Vacuum reference - Revision history

Csheldon: /* Step 2: Selecting structures at random */

2026-09-18T13:51:44Z

Step 2: Selecting structures at random

← Older revision		Revision as of 13:51, 18 September 2026
Line 919:		Line 919:
	path1 = "$PATH_TO_MD_DIRECTORIES/"		path1 = "$PATH_TO_MD_DIRECTORIES/"
	path2 = "$PATH_TO_CORELEVEL_DIRECTORIES/"		path2 = "$PATH_TO_CORELEVEL_DIRECTORIES/"
	~~systems = ["128H2O_slab", "Fe2P_64H2O", "Fe3P_64H2O"]~~

	def random_md_structures(path1, path2, systems, n, cutoff):		def random_md_structures(path1, path2, systems, n, cutoff):

Csheldon: Select 200 structures for the redox systems and 400 for the slab, matching the results quoted later

2026-09-18T13:01:21Z

Select 200 structures for the redox systems and 400 for the slab, matching the results quoted later

← Older revision		Revision as of 13:01, 18 September 2026
Line 910:		Line 910:
	From each of the MD calculations in Step 1, select 100-200 random structures (for the slab, use about 400), i.e., {{FILE\|POSCAR}} files, excluding the first 20000 steps to allow for time to equilibrate. Save these structures to three directories (one for each system), i.e., <code>Fe3P_64H2O</code>, <code>Fe2P_64H2O</code>, and <code>128H2O_slab</code>.		From each of the MD calculations in Step 1, select 100-200 random structures (for the slab, use about 400), i.e., {{FILE\|POSCAR}} files, excluding the first 20000 steps to allow for time to equilibrate. Save these structures to three directories (one for each system), i.e., <code>Fe3P_64H2O</code>, <code>Fe2P_64H2O</code>, and <code>128H2O_slab</code>.

	Using the following Python script, you can select 200 random structures ~~from~~ each ~~directory~~, omitting the first 20000 MD steps to allow equilibration:		Using the following Python script, you can select 200 random structures for each redox system and 400 for the slab, omitting the first 20000 MD steps to allow equilibration:

	<syntaxhighlight lang="python">		<syntaxhighlight lang="python">
Line 933:		Line 933:
	write(path2 + system + "/POSCAR." + str(i+1), atoms_list[j], format="vasp")		write(path2 + system + "/POSCAR." + str(i+1), atoms_list[j], format="vasp")

	random_md_structures(path1, path2, ~~systems~~, 200, 20000)		random_md_structures(path1, path2, ["Fe2P_64H2O", "Fe3P_64H2O"], 200, 20000)
			random_md_structures(path1, path2, ["128H2O_slab"], 400, 20000)
	</syntaxhighlight>		</syntaxhighlight>

	which writes ~~200 files named~~ <code>POSCAR.1</code> ~~to <code>POSCAR.200</code>~~ in each directory. Converging the local potential of the slab needs about 3000 structures {{Cite\|jinnouchi:ncm:2024}}; 400 are used here, which is enough to reach the literature value to within 0.1 eV.		which writes <code>POSCAR.1</code> onwards in each directory, 200 for the redox systems and 400 for the slab. Converging the local potential of the slab needs about 3000 structures {{Cite\|jinnouchi:ncm:2024}}; 400 are used here, which is enough to reach the literature value to within 0.1 eV.

	=== Step 3: Calculating the O<sub>1s</sub> energies and the local potential with a GGA ===		=== Step 3: Calculating the O<sub>1s</sub> energies and the local potential with a GGA ===

Csheldon: Attribute the 3000-structure figure to the paper and state the 400 used here beside the selection script

2026-09-18T12:54:06Z

Attribute the 3000-structure figure to the paper and state the 400 used here beside the selection script

← Older revision		Revision as of 12:54, 18 September 2026
Line 6:		Line 6:
	[[File:Redox_structures.png\|800px\|thumb\|center\|'''Figure 1'''. Top left: Fe<sup>3+</sup> in 64 H<sub>2</sub>O (<code>Fe3P_64H2O</code>), top center: Fe<sup>2+</sup> in 64 H<sub>2</sub>O (<code>Fe2P_64H2O</code>), top right: bulk H<sub>2</sub>O with 64 molecules (<code>64H2O_bulk</code>), and bottom: H<sub>2</sub>O slab with 128 molecules (<code>128H2O_slab</code>).]]		[[File:Redox_structures.png\|800px\|thumb\|center\|'''Figure 1'''. Top left: Fe<sup>3+</sup> in 64 H<sub>2</sub>O (<code>Fe3P_64H2O</code>), top center: Fe<sup>2+</sup> in 64 H<sub>2</sub>O (<code>Fe2P_64H2O</code>), top right: bulk H<sub>2</sub>O with 64 molecules (<code>64H2O_bulk</code>), and bottom: H<sub>2</sub>O slab with 128 molecules (<code>128H2O_slab</code>).]]

	The <code>128H2O_slab</code> calculations are more expensive, since so many structures ~~(~3000)~~ are required to ~~achieve convergence~~. Only the H<sub>2</sub>O molecules' O<sub>1s</sub> levels in the center of the slab are used, shaded in gray in Fig. 1, representing the ''bulk''. This only needs to be done once to determine the chemical potential relative to vacuum, and can then be used for any half-cell reaction within <code>64H2O_bulk</code> and with the same method. In contrast, repeat the <code>Fe2P/Fe3P_64H2O</code> calculations for each reaction. The {{FILE\|POSCAR}} files are described on more detail in the [[Thermodynamic integration: redox potential\|redox potential overview page]].		The <code>128H2O_slab</code> calculations are more expensive, since many structures are required to converge the local potential; Ref. {{Cite\|jinnouchi:ncm:2024}} uses ~3000. Only the H<sub>2</sub>O molecules' O<sub>1s</sub> levels in the center of the slab are used, shaded in gray in Fig. 1, representing the ''bulk''. This only needs to be done once to determine the chemical potential relative to vacuum, and can then be used for any half-cell reaction within <code>64H2O_bulk</code> and with the same method. In contrast, repeat the <code>Fe2P/Fe3P_64H2O</code> calculations for each reaction. The {{FILE\|POSCAR}} files are described on more detail in the [[Thermodynamic integration: redox potential\|redox potential overview page]].
	=== {{FILE\|POSCAR}}s ===		=== {{FILE\|POSCAR}}s ===

Line 908:		Line 908:
	=== Step 2: Selecting structures at random ===		=== Step 2: Selecting structures at random ===

	From each of the MD calculations in Step 1, select 100-200 random structures (for the slab, use about 400), i.e., {{FILE\|POSCAR}} files, excluding the first 20000 steps to allow for time to equilibrate. Save these structures to three directories (one for each system), i.e., <code>Fe3P_64H2O</code>, <code>Fe2P_64H2O</code>, and <code>128H2O_slab</code>~~. This example uses 200 structures for the redox systems and 400 for the slab, which reproduces the literature to within 0.1 eV; about 3000 are needed for a fully converged local potential~~.		From each of the MD calculations in Step 1, select 100-200 random structures (for the slab, use about 400), i.e., {{FILE\|POSCAR}} files, excluding the first 20000 steps to allow for time to equilibrate. Save these structures to three directories (one for each system), i.e., <code>Fe3P_64H2O</code>, <code>Fe2P_64H2O</code>, and <code>128H2O_slab</code>.

	Using the following Python script, you can select 200 random structures from each directory, omitting the first 20000 MD steps to allow equilibration:		Using the following Python script, you can select 200 random structures from each directory, omitting the first 20000 MD steps to allow equilibration:
Line 936:		Line 936:
	</syntaxhighlight>		</syntaxhighlight>

	which writes 200 files named <code>POSCAR.1</code> to <code>POSCAR.200</code>.		which writes 200 files named <code>POSCAR.1</code> to <code>POSCAR.200</code> in each directory. Converging the local potential of the slab needs about 3000 structures {{Cite\|jinnouchi:ncm:2024}}; 400 are used here, which is enough to reach the literature value to within 0.1 eV.

	=== Step 3: Calculating the O<sub>1s</sub> energies and the local potential with a GGA ===		=== Step 3: Calculating the O<sub>1s</sub> energies and the local potential with a GGA ===

Csheldon: Give the slab potential its own symbol; fix the script usage strings, units, structure counts and file naming

2026-09-18T12:47:46Z

Give the slab potential its own symbol; fix the script usage strings, units, structure counts and file naming

← Older revision		Revision as of 12:47, 18 September 2026
Line 908:		Line 908:
	=== Step 2: Selecting structures at random ===		=== Step 2: Selecting structures at random ===

	From each of the MD calculations in Step 1, select 100-200 random structures (for the slab, use about ~~3000~~), i.e., {{FILE\|POSCAR}} files, excluding the first 20000 steps to allow for time to equilibrate. Save these structures to three directories (one for each system), i.e., <code>Fe3P_64H2O</code>, <code>Fe2P_64H2O</code>, and <code>128H2O_slab</code>.		From each of the MD calculations in Step 1, select 100-200 random structures (for the slab, use about 400), i.e., {{FILE\|POSCAR}} files, excluding the first 20000 steps to allow for time to equilibrate. Save these structures to three directories (one for each system), i.e., <code>Fe3P_64H2O</code>, <code>Fe2P_64H2O</code>, and <code>128H2O_slab</code>. This example uses 200 structures for the redox systems and 400 for the slab, which reproduces the literature to within 0.1 eV; about 3000 are needed for a fully converged local potential.

	Using the following Python script, you can select 200 random structures from each directory, omitting the first 20000 MD steps to allow equilibration:		Using the following Python script, you can select 200 random structures from each directory, omitting the first 20000 MD steps to allow equilibration:
Line 936:		Line 936:
	</syntaxhighlight>		</syntaxhighlight>

	which ~~generates~~ 200 ~~{{FILE\|~~POSCAR}}.~~m files~~.		which writes 200 files named <code>POSCAR.1</code> to <code>POSCAR.200</code>.

	=== Step 3: Calculating the O<sub>1s</sub> energies and the local potential with a GGA ===		=== Step 3: Calculating the O<sub>1s</sub> energies and the local potential with a GGA ===

	For each of the structures that you have generated, perform a first-principles (FP) calculation with <code>{{TAG\|ICORELEVEL}} = 1</code>, {{TAG\|LVHAR\|.TRUE.}}, and {{TAG\|WRT_POTENTIAL\|hartree ionic}}. The energies of the O<sub>1s</sub> levels, and the local potential will be written, which can later be used to find the vacuum reference. An example {{FILE\|INCAR}} for RPBE+D3 is provided below:		For each of the structures that you have generated, perform a first-principles (FP) calculation with <code>{{TAG\|ICORELEVEL}} = 1</code>, {{TAG\|LVHAR\|.TRUE.}}, and {{TAG\|WRT_POTENTIAL\|hartree ionic}}. The energies of the O<sub>1s</sub> levels, and the local potential will be written, which can later be used to find the vacuum reference. An example {{FILE\|INCAR}} for RPBE+D3 is provided below: {{NB\|mind\|{{TAG\|WRT_POTENTIAL}} is available from VASP 6.4.3. For older versions use {{TAG\|LVHAR\|.TRUE.}} and read the potential from {{FILE\|LOCPOT}}.}}

	{{TAGBL\|ENCUT}} = 520.0		{{TAGBL\|ENCUT}} = 520.0
Line 973:		Line 973:
	{{TAGBL\|MAGMOM}} = 1280 640 1*5		{{TAGBL\|MAGMOM}} = 1280 640 1*5

	An example bash script is given below for executing the calculations. ~~Make sure to use~~ the ~~supporting~~ Python ~~scripts (~~<code>O1s_level.py</code> and <code>vacuum_ref.py</code>).		An example bash script is given below for executing the calculations. Save the two Python blocks below as <code>O1s_level.py</code> and <code>vacuum_ref.py</code> in the directory you run from.
	{{NB\|important\|For the slab, only the central 10 Å is used for calculating the average local potential and the average O<sub>1s</sub> levels. For the redox systems (<code>Fe3P_64H2O</code> and <code>Fe2P_64H2O</code>), only the O<sub>1s</sub> levels outside of a radius of half the cell parameter are included.}}		{{NB\|important\|For the slab, only the central 10 Å is used for calculating the average local potential and the average O<sub>1s</sub> levels. For the redox systems (<code>Fe3P_64H2O</code> and <code>Fe2P_64H2O</code>), only the O<sub>1s</sub> levels outside of a radius of half the cell parameter are included.}}

Line 1,065:		Line 1,065:
	return(np.average(o1s_all))		return(np.average(o1s_all))

	# Expecting: python3 ~~vacuum_ref~~.py z_mid z_diff system		# Expecting: python3 O1s_level.py z_mid z_diff system element
	if len(sys.argv) != 5:		if len(sys.argv) != 5:
	print("Usage: python3 ~~vacuum_ref~~.py z_mid z_diff system element")		print("Usage: python3 O1s_level.py z_mid z_diff system element")
	sys.exit(1)		sys.exit(1)

Line 1,355:		Line 1,355:
	</syntaxhighlight>		</syntaxhighlight>
	</div>		</div>
	{{NB\|important\|~~Make sure to first download and save the <code>O1s_level.py</code> and <code>vacuum_ref.py</code> scripts.~~ Calculating these values while running the jobs will save a non-trivial amount of disk space. Alternatively, you can save all of the {{FILE\|OUTCAR}} and {{FILE\|vaspout.h5}} files to process later.}}		{{NB\|important\|Calculating these values while running the jobs will save a non-trivial amount of disk space. Alternatively, you can save all of the {{FILE\|OUTCAR}} and {{FILE\|vaspout.h5}} files to process later.}}

	These scripts will create a <code>vac_ref.dat</code> file containing the number, the average 1s energy levels for O in water molecules, and the average local potential for each structure:		These scripts will create a <code>vac_ref.dat</code> file containing the number, the average 1s energy levels for O in water molecules, and the average local potential for each structure:
Line 1,413:		Line 1,413:

	=== Step 5: Referencing the O<sub>1s</sub> level to the vacuum ===		=== Step 5: Referencing the O<sub>1s</sub> level to the vacuum ===
	The ~~value~~ of the ~~local potential~~ <math>\~~mu_~~{\mathrm{~~vac~~}}</math> ''must'' be included to obtain an absolute redox potential; otherwise, comparing between different cells (as in this example) or between different systems ''cannot'' be done as no common reference exists. It is important to include this because the local potential means that the vacuum reference is not equal to 0 eV, it is more negative ~-1.8 eV:		The averaged local potential of the slab, <math>\bar{\phi}_{\mathrm{slab}}</math>, ''must'' be included to obtain an absolute redox potential; otherwise, comparing between different cells (as in this example) or between different systems ''cannot'' be done as no common reference exists. It is important to include this because the local potential means that the vacuum reference is not equal to 0 eV, it is more negative ~-1.8 eV:

	[[File:Local_potential.png\|600px\|thumb\|center\|'''Figure 2'''. The local potential for the water slab (<code>128H2O_slab</code>). Increasing from 100 to 400 structures partially smooths out the potential, but much larger numbers of structures are required to smooth it out completely (cf. the right-hand side of Fig. 1b in {{Cite\|jinnouchi:ncm:2024}}). You can see that the local potential is ~-1.8 eV. This is the missing term from the previous section.]]		[[File:Local_potential.png\|600px\|thumb\|center\|'''Figure 2'''. The local potential for the water slab (<code>128H2O_slab</code>). Increasing from 100 to 400 structures partially smooths out the potential, but much larger numbers of structures are required to smooth it out completely (cf. the right-hand side of Fig. 1b in {{Cite\|jinnouchi:ncm:2024}}). You can see that the local potential is ~-1.8 eV. This is the missing term from the previous section.]]

	You can calculate the local potential using {{TAG\|LVHAR\|.TRUE.}}. This saves the potential to {{FILE\|LOCPOT}}, from which you could perform all of the following steps to calculate the average. Instead, {{TAG\|WRT_POTENTIAL\|hartree ionic}} saves the local potential to {{FILE\|vaspout.h5}} ~~and analyze using~~ {{py4vasp}}. This gives		You can calculate the local potential using {{TAG\|LVHAR\|.TRUE.}}. This saves the potential to {{FILE\|LOCPOT}}, from which you could perform all of the following steps to calculate the average. Instead, {{TAG\|WRT_POTENTIAL\|hartree ionic}} saves the local potential to {{FILE\|vaspout.h5}}, which can be analyzed with {{py4vasp}}. This gives

	{\|cellpadding="4" cellspacing="0" border="1"		{\|cellpadding="4" cellspacing="0" border="1"
Line 1,430:		Line 1,430:
	\| 400		\| 400
	\|-		\|-
	\| <math>-\langle \~~mu_~~{\mathrm{~~vac~~}} \rangle</math>, 128H<sub>2</sub>O slab		\| <math>\langle \bar{\phi}_{\mathrm{slab}} \rangle</math>, 128H<sub>2</sub>O slab
	\| -1.686		\| -1.686
	\| -1.735		\| -1.735
Line 1,454:		Line 1,454:
	\| Std POTCAR		\| Std POTCAR
	\|-		\|-
	\| <math>~~-\mu_{\mathrm{vac}} +~~ \langle \epsilon_{1s,slab} \rangle </math>, 128H<sub>2</sub>O slab		\| <math>\langle \epsilon_{1s,slab} \rangle + \bar{\phi}_{\mathrm{slab}}</math>, 128H<sub>2</sub>O slab
	\| -511.12		\| -511.12
	\| -511.20		\| -511.20
Line 1,464:		Line 1,464:

	=== Step 6: Comparing the local potential in the slab and the bulk ===		=== Step 6: Comparing the local potential in the slab and the bulk ===
	The ~~redox level~~ <math>e \Delta \bar{\phi}</math> (in V) can now be obtained ~~(note that e=1 for~~ this half-cell reaction):		The vacuum alignment <math>e \Delta \bar{\phi}</math> (in eV) can now be obtained; one electron is transferred in this half-cell reaction, so <math>n=1</math>:

	<math>e \Delta \bar{\phi} = \frac{[\langle \epsilon_{1s,Fe^{3+}} \rangle + \langle \epsilon_{1s,Fe^{2+}} \rangle]}{2} - \langle \epsilon_{1s,slab} \rangle + \~~mu_~~{\mathrm{~~vac~~}}</math>		<math>e \Delta \bar{\phi} = \frac{[\langle \epsilon_{1s,Fe^{3+}} \rangle + \langle \epsilon_{1s,Fe^{2+}} \rangle]}{2} - \left( \langle \epsilon_{1s,slab} \rangle + \bar{\phi}_{\mathrm{slab}} \right)</math>

	<math>= \frac{[(-507.45) + (-507.50)]}{2} - (-509.45 + -1.73) = 3.71</math> (this example)		<math>= \frac{[(-507.45) + (-507.50)]}{2} - (-509.45 - 1.73) = 3.71</math> eV (this example)

	<math>e \Delta \bar{\phi}_{\mathrm{Lit.}} = \frac{[(-507.43) + (-507.50)]}{2} - (-511.23) = 3.74</math> (for Std. {{FILE\|POTCAR}} provided by R. Jinnouchi; GW {{FILE\|POTCAR}}s in lit. {{Cite\|jinnouchi:ncm:2024}})		<math>e \Delta \bar{\phi}_{\mathrm{Lit.}} = \frac{[(-507.43) + (-507.50)]}{2} - (-511.23) = 3.74</math> (for Std. {{FILE\|POTCAR}} provided by R. Jinnouchi; GW {{FILE\|POTCAR}}s in lit. {{Cite\|jinnouchi:ncm:2024}})

Csheldon: Point at the published title; hold the unpublished density-functionals link; make categories live

2026-09-18T08:07:32Z

Point at the published title; hold the unpublished density-functionals link; make categories live

← Older revision		Revision as of 08:07, 18 September 2026
Line 1,486:		Line 1,486:
	* [[Thermodynamic integration: redox potential]]		* [[Thermodynamic integration: redox potential]]
	* [[Thermodynamic integration between machine-learned force fields]]		* [[Thermodynamic integration between machine-learned force fields]]
	* [[~~Construction:~~Thermodynamic integration between a machine-learned force field and a density functional\|Thermodynamic integration between a machine-learned force field and a density functional]]		* [[Thermodynamic integration between a machine-learned force field and a density functional\|Thermodynamic integration between a machine-learned force field and a density functional]]
	;Theory		;Theory
	* [[Thermodynamic integration]]		* [[Thermodynamic integration]]

Csheldon: Point the links at the published titles; categories made live

2026-09-18T08:03:53Z

Point the links at the published titles; categories made live

← Older revision		Revision as of 08:03, 18 September 2026
Line 1:		Line 1:
	A vacuum reference for the Fe<sup>2+</sup>/Fe<sup>3+</sup> half-cell is obtained from a water slab, by referencing its local potential to the O 1s states of the oxygen atoms in its water molecules. In each of the solvated ion cells, the O 1s level of the water molecules is taken as the same reference, so the energies of the solvated cells can be placed on a common vacuum scale {{Cite\|jinnouchi:ncm:2024}}. The derivation and the approximations behind it are given in [[~~Construction:~~Vacuum reference: theory\|Vacuum reference: theory]].		A vacuum reference for the Fe<sup>2+</sup>/Fe<sup>3+</sup> half-cell is obtained from a water slab, by referencing its local potential to the O 1s states of the oxygen atoms in its water molecules. In each of the solvated ion cells, the O 1s level of the water molecules is taken as the same reference, so the energies of the solvated cells can be placed on a common vacuum scale {{Cite\|jinnouchi:ncm:2024}}. The derivation and the approximations behind it are given in [[Vacuum reference: theory\|Vacuum reference: theory]].
	{{NB\|mind\|Although the absolute reference is taken here for the electrochemical potential, it is also possible to do this for other examples, such as for electron emission or catalytic processes (swap out the redox level for the valence band maximum - VBM).}}		{{NB\|mind\|Although the absolute reference is taken here for the electrochemical potential, it is also possible to do this for other examples, such as for electron emission or catalytic processes (swap out the redox level for the valence band maximum - VBM).}}

Line 6:		Line 6:
	[[File:Redox_structures.png\|800px\|thumb\|center\|'''Figure 1'''. Top left: Fe<sup>3+</sup> in 64 H<sub>2</sub>O (<code>Fe3P_64H2O</code>), top center: Fe<sup>2+</sup> in 64 H<sub>2</sub>O (<code>Fe2P_64H2O</code>), top right: bulk H<sub>2</sub>O with 64 molecules (<code>64H2O_bulk</code>), and bottom: H<sub>2</sub>O slab with 128 molecules (<code>128H2O_slab</code>).]]		[[File:Redox_structures.png\|800px\|thumb\|center\|'''Figure 1'''. Top left: Fe<sup>3+</sup> in 64 H<sub>2</sub>O (<code>Fe3P_64H2O</code>), top center: Fe<sup>2+</sup> in 64 H<sub>2</sub>O (<code>Fe2P_64H2O</code>), top right: bulk H<sub>2</sub>O with 64 molecules (<code>64H2O_bulk</code>), and bottom: H<sub>2</sub>O slab with 128 molecules (<code>128H2O_slab</code>).]]

	The <code>128H2O_slab</code> calculations are more expensive, since so many structures (~3000) are required to achieve convergence. Only the H<sub>2</sub>O molecules' O<sub>1s</sub> levels in the center of the slab are used, shaded in gray in Fig. 1, representing the ''bulk''. This only needs to be done once to determine the chemical potential relative to vacuum, and can then be used for any half-cell reaction within <code>64H2O_bulk</code> and with the same method. In contrast, repeat the <code>Fe2P/Fe3P_64H2O</code> calculations for each reaction. The {{FILE\|POSCAR}} files are described on more detail in the [[~~Construction:~~Thermodynamic integration: redox potential\|redox potential overview page]].		The <code>128H2O_slab</code> calculations are more expensive, since so many structures (~3000) are required to achieve convergence. Only the H<sub>2</sub>O molecules' O<sub>1s</sub> levels in the center of the slab are used, shaded in gray in Fig. 1, representing the ''bulk''. This only needs to be done once to determine the chemical potential relative to vacuum, and can then be used for any half-cell reaction within <code>64H2O_bulk</code> and with the same method. In contrast, repeat the <code>Fe2P/Fe3P_64H2O</code> calculations for each reaction. The {{FILE\|POSCAR}} files are described on more detail in the [[Thermodynamic integration: redox potential\|redox potential overview page]].
	=== {{FILE\|POSCAR}}s ===		=== {{FILE\|POSCAR}}s ===

Line 1,484:		Line 1,484:
	== Related tags and articles ==		== Related tags and articles ==
	;How-tos		;How-tos
	* [[~~Construction:~~Thermodynamic integration: redox potential]]		* [[Thermodynamic integration: redox potential]]
	* [[~~Construction:~~Thermodynamic integration between ~~MLFFs~~]]		* [[Thermodynamic integration between machine-learned force fields]]
	* [[Construction:Thermodynamic integration between ~~MLFF~~ and ~~GGAs~~]]		* [[Construction:Thermodynamic integration between a machine-learned force field and a density functional\|Thermodynamic integration between a machine-learned force field and a density functional]]
	;Theory		;Theory
	* [[Thermodynamic integration]]		* [[Thermodynamic integration]]
Line 1,497:		Line 1,497:
	<references/>		<references/>

	~~<!--~~[[Category:Howto]][[Category:Advanced molecular-dynamics sampling]][[Category:Electrostatics]][[Category:Potential]][[Category:Electrochemistry]]-->		[[Category:Howto]][[Category:Advanced molecular-dynamics sampling]][[Category:Electrostatics]][[Category:Potential]]
			<!--[[Category:Electrochemistry]]-->

Csheldon: Csheldon moved page Construction:Vacuum reference to Vacuum reference

2026-09-18T08:00:25Z

Csheldon moved page Construction:Vacuum reference to Vacuum reference

← Older revision	Revision as of 08:00, 18 September 2026
(No difference)

Csheldon: Use the friendly imperative where the sentence tells the reader to do something

2026-09-17T16:06:43Z

Use the friendly imperative where the sentence tells the reader to do something

Csheldon: Upright multi-letter math subscripts; write the Gamma point in prose; tidy Latin abbreviations

2026-09-17T16:03:16Z

Upright multi-letter math subscripts; write the Gamma point in prose; tidy Latin abbreviations

← Older revision		Revision as of 16:03, 17 September 2026
Line 3:		Line 3:

	== Input files ==		== Input files ==
	The vacuum alignment is obtained from calculations on three different systems: Fe<sup>3+</sup> in 64 H<sub>2</sub>O (<code>Fe3P_64H2O</code>), Fe<sup>2+</sup> in 64 H<sub>2</sub>O (<code>Fe2P_64H2O</code>), H<sub>2</sub>O slab with 128 molecules (<code>128H2O_slab</code>)~~. I.e.~~, <math>[\mathrm{Fe}(\mathrm{H}_2\mathrm{O})_n]^{3+}</math>, <math>[\mathrm{Fe}(\mathrm{H}_2\mathrm{O})_n]^{2+}</math>, and the water slab.		The vacuum alignment is obtained from calculations on three different systems: Fe<sup>3+</sup> in 64 H<sub>2</sub>O (<code>Fe3P_64H2O</code>), Fe<sup>2+</sup> in 64 H<sub>2</sub>O (<code>Fe2P_64H2O</code>), H<sub>2</sub>O slab with 128 molecules (<code>128H2O_slab</code>), that is, <math>[\mathrm{Fe}(\mathrm{H}_2\mathrm{O})_n]^{3+}</math>, <math>[\mathrm{Fe}(\mathrm{H}_2\mathrm{O})_n]^{2+}</math>, and the water slab.
	[[File:Redox_structures.png\|800px\|thumb\|center\|'''Figure 1'''. Top left: Fe<sup>3+</sup> in 64 H<sub>2</sub>O (<code>Fe3P_64H2O</code>), top center: Fe<sup>2+</sup> in 64 H<sub>2</sub>O (<code>Fe2P_64H2O</code>), top right: bulk H<sub>2</sub>O with 64 molecules (<code>64H2O_bulk</code>), and bottom: H<sub>2</sub>O slab with 128 molecules (<code>128H2O_slab</code>).]]		[[File:Redox_structures.png\|800px\|thumb\|center\|'''Figure 1'''. Top left: Fe<sup>3+</sup> in 64 H<sub>2</sub>O (<code>Fe3P_64H2O</code>), top center: Fe<sup>2+</sup> in 64 H<sub>2</sub>O (<code>Fe2P_64H2O</code>), top right: bulk H<sub>2</sub>O with 64 molecules (<code>64H2O_bulk</code>), and bottom: H<sub>2</sub>O slab with 128 molecules (<code>128H2O_slab</code>).]]

Line 867:		Line 867:

	=== {{FILE\|KPOINTS}} ===		=== {{FILE\|KPOINTS}} ===
	~~The~~ Gamma-point ~~only~~ is used ~~for~~ the {{FILE\|KPOINTS}} file:		Only the Γ point is used, so the {{FILE\|KPOINTS}} file is:

	Gamma-point only		Gamma-point only
Line 1,470:		Line 1,470:
	<math>= \frac{[(-507.45) + (-507.50)]}{2} - (-509.45 + -1.73) = 3.71</math> (this example)		<math>= \frac{[(-507.45) + (-507.50)]}{2} - (-509.45 + -1.73) = 3.71</math> (this example)

	<math>e \Delta \bar{\phi}_{Lit.} = \frac{[(-507.43) + (-507.50)]}{2} - (-511.23) = 3.74</math> (for Std. {{FILE\|POTCAR}} provided by R. Jinnouchi; GW {{FILE\|POTCAR}}s in lit. {{Cite\|jinnouchi:ncm:2024}})		<math>e \Delta \bar{\phi}_{\mathrm{Lit.}} = \frac{[(-507.43) + (-507.50)]}{2} - (-511.23) = 3.74</math> (for Std. {{FILE\|POTCAR}} provided by R. Jinnouchi; GW {{FILE\|POTCAR}}s in lit. {{Cite\|jinnouchi:ncm:2024}})

	These values agree to within 0.1 eV of the literature, having used only 400 structures for the slab, rather than the 3000 in the paper. Having obtained <math>e \Delta \bar{\phi}</math>, you now have the reference to the vacuum, necessary for calculating the absolute redox potentials.		These values agree to within 0.1 eV of the literature, having used only 400 structures for the slab, rather than the 3000 in the paper. Having obtained <math>e \Delta \bar{\phi}</math>, you now have the reference to the vacuum, necessary for calculating the absolute redox potentials.

Csheldon: Use the friendly imperative for reader actions and impersonal phrasing for reported results

2026-09-17T15:54:39Z

Use the friendly imperative for reader actions and impersonal phrasing for reported results

← Older revision		Revision as of 16:06, 17 September 2026
Line 6:		Line 6:
	[[File:Redox_structures.png\|800px\|thumb\|center\|'''Figure 1'''. Top left: Fe<sup>3+</sup> in 64 H<sub>2</sub>O (<code>Fe3P_64H2O</code>), top center: Fe<sup>2+</sup> in 64 H<sub>2</sub>O (<code>Fe2P_64H2O</code>), top right: bulk H<sub>2</sub>O with 64 molecules (<code>64H2O_bulk</code>), and bottom: H<sub>2</sub>O slab with 128 molecules (<code>128H2O_slab</code>).]]		[[File:Redox_structures.png\|800px\|thumb\|center\|'''Figure 1'''. Top left: Fe<sup>3+</sup> in 64 H<sub>2</sub>O (<code>Fe3P_64H2O</code>), top center: Fe<sup>2+</sup> in 64 H<sub>2</sub>O (<code>Fe2P_64H2O</code>), top right: bulk H<sub>2</sub>O with 64 molecules (<code>64H2O_bulk</code>), and bottom: H<sub>2</sub>O slab with 128 molecules (<code>128H2O_slab</code>).]]

	The <code>128H2O_slab</code> calculations are more expensive, since so many structures (~3000) are required to achieve convergence. Only the H<sub>2</sub>O molecules' O<sub>1s</sub> levels in the center of the slab are used, shaded in gray in Fig. 1, representing the ''bulk''. This only needs to be done once to determine the chemical potential relative to vacuum, and can then be used for any half-cell reaction within <code>64H2O_bulk</code> and with the same method. ~~The~~ <code>Fe2P/Fe3P_64H2O</code> calculations~~, on the other hand, need to be repeated~~ for each reaction. The {{FILE\|POSCAR}} files are described on more detail in the [[Construction:Thermodynamic integration: redox potential\|redox potential overview page]].		The <code>128H2O_slab</code> calculations are more expensive, since so many structures (~3000) are required to achieve convergence. Only the H<sub>2</sub>O molecules' O<sub>1s</sub> levels in the center of the slab are used, shaded in gray in Fig. 1, representing the ''bulk''. This only needs to be done once to determine the chemical potential relative to vacuum, and can then be used for any half-cell reaction within <code>64H2O_bulk</code> and with the same method. In contrast, repeat the <code>Fe2P/Fe3P_64H2O</code> calculations for each reaction. The {{FILE\|POSCAR}} files are described on more detail in the [[Construction:Thermodynamic integration: redox potential\|redox potential overview page]].
	=== {{FILE\|POSCAR}}s ===		=== {{FILE\|POSCAR}}s ===

Line 884:		Line 884:
	=== Step 1: Running the MD simulation with an MLFF ===		=== Step 1: Running the MD simulation with an MLFF ===

	Perform a [[:Category:Molecular dynamics\|molecular dynamics (MD)]] simulation for each system using a Langevin thermostat. ~~The~~ following {{FILE\|INCAR}} file ~~can be used~~:		Perform a [[:Category:Molecular dynamics\|molecular dynamics (MD)]] simulation for each system using a Langevin thermostat. Use the following {{FILE\|INCAR}} file:

	#Molecular dynamics		#Molecular dynamics
Line 908:		Line 908:
	=== Step 2: Selecting structures at random ===		=== Step 2: Selecting structures at random ===

	From each of the MD calculations in Step 1, select 100-200 random structures (for the slab, ~3000 ~~should be used~~), i.e., {{FILE\|POSCAR}} files, excluding the first 20000 steps to allow for time to equilibrate. Save these structures to three directories (one for each system), i.e., <code>Fe3P_64H2O</code>, <code>Fe2P_64H2O</code>, and <code>128H2O_slab</code>.		From each of the MD calculations in Step 1, select 100-200 random structures (for the slab, use about 3000), i.e., {{FILE\|POSCAR}} files, excluding the first 20000 steps to allow for time to equilibrate. Save these structures to three directories (one for each system), i.e., <code>Fe3P_64H2O</code>, <code>Fe2P_64H2O</code>, and <code>128H2O_slab</code>.

	Using the following Python script, you can select 200 random structures from each directory, omitting the first 20000 MD steps to allow equilibration:		Using the following Python script, you can select 200 random structures from each directory, omitting the first 20000 MD steps to allow equilibration:
Line 1,370:		Line 1,370:
	=== Step 4: Averaging the O<sub>1s</sub> energy levels ===		=== Step 4: Averaging the O<sub>1s</sub> energy levels ===

	Once the FP calculations have finished, extract the O<sub>1s</sub> energy levels <math>\epsilon_{1s}</math>. You can find them listed in {{FILE\|OUTCAR}} after <code>the core state eigenenergies</code>. For each structure, average these energy levels over all O, except for <code>128H2O_slab</code>, where only ~~O's~~ within the central 10 <math>\AA</math> of the cell ~~should be~~ used to mimic the bulk. With the average O<sub>1s</sub> for each individual structure, average over all structures in the ensemble, to get a single value for each system.		Once the FP calculations have finished, extract the O<sub>1s</sub> energy levels <math>\epsilon_{1s}</math>. You can find them listed in {{FILE\|OUTCAR}} after <code>the core state eigenenergies</code>. For each structure, average these energy levels over all O, except for <code>128H2O_slab</code>, where only the oxygen atoms within the central 10 <math>\AA</math> of the cell are used, to mimic the bulk. With the average O<sub>1s</sub> for each individual structure, average over all structures in the ensemble, to get a single value for each system.

	If you used the run script structure suggested above, then you will have a <code>vac_ref.dat</code> file. In the first column, there is the structure number, followed by the average O<sub>1s</sub> level for water molecules, then the average local potential <math>\mu</math>.		If you used the run script structure suggested above, then you will have a <code>vac_ref.dat</code> file. In the first column, there is the structure number, followed by the average O<sub>1s</sub> level for water molecules, then the average local potential <math>\mu</math>.
Line 1,437:		Line 1,437:
	\|}		\|}

	~~The~~ average local potential only ~~needs to be considered~~ relative to the vacuum, ~~i.e.~~, for the center of the water slab. The bulk and redox systems do not need to be corrected (cf. Fig. 2 below, and Fig. 1c and Eq. 7 of {{Cite\|jinnouchi:ncm:2024}}).		Consider the average local potential only relative to the vacuum, that is, for the center of the water slab. The bulk and redox systems do not need to be corrected (cf. Fig. 2 below, and Fig. 1c and Eq. 7 of {{Cite\|jinnouchi:ncm:2024}}).

	The updated O<sub>1s</sub> energy for the water slab is then much closer to the literature:		The updated O<sub>1s</sub> energy for the water slab is then much closer to the literature:

← Older revision		Revision as of 15:54, 17 September 2026
Line 1:		Line 1:
	A vacuum reference for the Fe<sup>2+</sup>/Fe<sup>3+</sup> half-cell is obtained from a water slab, by referencing its local potential to the O 1s states of the oxygen atoms in its water molecules. In each of the solvated ion cells, the O 1s level of the water molecules is taken as the same reference, so the energies of the solvated cells can be placed on a common vacuum scale {{Cite\|jinnouchi:ncm:2024}}. The derivation and the approximations behind it are given in [[Construction:Vacuum reference: theory\|Vacuum reference: theory]].		A vacuum reference for the Fe<sup>2+</sup>/Fe<sup>3+</sup> half-cell is obtained from a water slab, by referencing its local potential to the O 1s states of the oxygen atoms in its water molecules. In each of the solvated ion cells, the O 1s level of the water molecules is taken as the same reference, so the energies of the solvated cells can be placed on a common vacuum scale {{Cite\|jinnouchi:ncm:2024}}. The derivation and the approximations behind it are given in [[Construction:Vacuum reference: theory\|Vacuum reference: theory]].
	{{NB\|mind\|Although ~~we take~~ the absolute reference for the electrochemical potential ~~here~~, it is also possible to do this for other examples, such as for electron emission or catalytic processes (swap out the redox level for the valence band maximum - VBM).}}		{{NB\|mind\|Although the absolute reference is taken here for the electrochemical potential, it is also possible to do this for other examples, such as for electron emission or catalytic processes (swap out the redox level for the valence band maximum - VBM).}}

	== Input files ==		== Input files ==
Line 904:		Line 904:
	{{TAGBL\|ML_MODE}} = run		{{TAGBL\|ML_MODE}} = run

	~~In this case, we have performed~~ a 100000 fs MD simulation (100000 steps with a time step of 1 fs), using a [[Langevin thermostat]].		This example uses a 100000 fs MD simulation (100000 steps with a time step of 1 fs), using a [[Langevin thermostat]].

	=== Step 2: Selecting structures at random ===		=== Step 2: Selecting structures at random ===
Line 1,372:		Line 1,372:
	Once the FP calculations have finished, extract the O<sub>1s</sub> energy levels <math>\epsilon_{1s}</math>. You can find them listed in {{FILE\|OUTCAR}} after <code>the core state eigenenergies</code>. For each structure, average these energy levels over all O, except for <code>128H2O_slab</code>, where only O's within the central 10 <math>\AA</math> of the cell should be used to mimic the bulk. With the average O<sub>1s</sub> for each individual structure, average over all structures in the ensemble, to get a single value for each system.		Once the FP calculations have finished, extract the O<sub>1s</sub> energy levels <math>\epsilon_{1s}</math>. You can find them listed in {{FILE\|OUTCAR}} after <code>the core state eigenenergies</code>. For each structure, average these energy levels over all O, except for <code>128H2O_slab</code>, where only O's within the central 10 <math>\AA</math> of the cell should be used to mimic the bulk. With the average O<sub>1s</sub> for each individual structure, average over all structures in the ensemble, to get a single value for each system.

	If you used the run script structure ~~that we~~ suggested above, then you will have a <code>vac_ref.dat</code> file. In the first column, there is the structure number, followed by the average O<sub>1s</sub> level for water molecules, then the average local potential <math>\mu</math>.		If you used the run script structure suggested above, then you will have a <code>vac_ref.dat</code> file. In the first column, there is the structure number, followed by the average O<sub>1s</sub> level for water molecules, then the average local potential <math>\mu</math>.

	You have now obtained the O<sub>1s</sub> level. ~~Our~~ values for different numbers of structures compared to the literature values are compiled in the following table:		You have now obtained the O<sub>1s</sub> level. The values obtained for different numbers of structures, compared with the literature values, are compiled in the following table:

	{\|cellpadding="6" cellspacing="0" border="1"		{\|cellpadding="6" cellspacing="0" border="1"
Line 1,410:		Line 1,410:
	<sup>a</sup>The number of structures used in Jinnouchi et al. is 200 for bulk water and redox systems, and 3000 for the water slab, cf. Supplementary Table 4 for the averaged 1s levels of O atoms in Jinnouchi et al. for reference {{Cite\|jinnouchi:ncm:2024}}.		<sup>a</sup>The number of structures used in Jinnouchi et al. is 200 for bulk water and redox systems, and 3000 for the water slab, cf. Supplementary Table 4 for the averaged 1s levels of O atoms in Jinnouchi et al. for reference {{Cite\|jinnouchi:ncm:2024}}.

	There is a small difference between the literature for the bulk and redox systems using 100 structures, which is narrowed when using 200. The difference for the water slab is much larger, as the reference with respect to the vacuum has not been given, hence the missing ~1.8 eV. The average O<sub>1s</sub> level is converged for the water slab, which ~~tells us~~ that ~~it is~~ the local potential ~~that~~ requires many structures to converge.		There is a small difference between the literature for the bulk and redox systems using 100 structures, which is narrowed when using 200. The difference for the water slab is much larger, as the reference with respect to the vacuum has not been given, hence the missing ~1.8 eV. The average O<sub>1s</sub> level is converged for the water slab, which shows that the local potential requires many structures to converge.

	=== Step 5: Referencing the O<sub>1s</sub> level to the vacuum ===		=== Step 5: Referencing the O<sub>1s</sub> level to the vacuum ===
Line 1,417:		Line 1,417:
	[[File:Local_potential.png\|600px\|thumb\|center\|'''Figure 2'''. The local potential for the water slab (<code>128H2O_slab</code>). Increasing from 100 to 400 structures partially smooths out the potential, but much larger numbers of structures are required to smooth it out completely (cf. the right-hand side of Fig. 1b in {{Cite\|jinnouchi:ncm:2024}}). You can see that the local potential is ~-1.8 eV. This is the missing term from the previous section.]]		[[File:Local_potential.png\|600px\|thumb\|center\|'''Figure 2'''. The local potential for the water slab (<code>128H2O_slab</code>). Increasing from 100 to 400 structures partially smooths out the potential, but much larger numbers of structures are required to smooth it out completely (cf. the right-hand side of Fig. 1b in {{Cite\|jinnouchi:ncm:2024}}). You can see that the local potential is ~-1.8 eV. This is the missing term from the previous section.]]

	You can calculate the local potential using {{TAG\|LVHAR\|.TRUE.}}. This saves the potential to {{FILE\|LOCPOT}}, from which you could perform all of the following steps to calculate the average. ~~We instead used~~ {{TAG\|WRT_POTENTIAL\|hartree ionic}} ~~to save~~ the local potential to {{FILE\|vaspout.h5}} and analyze using {{py4vasp}}. This gives		You can calculate the local potential using {{TAG\|LVHAR\|.TRUE.}}. This saves the potential to {{FILE\|LOCPOT}}, from which you could perform all of the following steps to calculate the average. Instead, {{TAG\|WRT_POTENTIAL\|hartree ionic}} saves the local potential to {{FILE\|vaspout.h5}} and analyze using {{py4vasp}}. This gives

	{\|cellpadding="4" cellspacing="0" border="1"		{\|cellpadding="4" cellspacing="0" border="1"
Line 1,472:		Line 1,472:
	<math>e \Delta \bar{\phi}_{Lit.} = \frac{[(-507.43) + (-507.50)]}{2} - (-511.23) = 3.74</math> (for Std. {{FILE\|POTCAR}} provided by R. Jinnouchi; GW {{FILE\|POTCAR}}s in lit. {{Cite\|jinnouchi:ncm:2024}})		<math>e \Delta \bar{\phi}_{Lit.} = \frac{[(-507.43) + (-507.50)]}{2} - (-511.23) = 3.74</math> (for Std. {{FILE\|POTCAR}} provided by R. Jinnouchi; GW {{FILE\|POTCAR}}s in lit. {{Cite\|jinnouchi:ncm:2024}})

	We agree to within 0.1 eV of the literature, having used only 400 structures for the slab, rather than the 3000 in the paper. Having obtained <math>e \Delta \bar{\phi}</math>, you now have the reference to the vacuum, necessary for calculating the absolute redox potentials.		These values agree to within 0.1 eV of the literature, having used only 400 structures for the slab, rather than the 3000 in the paper. Having obtained <math>e \Delta \bar{\phi}</math>, you now have the reference to the vacuum, necessary for calculating the absolute redox potentials.

	In the next stage, you will perform TI from Fe<sup>3+</sup> to Fe<sup>2+</sup> using MLFFs.		In the next stage, you will perform TI from Fe<sup>3+</sup> to Fe<sup>2+</sup> using MLFFs.
Line 1,478:		Line 1,478:

	== Recommendations and advice ==		== Recommendations and advice ==
	*~~We recommend reading~~ the local potential from {{FILE\|vaspout.h5}}, rather than {{FILE\|LOCPOT}}, as you can take advantage of the HDF5 tools for data extraction.		*Read the local potential from {{FILE\|vaspout.h5}}, rather than {{FILE\|LOCPOT}}, as you can take advantage of the HDF5 tools for data extraction.
	*3000 slab calculations will create a lot of large {{FILE\|OUTCAR}}, {{FILE\|vaspout.h5}}, and {{FILE\|LOCPOT}} files (several 100 GBs). Test your setup using the Python scripts in your run script to avoid this.		*3000 slab calculations will create a lot of large {{FILE\|OUTCAR}}, {{FILE\|vaspout.h5}}, and {{FILE\|LOCPOT}} files (several 100 GBs). Test your setup using the Python scripts in your run script to avoid this.
	*~~We recommend using~~ a stochastic thermostat, such as the Langevin thermostat ({{TAG\|MDALGO}}=3), because it generally provides more ergodic sampling than deterministic alternatives.		*Use a stochastic thermostat, such as the Langevin thermostat ({{TAG\|MDALGO}}=3), because it generally provides more ergodic sampling than deterministic alternatives.

	== Related tags and articles ==		== Related tags and articles ==