},{\mathrm{i}}\omega) \right\} </math>.

The sum over reciprocal lattice vectors has to be truncated at some <math>\mathbf{G}_{\mathrm{max}}</math>, determined by <math>\frac{\hbar^2|{\mathbf{G}}+{\mathbf{q}}|^2}{2\mathrm{m}_e}</math> < {{TAG|ENCUTGW}}, which can be set in the {{TAG|INCAR}} file. The default value is <math>\frac{2}{3}\times</math> {{TAG|ENCUT}}, which experience has taught us not to change. For systematic convergence tests, instead increase {{TAG|ENCUT}} and repeat steps 1 to 4, but be aware that the "maximum number of plane-waves" changes when {{TAG|ENCUT}} is increased. Note that it is virtually impossible, to converge absolute correlation energies. Rather concentrate on relative energies (e.g. energy differences between two solids, or between a solid and the constituent atoms).

Since correlation energies converge very slowly with respect to <math>\mathbf{G}_{\mathrm max }</math>, VASP automatically extrapolates to the infinite basis set limit using a linear regression to the equation: {{cite|harl:2008}}{{cite|harl:2010}}{{cite|klimes:2014}}

<math>E_{\mathrm{c}}({\mathbf{G}})=E_{\mathrm{c}}(\infty)+\frac{A}{{\mathbf{G}}^3}</math>.

Furthermore, the Coulomb kernel is smoothly truncated between {{TAG|ENCUTGWSOFT}} and {{TAG|ENCUTGW}} using a simple cosine like window function (Hann window function). The default for {{TAG|ENCUTGWSOFT}} is 0.8<math>\times</math>{{TAG|ENCUTGW}} (again we do not recommend to change this default).

The integral over <math>\omega</math> is evaluated by means of a highly accurate minimax integration.{{cite|kaltak:2014}} The number of <math>\omega</math> points is determined by the flag {{TAG|NOMEGA}}, whereas the energy range of transitions is determined by the band gap and the energy difference between the lowest occupied and highest unoccupied one-electron orbital. VASP determines these values automatically (from vasp.5.4.1 on), and the user should only carefully converge with respect to the number of frequency points {{TAG|NOMEGA}}. A good choice is usually {{TAG|NOMEGA}}=12, however, for large gap systems one might obtain <math>\mu</math>eV convergence per atom already using 8 points, whereas for metals up to {{TAG|NOMEGA}}=24 frequency points are sometimes necessary, in particular, for large unit cells.

Strictly adhere to the steps outlines above. Specifically, be aware that steps two and three require the {{TAG|WAVECAR}} file generated in step one, whereas step four requires the {{TAG|WAVECAR}} and {{TAG|WAVEDER}} file generated in step three (generated by setting {{TAG|LOPTICS}}=''.TRUE.'').
<noinclude>
== References ==
<references/>
</noinclude>