Coder Social home page Coder Social logo

vaspsol's Introduction

vaspsol

New VASPsol ++

VASPsol is a community driven code implementing implicit solvents described by continuum models. A new version VASPsol++, has been released.This page will be maintained, however, all updates to VASPsol's code will now feature at this new repository : https://github.com/VASPsol/VASPsol/

VAPSsol ++ Paper : https://pubs.aip.org/aip/jcp/article-abstract/159/23/234117/2930317/An-implicit-electrolyte-model-for-plane-wave?redirectedFrom=fulltext

VASPsol (2014)

We have implemented an implicit solvation model that describes the effect of electrostatics, cavitation, and dispersion on the interaction between a solute and solvent into the plane-wave DFT code VASP. Our implementation provides a computationally efficient means to calculate the effects of solvation on molecules and crystal surfaces as well as reaction barriers. The strength of our solvation model implementation is its capability to handle large periodic systems such as metal and semiconductor surfaces and its interoperability with standard ultrasoft pseudopotential and projector-augmented wave potential libraries.

Prerequisites

VASP versions 5.2.12 or 5.3.3 or 5.3.5 or >=5.4.1.

Compiler and library requirements are the same as that of VASP ([vasp wiki] (http://cms.mpi.univie.ac.at/wiki/index.php/Installing_VASP))

Installation

For VASP versions = 5.2.12 or 5.3.3 or 5.3.5:

Note:

VASP license prohibits us from distributing the patch files on a public platform like github. If you would like to use VASPsol with VASP versions 5.2.12, 5.3.3 or 5.3.5, please contact Dr. Richard Hennig (rhennig[at]mse.ufl.edu) or me (km468[at]cornell.edu) to obtain the required patch file. Sorry for the inconvenience

  • Apply the appropriate interface patch to the original VASP source code. There are 3 interface_patch files available in the VASPsol/patches folder, one for each supported versions of VASP.
    cd <VASP src directory>
    patch -p1 < <path to the interface patch file>
  • After applying the patch, copy VASPsol/src/solvation.F file to the VASP source directory:
  • In the original VASP Makefile, put solvation.o object file name before pot.o in that order.
  • make clean
  • make

Note:

The file VASPsol/src/solvation.F combines all the modules defined in the files in the folder VASPsol/src/modules so as to conform with the new builtin interface for solvation in vasp(versions >= 5.4.1). The files in the VASPsol/src/modules folder do not play any role in the installation procedure.

For VASP versions >= 5.4.1:

These versions of VASP support solvation calculations out of the box. The only thing that the user needs to do is to update the solvation routines by copying the VASPsol/src/solvation.F file to the src/ folder in the original VASP folder, vasp.5.4.X/src/ and follow the install instructions for the compilation of the original VASP.

Note on versions >=5.4.4:

It seems like the complementary error function definition has been refactored a bit in the newer versions. Inorder to ensure the solvation module's compatibility with the newer versions of vasp please add -Dsol_compat option to the list of precompiler options(CPP_OPTIONS) in the makefile.

Note on binaries:

VASP.5.4.1 by default compiles all the three binaries: standard one(vasp_std), gamma version(vasp_gamma) and the one for noncollinear calculations(vasp_ncl). Some users have reported in the past that the version 5.4.1 compilation fails after updating the solvation related files. This was due to the fact that the solvation modules were written and tested using the real to complex FFTs(same as vasp_std and vasp_gamma) whereas the non-collinear compilation uses complex to complex FFTs. The issue has been fixed and all three binaries will compile fine without any errors.

Note on running Poisson-Boltzmann solver with 5.4.1:

To compute the FERMI_SHIFT as mentioned in the documentation in version 5.4.1, minor modification is required to the pot.F file. This can be done by applying the patch file pbz_patch_541 provided in the src/patches folder as follows.

    cd vasp.5.4.1
    patch src/pot.F < <path to the pbz_patch_541 patch file>

For VASP6 versions >= 6.1.0:

For VASP versions > 6.1.0 there is currently a patch in /VASPsol/patches/VASPsol6.patch. This patch applies the FERMI_SHIFT patch to the pot.F as well as providing additional subroutines. This patch was developed by user shk11 in Issue #43.

  • copy solvation.F from path_to_VASPsol/src/solvation.F to path_to_VASP6_install/src/
  • Navigate to path_to_VASP6_install/src/ and modify the .objects file by ensuring that “solvation.o” appears before “pot.o”
  • Set the CPP option "-Dsol_compat" in the VASP makefile.include file.
  • Compile the code as described in the instructions on the VASP wiki.
    cd vasp.6.1.0
    patch -p0 < <path to the pbz_patch_610 patch file>

Usage

Please see the file USAGE.md in the docs folder for instructions and the examples folder for some sample calculations.

License

VASPSol is released under the terms of Apache version 2.0 license:

Copyright 2013-2015 Kiran Mathew and Richard Hennig

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.

For discussions and feedback

Join our mailing list: https://groups.google.com/forum/#!forum/vaspsol

Contributing

Please contact Dr. Richard Hennig (rhennig[at]mse.ufl.edu)

Authors

Kiran Mathew

Dr. Richard Hennig

How to cite

DOI for the VASPsol Github repository:

DOI

If you use the VASPsol package in your research, please cite the following papers:

  • Implicit solvation model for density-functional study of nanocrystal surfaces and reaction pathways. K. Mathew, R. Sundararaman, K. Letchworth-Weaver, T. A. Arias, and R. G. Hennig, J. Chem. Phys. 140, 084106 (2014), [doi:10.1063/1.4865107] (https://doi.org/10.1063/1.4865107).

  • Implicit self-consistent electrolyte model in plane-wave density-functional theory. K. Mathew, V. S. C. Kolluru, S. Mula, S. N. Steinmann, and R. G. Hennig, J. Chem. Phys. 151, 234101 (2019), [doi:10.1063/1.5132354] (https://doi.org/10.1063/1.5132354).

BibTex entry for the Github repository and the publications::

   @misc{VASPsol-Software,
     title        = {VASPsol: Implicit solvation and electrolyte model for density-functional theory},
     author       = {K. Mathew and V. S. Chaitanya Kolluru and R. G. Hennig},
     year         = 2018,
     publisher    = {GitHub},
     journal      = {GitHub repository},
     howpublished = {\url{https://github.com/henniggroup/VASPsol}},
     url          = {https://github.com/henniggroup/VASPsol},
     doi          = {10.5281/zenodo.2555053}
   }
   
   @article{VASPsol2014-Dielectric,
     title        = {Implicit solvation model for density-functional study of nanocrystal surfaces
                     and reaction pathways.},
     author       = {K. Mathew and R. Sundararaman and K. Letchworth-Weaver and T. A. Arias and
                     R. G. Hennig},
     year         = 2014,
     journal      = {J. Chem. Phys.},
     volume       = 140,
     pages        = {084106},
     doi          = {10.1063/1.4865107}
   }
   
   @article{VASPsol2019-Electrolyte,
     title        = {Implicit self-consistent electrolyte model in plane-wave density-functional theory.},
     author       = {K. Mathew and V. S. C. Kolluru and S. Mula and S. N. Steinmann and R. G. Hennig},
     year         = 2019,
     journal      = {J. Chem. Phys.},
     volume       = 151,
     pages        = {234101},
     doi          = {10.1063/1.5132354}
   }

vaspsol's People

Contributors

ericfonseca95 avatar javierbertoli avatar matk86 avatar rhennig avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

vaspsol's Issues

How to set 1M electrolyte in INCAR using vaspsol?

I saw in the article "Implicit self-consistent description of electrolyte in plane-wave density-functional
theory" in 2016 that 1M electrolyte was used. Could you please tell me how to set it? The parameters you mentioned in INCAR have nothing to do with the electrolyte concentration.

Unable to build VASP 5.4.4 with new Intel OneAPI compilers, mpi mkl

I am using the following versions of Intel OneAPI compilers:
Compilers 2022.1.0
MPI 2021.6.0
MKL 2022.1.0
When running "make all", I get the following error:

solvation.o: In function `pot_lpcm_k_mp_createcavity_':
solvation.f90:(.text+0x14a6d): undefined reference to `errfc_'
make[2]: *** [makefile:145: vasp] Error 1
make[2]: Leaving directory '/.../vaspsol/vasp.5.4.4/build/std'
cp: cannot stat 'vasp': No such file or directory
make[1]: *** [makefile:142: all] Error 1
make[1]: Leaving directory '/.../vaspsol/vasp.5.4.4/build/std'
make: *** [makefile:10: std] Error 2

Not sure how to resolve this issue.

Water molecules at the intersection of implicit and explicit solvents behave abnormally

Dear developer,
I recently use VASPsol to simulate the MD of the interface system. At first the water molecules are behaving normally, but then they start to move towards the vacuum layer(pic below). Is this a normal phenomenon? If that's not the case, then how do I set up the explicit solvent/implicit solvent interaction layer to be more stable. Any suggestions are welcome.

pic:
The beginning of the MD
image

The end of the MD.
image

And here is my INCAR:

ISTART = 0
ICHARG = 2
INIWAV = 1

ISPIN = 2

ENCUT = 400
NELM = 200
NELMIN = 4
EDIFF = 1E-05
PREC = Normal
LREAL = Auto

#for vaspsol
LSOL = .TRUE.
EB_k = 78.4
LAMBDA_D_K = 3.04

EDIFFG = -0.02
NSW = 1000
IBRION = 0
ISIF = 2
ISYM = 0
POTIM = 1
TEBEG = 300
TEEND = 300
SMASS = 2
NBLOCK = 1
KBLOCK= 100
MDALGO = 2

ISMEAR = 0
SIGMA = 0.05

ALGO = VERY FAST
LPLANE = .TRUE.
LSCALU = .FALSE.
NCORE = 4
NSIM = 4
POMASS = 55.847 12.011 14.001 16.000 2.000
IWAVPR=12

LWAVE = .FALSE.
LCHARG = .FALSE.
LVTOT = .FALSE.
LELF = .FALSE.
IVDW= 12

VASPsol convergence issue for CeO2(111)

Dear Prof. Hennig,

I tried applying VASPsol to a cerium dioxide surface (CeO2(111); clean, without any adsorbate on it), and the wave function failed to converge despite my best efforts. To my knowledge, it works well for many other oxides such as titania, nickel oxide, and iridium oxide, and indeed I tried VASPsol with titanium using the same settings and did not have any convergence issues. Below I attach the relevant files for the ceria job. I would very much appreciate any help you can give me for getting this system converged.

POTCAR (using the 2012 PAW_PBE set)

KPOINTS:
Automatic mesh
0
Gamma
2 2 1
0 0 0

INCAR:
System = CeO2(111) 4x4x3 slab

LSOL = .TRUE.

NPAR = 1
GGA = 91
VOSKOWN= 1
ENCUT = 400.00 eV
LPLANE = .TRUE.
LREAL = Auto
ALGO = Fast
PREC = Normal
NELM = 70
EDIFF = 1e-5
IDIPOL = 3
ISMEAR = 0
SIGMA = 0.05
ISPIN = 1
ISIF = 2
NSW = 0
LDAU = .TRUE.
LDAUTYPE = 2
LDAUL = 3 -1
LDAUU = 6 0
LDAUJ = 1 0

POSCAR:
CeO2(111)
3.884844655838890
2.0000000000000000 0.0000000000000000 0.0000000000000000
-1.0000000000000000 1.7320508075688772 0.0000000000000000
0.0000000000000000 0.0000000000000000 5.1775494819768904
Ce O
12 24
Selective dynamics
Direct
0.1666666666666643 0.3333333333333357 0.0394248564774671 F F F
0.6666666666666643 0.3333333333333357 0.0394248564774671 F F F
0.1666666666666643 0.8333333333333357 0.0394248564774671 F F F
0.6666666666666643 0.8333333333333357 0.0394248564774671 F F F
0.3333333333333357 0.1666666666666643 0.1971242823873425 F F F
0.8333333333333357 0.1666666666666643 0.1971242823873425 F F F
0.3333333333333357 0.6666666666666643 0.1971242823873425 F F F
0.8333333333333357 0.6666666666666643 0.1971242823873425 F F F
0.0000000000000000 0.0000000000000000 0.3532008151332775 T T T
0.5000000000000000 0.0000000000000000 0.3532008151332775 T T T
0.0000000000000000 0.5000000000000000 0.3532008151332775 T T T
0.5000000000000000 0.5000000000000000 0.3532008151332775 T T T
0.0000000000000000 0.0000000000000000 0.0000000000000000 F F F
0.5000000000000000 0.0000000000000000 0.0000000000000000 F F F
0.0000000000000000 0.5000000000000000 0.0000000000000000 F F F
0.5000000000000000 0.5000000000000000 0.0000000000000000 F F F
0.3333333333333357 0.1666666666666643 0.0788497129549413 F F F
0.8333333333333357 0.1666666666666643 0.0788497129549413 F F F
0.3333333333333357 0.6666666666666643 0.0788497129549413 F F F
0.8333333333333357 0.6666666666666643 0.0788497129549413 F F F
0.1666666666666643 0.3333333333333357 0.1576994259098754 F F F
0.6666666666666643 0.3333333333333357 0.1576994259098754 F F F
0.1666666666666643 0.8333333333333357 0.1576994259098754 F F F
0.6666666666666643 0.8333333333333357 0.1576994259098754 F F F
0.0000000000000000 0.0000000000000000 0.2365491388648095 F F F
0.5000000000000000 0.0000000000000000 0.2365491388648095 F F F
0.0000000000000000 0.5000000000000000 0.2365491388648095 F F F
0.5000000000000000 0.5000000000000000 0.2365491388648095 F F F
0.3333333333333357 0.1666666666666643 0.3137089924832708 T T T
0.8333333333333357 0.1666666666666643 0.3137089924832708 T T T
0.3333333333333357 0.6666666666666643 0.3137089924832708 T T T
0.8333333333333357 0.6666666666666643 0.3137089924832708 T T T
0.1666666666666643 0.3333333333333357 0.3918096184401207 T T T
0.6666666666666643 0.3333333333333357 0.3918096184401207 T T T
0.1666666666666643 0.8333333333333357 0.3918096184401207 T T T
0.6666666666666643 0.8333333333333357 0.3918096184401207 T T T

Excerpt of OSZICAR:
N E dE d eps ncg rms rms(c)
SOL: 1 -0.52812E+03 0.60794E-01 -0.52806E+03 103
DAV: 1 -0.820702626470E+03 -0.82070E+03 -0.32056E+01 1280 0.181E+01 0.988E+00
SOL: 2 -0.81573E+03 0.61764E-01 -0.81566E+03 119
RMM: 2 -0.110938402880E+04 -0.28868E+03 -0.54493E+01 1092 0.186E+01 0.556E+00
SOL: 3 -0.10752E+04 0.63630E-01 -0.10752E+04 121
RMM: 3 -0.377120855499E+04 -0.26618E+04 0.26545E+05 1096 0.392E+03 0.335E+02
SOL: 4 -0.10999E+04 0.22198E+00 -0.10997E+04 121
RMM: 4 -0.860774897451E+03 0.29104E+04 -0.32067E+03 1289 0.169E+02 0.147E+02
SOL: 5 -0.10494E+04 0.15546E+00 -0.10492E+04 118
RMM: 5 -0.112374524039E+04 -0.26297E+03 -0.26885E+03 1271 0.161E+02 0.347E+01
SOL: 6 -0.10422E+04 0.95755E-01 -0.10421E+04 112
RMM: 6 0.300259658379E+04 0.41263E+04 0.10527E+05 865 0.283E+03 0.261E+02
SOL: 7 -0.87971E+03 0.20732E+00 -0.87950E+03 117
RMM: 7 -0.918736494701E+03 -0.39213E+04 -0.66175E+03 1072 0.443E+02 0.186E+02
SOL: 8 -0.91527E+03 0.16960E+00 -0.91510E+03 116
RMM: 8 -0.103056369799E+04 -0.11183E+03 -0.40561E+03 1320 0.178E+02 0.508E+01
SOL: 9 -0.10156E+04 0.13218E+00 -0.10155E+04 119
RMM: 9 0.567275679760E+04 0.67033E+04 0.22001E+05 848 0.379E+03 0.271E+02
SOL: 10 -0.11002E+04 0.21664E+00 -0.10999E+04 120
RMM: 10 0.579347565339E+04 0.12072E+03 -0.31253E+04 782 0.884E+03 0.667E+02
SOL: 11 -0.12206E+04 0.27399E+00 -0.12203E+04 122
RMM: 11 -0.963696014479E+03 -0.67572E+04 -0.55210E+03 1195 0.252E+02 0.108E+02
SOL: 12 -0.10984E+04 0.24188E+00 -0.10981E+04 120
RMM: 12 -0.174837438657E+04 -0.78468E+03 -0.87259E+03 868 0.806E+02 0.116E+02
SOL: 13 -0.10492E+04 0.27924E+00 -0.10489E+04 117
RMM: 13 -0.104449713950E+04 0.70388E+03 -0.22427E+03 1292 0.149E+02 0.695E+01
SOL: 14 -0.10493E+04 0.23889E+00 -0.10491E+04 113
RMM: 14 -0.218489173678E+04 -0.11404E+04 -0.61078E+03 982 0.708E+02 0.120E+02
SOL: 15 -0.91098E+03 0.17805E+00 -0.91080E+03 104
RMM: 15 -0.100270769556E+04 0.11822E+04 -0.26748E+03 1279 0.149E+02 0.649E+01
SOL: 16 -0.80444E+03 0.16273E+00 -0.80427E+03 119
RMM: 16 0.566190980108E+04 0.66646E+04 0.25522E+05 1164 0.646E+03 0.680E+02
SOL: 17 -0.85792E+03 0.28412E+00 -0.85763E+03 123
RMM: 17 -0.483082404742E+03 -0.61450E+04 -0.61611E+03 1274 0.265E+02 0.127E+02
SOL: 18 -0.69718E+03 0.17535E+00 -0.69700E+03 125
RMM: 18 0.250364473675E+06 0.25085E+06 0.25895E+06 921 0.245E+04 0.221E+03
SOL: 19 -0.11238E+04 0.42367E+00 -0.11234E+04 124
RMM: 19 0.290791016175E+04 -0.24746E+06 -0.25034E+04 1298 0.151E+03 0.132E+03
SOL: 20 -0.92575E+03 0.32067E+00 -0.92543E+03 121
RMM: 20 0.121484422355E+04 -0.16931E+04 -0.27968E+04 867 0.338E+03 0.122E+03
SOL: 21 -0.61003E+03 0.33302E+00 -0.60970E+03 122
RMM: 21 -0.353509281836E+02 -0.12502E+04 -0.21922E+04 1348 0.132E+03 0.106E+03
...

I used previously converged WAVECAR and CHGCAR without VASPsol but that did not make any difference. I also tried various other things, such as turning off DFT+U, changing to gamma-point only, increasing the k-point density or cutoff energy, using an earlier set of PAW potentials, but nothing has helped.

Best,
Neil

How to control the constant potential ?

  1. If a want to perform the calculation under a constant potential (e.g. 1V versus SHE), Are the following steps right ?
    0). Let the target potentail be Φ0,the current potentail is Φ. And
    Φ = W - 4.6
    Where, W is the work function of the slab, 4.6 eV is the potential of SHE.
    1). Get the capacitance C by finite difference method: increase NELECT by a small number δn, and get the difference of the work function δW, the C = δn / δW.
    2). When a SCF loop is done, get Φ,and the number of electrons we need to adjust ΔN is
    ΔN = C*(Φ - Φ0)
  2. In src dir, I find a solvation.F and src/modules also contains a solvation.F, what's their difference ?

How to set FLUCAR for JDFTx

When I read solvation.F code I notice that it can use JDFTx interface, but an input file JDFTx is inquired. Is FLUCAR the same as the input file of JDFTx ?

Patches for VASP 6.3

Thanks for providing these additions to the community. We want to compile VASPsol with VASP 6.3 and noticed that the patches need to be updated. @shk11 I noticed you produced patches for 6.1 and 6.2. Can you also provide patches for VASP 6.3? Is it possible to put the suggested patches in the repo itself?

Tips for convergence

Hi,
First of all, thank VASPsol team for your contribution to computational chemistry and materials science.
It's a really great job!

I'm interested in understanding solvation effects in the catalytic activities on the metal-oxide surface, so I'm trying to apply VASPsol code into pre-converged (at vacuum) systems. While investigated several catalyst structures, I found that the electronic relaxations were not converged, in particular, for the complicated structures where the surface of metal-oxide was partially covered with other species (such as TMDs).

Even though I tried to tweak the original VASP options to control the convergence at the electronic relaxation step (algorithms, # of bands, mixing parameters...), the energy badly fluctuated at some point.

Is there any general procedure I can try to achieve electronic convergences for VASPsol run?

Since I could sometimes get the convergence for the lower EB_K (~20) than water, I think it's maybe because of too complicated structures for the large dielectric effects in water. But, it will be extremely thankful if I can get any information on the convergence in VASPsol.

Thank you.

Speed of solvation step with VASPsol

I am wondering if others find that the solvation steps within the SCF loop take longer than the DFT steps. I'm trying to decide if this makes sense computationally or if the efficiency of the code could/should be improved.

Inconsistency between FERMI_SHIFT and electrostatic potential in the middle of vacuum

Hi,
I am currently repeating the grand canonical potential test in J. Chem. Phys. 151, 234101 (2019) (for Cu-100 surface). I noticed that a correction to total energy QV is needed, which is the net charge and "the electrostatic potential to align the potential in the electrolyte region to 0".
But I found that the FERMI_SHIFT in the output file is a little different with the integrated average planar electrostatic potential in the middle of the vacuum along z direction. The FERMI_SHIFT is nearly constant, while the calculated potential would change linearly with the net charge.
Is this just a mathematical error or something wrong with my output?

Comparison

Best,
Zhichao

FERMI_SHIFT calculation in VASPSol

In solvation.F, fermi_shift is calculated by VACPOT_POIS - VACPOT_PSP. Both VACPOT_POIS and VACPOT_PSP are some electrostatic potential where the electron density reaches a minimum.
VACPOT_PSP is calculated from Hartree potential and ionic potential (with the local part of ionic pseudopotential) from POTLOK subroutine in pot.F, while VACPOT_POIS is calculated from the Hatree potential generated by total charge density (electron density plus ionic gaussian density, without local pseudo potential).
So it seems that VACPOT_POIS is a potential reference. Can anyone tell me why ? Thanks.

Fermi energy changes with the value of parameter NC_K

Dear developers of VASPsol,
I am doing some research about the explicit/implicit solvent model with VASPsol code, in which I build a slab with 1 or 2 layers of explicit ice-like water above it. I have tried to modify the parameter NC_K with the purpose to make sure the counter-ion would not seep into the explicit molecules region. However, I find that the Fermi energy will dramatically change with the variation of the value of NC_K, as follows:
e.g (NC_K, E-fermi, Vaccum level)
(2.5E-5, -5.7899, -0.368)
(2.5E-4, -5.4576, -0.369)
(2.5E-6, -4.9104, -0.368)
This is a puzzle, I wonder if the variation of Fermi energy is reasonable and how can I obtain the correct PZC by adjusting NC_K

Some internal variables declared as REAL(q) instead of RGRID

Hello,

I noticed that solvation.F would not compile properly when VASP is compiled in complex mode (e.g. for noncollinear calculations). Here's a diff that shows how I fixed the problem in src/solvation.F:

diff --git a/src/solvation.F b/src/solvation.F
index 4a2492e..853bafc 100644
--- a/src/solvation.F
+++ b/src/solvation.F
@@ -91,7 +91,7 @@ SUBROUTINE Vcorrection_lpcm(SP_GRIDC, SP_LATT_CUR, SP_P, &
  COMPLEX(q),ALLOCATABLE :: CWORK_K(:)
 #endif
  !vacuum level
- REAL(q),ALLOCATABLE :: RHOB(:), RHOION(:)
+ RGRID,ALLOCATABLE :: RHOB(:), RHOION(:)

  ALLOCATE(SP_d_eps(DIMREAL(SP_GRIDC%MPLWV)) )
  ALLOCATE(SP_kappa2(DIMREAL(SP_GRIDC%MPLWV)))
@@ -1663,7 +1663,8 @@ SUBROUTINE COMPUTE_RHOB(GRIDC, LATT_CUR, T_INFO, EPSILON, PHI, RHOB)
   TYPE(type_info),INTENT(IN) :: T_INFO

   COMPLEX(q) :: PHI(GRIDC%MPLWV)
-  REAL(q) :: EPSILON(DIMREAL(GRIDC%MPLWV)), RHOB(DIMREAL(GRIDC%MPLWV))
+  REAL(q) :: EPSILON(DIMREAL(GRIDC%MPLWV))
+  RGRID :: RHOB(DIMREAL(GRIDC%MPLWV))

   COMPLEX(q),ALLOCATABLE :: gradphi_x_in(:)
   COMPLEX(q),ALLOCATABLE :: gradphi_y_in(:)
@@ -1719,9 +1720,9 @@ SUBROUTINE COMPUTE_RHOION(GRIDC, LATT_CUR, T_INFO, KAPPA2, PHI,RHOION)

   COMPLEX(q) :: PHI(GRIDC%MPLWV)
   REAL(q) :: KAPPA2(DIMREAL(GRIDC%MPLWV))
-  REAL(q) :: RHOION(DIMREAL(GRIDC%MPLWV))
+  RGRID :: RHOION(DIMREAL(GRIDC%MPLWV))

-  REAL(q),ALLOCATABLE :: R_PHI(:)
+  RGRID,ALLOCATABLE :: R_PHI(:)

   INTEGER(q) :: I

I have not tested this new code, but at least it compiles now. Note: This is with VASP 5.3.5.

Grand-canonical energy calculations of charged systems

I have been been attempting to repeat the calculations performed on the Pt(111) slab from the JCP 2019 (doi.org/10.1063/1.5132354) paper.

Due to what is likely a trivial misconception, I am having trouble simulating your graph of the grand-canoncial energy of a charged slab shown in Fig. 2.

I have firstly corrected for the DFT energy of the charged slab based on Q V,
"where Q is the net charge of the simulation cell and V the shift in reference potential, e.g., V = FERMI_SHIFT."
(If I have increased NELECT by 0.05 I take the charge as -0.05).

This then gives me a linear dependance of the energy of the system with respect to the charge (q).

I then note the reference to "The grand canonical electronic energy, F, is the Legendre transformation of the free energy, A, of the system, F(U) = A(n)-n_electrode x U" where n_electrode is the net charge of the electrode slab and U is the applied potential.

But I am unsure if this needs to be applied and if so I am still not getting the quadratic behaviour. I am using the Pt(111) surface from Materials Project (9 atoms) with a vacuum of 30 Angstroms.

Much appreciated
Alex Aziz

Can VASPsol simulate solvated effects on the adsorption of radionuclide ions on 2D materials

Dear developers of VASPsol,

I am working on modeling and simulation of radionuclide ions (U (VI), Sr (II), Th(IV) etc.) on graphene oxides. I would like to investigate the effect of solvent on the adsorption configurations of radionuclide ions on graphene oxides. I understand from the articles that VASPsol parameters were fit to a range of neutral molecules and surfaces. Can VASPsol be applied for simulating the solvated effects on ion-adsorbed graphene oxide systems by modifying the default parameters.
I would greatly appreciate any response on this.

Thanks in advance,
Seba

Solvation energy (Solvation Ediel_sol ) of ion-electron system

Dear Altruists

I am working with DMSO electrolyte in implicit water solvent. For solvation energy, we have to take the difference of ground state energy of vacuum and solvation state.
However, I am confused about the physical significance of "Ediel_sol" value in each electronic step how it contributes to the solvation energy. I am adding the plot here.
Screenshot from 2022-02-28 23-49-53

Thanks,
Musanna

Confusion about DFT energy correction using FERMI_SHIFT

Hi everyone,

When I do calculations about charged slab using VASPsol, I notice it need an energy correction for the final DFT energy from OUTCAR:

Furthermore, if you shift the electrostatic potential to align the potential in the
electrolyte region to zero, or any other value, you also need to add a correction to
the energy of your system. This correction is given by Q V, where  Q is the net
charge of the simulation cell and V the shift in reference potential, e.g.,
V = FERMI_SHIFT. 

if the charged slab has more electrons than neutral slab, Q will be a negative number, right? for example, "-5|e|", which |e| means the absolute value of an electron's charge. So I wonder whether the new corrected energy is : Enew = E(DFT) + Q * FERMI_SHIFT,or Enew = E(DFT) - Q * FERMI_SHIFT ?

Discontinuities in the Fermi Energy

I am working on Mg surfaces, having started with the 001 surface the results are consistent with what I would expect and I obtain the quadratic behaviour of the grand canonical energy versus electrode potential. Moving on to the 100 surface I am obtaining discontinuities in the Fermi energy as a function of charge (NELECT).

This wasn’t the case for my other calculations which exhibited an almost linear increase in the Fermi energy as a function of charge. At first, I thought this was a smearing or convergence issue but I have tried changing some of the parameters with little effect. The change is significantly large, see the table below was computed with a system of 12 atoms (1 atoms per layer with 24 electrons for the neutral slab).

I am wondering if you have come across anything similar or you had any suggestions?

No of Electrons Fermi Level (eV)
23.7500 -2.4600
23.8000 -2.4431
23.8500 -2.4311
23.9000 -2.4245
23.9500 -3.1135
24.0000 -2.6020
24.0500 -3.4477
24.1000 -3.4226
24.1500 -3.3600
24.2000 -3.2602
24.2500 -3.1266

Many thanks

Alex Aziz

How to set the parameters in the isolated system of charged adsorbates?

Dear developers of VASPsol,
I would like to use vaspsol to calculate the energy of an isolated charged system. I constructed a finite length carbon nanotube with charged peroxodisulfate ions adsorbed on it. The transition state calculations for the dissociation of this molecule were performed under vacuum, after which I want to calculate the solvation effect in a pure aqueous solution.
How can I set the parameters to calculate the energy, do I just need to add EB_k = 80, LSOL = .TRUE. for the solvation calculation and
IDIPOL=4?
The energy (with solvation calculation) of the transition state obtained from my calculation is lower than the initial state, which is obviously unreasonable (the reaction energy barrier is around 1 eV under vacuum conditions). Is this related to the fact that the system is charged?

Here are my INCAR parameters:

PREC = Accurate
ENCUT = 520.000
NPAR=4
ISYM=0
IDIPOL=4
LSOL = .TRUE.
EB_K = 80
IBRION = -1
NSW = 1

ISIF = 2
NELMIN = 2
EDIFF = 1.0e-05
EDIFFG = -0.03
VOSKOWN = 1
NBLOCK = 1
NWRITE = 1
NELECT = 690.0
ISYM = 0
NELM = 200
ALGO = Fast
IVDW = 12
VDW_S6 = 1.000
VDW_S8 = 0.7875
VDW_A1 = 0.4289
VDW_A2 = 4.4407
ISPIN = 1
INIWAV = 1
ISTART = 1
ICHARG = 0
LWAVE = T
LCHARG = .FALSE.
ADDGRID = .FALSE.
ISMEAR = 0
SIGMA = 0.05
LREAL = Auto
LSCALAPACK = .FALSE.
RWIGS = 0.77 0.32 0.73 1.02 0.99

Continuum solvation model with constant electrolyte concentration with different thickness

Dear Altruists

I have a rough metal surface on top of which I am considering the PCM solvation modeled by the VASPsol. I know that in VASPsol the thickness of the electrolyte is controlled by the ionic concentration of the electrolyte, which is set by LAMBDA_D_K (LAMBDA_D_K = 3.04 Å / sqrt(ionic concentration)).

In my calculation, I want to study the effect of electrolyte thickness with constant ionic concentration but different thicknesses. I would like to know that is it possible to use an electrolyte of constant concentration with different thicknesses on top of the metal surface by the VASPsol. If so, which tag should I consider for changing the thickness of the electrolyte and which one to keep the concentration constant?

Thank you very much for your kind dedication.

Sincerely
Arindam Sannyal

What are the units for surface tension "TAU"?

I would like to use VASPsol to simulate solvation in methanol. It appears that VASPsol will default to water as a solvent, and will default to a surface tension of 5.25E-4. However, I am not sure what the units are on this. I have checked eV/A^2, hartree/bohr^2, and every combination in between (including Rydbergs), and none of these seem to line up with the default value in VASPsol if the surface tension of water is taken to be 72.8 mN/m. Therefore, what are the units for "TAU"? What should I set as the value of TAU if I want to simulate methanol, which has a surface tension of about 22.5 mN/m?

Implicit self-consistent electrolyte model

As starting point, and since the iterative calculations with the electrolyte are computationally demanding, I decided to perform some test with a simple system by selecting a molecule (OCHO), before implementing on the actual systems. In this case, after performing a structural optimization, I include the solvent with the following parameters: NC_K = 0.0025, SIGMA_K = 0.6, TAU = 0.0 and EB_K = 78.4, obtaining a Fermi energy Ef=-6.9731eV.

The next step was modifying the total number of valence electrons, (in this case NELEC=17) by increasing the value of NELECT by 0.01 until NELECT=17.1 and decreasing by 0.01 until a final value of NELECT=16.9.

In this case, I would like to ask how to associate and calculate a potential associated to the shift of the Fermi level for each case for example:

for NELEC=17 the resulting fermi energy amounts to Ef=-6.9731 eV, then for NELECT= 17.01 the fermi energy amounts to Ef= -6.9598 eV , hence the shift in the energy level ( -6.9598 - (-6.9731)) corresponds to 1.33E-02 eV. Here, if I use the standard definition of Volt, I will need to divide the 1.33e-2 eV by the elementary charge of an electron (1.6e-19 V) is this correct?, I assume that I am missing something since the value does not have an adequate magnitude.

Thanks in advance,

Jose Guillermo

RHOION file and RHOB file

I am using LRHOION=.TRUE. and LRHOB=.TRUE. tags to generate RHOION and RHOB file. I think they are writing the information of counter-ion concentration and bound charge concentration in the electrolyte, respectively. However, I am not sure about the unit of the value written in RHOION and RHOB file. Is it mol/L or mM? Could you please explain what the exact meaning of the data in the RHOION and RHOB file is? Is the sum of all the data the opposite of net charge concentration of the solid solute? In your paper, Implicit self-consistent electrolyte model in plane-wave density-functional theory, the figure 5 showed an example of plot generated from RHOION file, so I am wondering how you processed the RHOION file to get the plot? Thank you!

VASPSol with GPU version of VASP

Dear developers,

As reported in the VASP Forum ( https://www.vasp.at/forum/viewtopic.php?f=7&t=17564 ), there is a problem with the VASPSol implementation and the GPU version of VASP.
The solvent subroutine is only called after the SCF cycle is converged, not at every step as in the CPU version. This gives different results between the CPU and GPU versions, making the VASPSol unusable at the VASP GPU version.
As the performance advantages of the GPU version are striking, it would be very useful to take a look at this bug, although, I don't know if it is on the VASP side or the VASPSol plugin.

Yours sincerely,
James Almeida.

Potential vs z calculated using VASPsol?

image
Can you explain why there is a sharp peak in the end of the simulation cell? Is it because the VASP is trying to set the average electrostatic potential to 0? The system is Cu(111) with CO adsorbed on it with 3 water molecules.

VASPsol for solid-solid interfaces

Hi,
I wish to know if it is possible to use the dielectric constant of a solid electrolyte as the dielectric in VASPsol to calculate the properties of a solid-solid interface.

Also, I have run a job using VASPsol
Tags:
LSOL = .TRUE.
EB_k = 5.5
TAU = 0
NC_K = 5E-5

However, I did not find any clue in the OUTCAR that VASP actually ran the VASPsol. I checked with the compilation and VASPsol should be compiled correctly.

Please, advice.

Thank you in advance.

Unable to build VASP 6.2.0 with GPU support

Hello,

I am unable to build VASP 6.2.0 with VASPSol and GPU support. Everything seems to compile successfully, but at the linking stage I get this error:

mpif90 -mkl=sequential -lstdc++ -o vasp c2f_interface.o cuda_interface.o simd.o base.o base_gpu.o profiling.o string.o tutor.o version.o vhdf5_base.o incar_reader.o reader_base.o openmp.o openacc_struct.o mpi_gpu.o mpi_shmem.o mathtools.o hamil_struct.o radial_struct.o pseudo_struct.o mgrid_struct.o wave_struct_gpu.o nl_struct_gpu.o mkpoints_struct.o poscar_struct.o afqmc_struct.o phonon_struct.o fock_glb.o chi_glb.o smart_allocate.o xml.o extpot_glb.o constant.o vdwforcefield_glb.o jacobi.o main_mpi.o openacc.o scala.o asa.o lattice.o poscar.o ini.o mgrid.o setex_struct.o xclib.o vdw_nl.o xclib_grad_gpu.o setex.o radial.o pseudo.o gridq.o ebs.o symlib.o mkpoints.o wave_gpu.o wave_mpi_gpu.o wave_high_gpu.o bext.o spinsym.o symmetry.o lattlib.o random.o nonl.o nonlr_gpu.o nonl_high.o nonl_high_gpu.o dfast.o dfast_gpu.o choleski2_gpu.o mix.o hamil.o hamil_gpu.o xcgrad_gpu.o xcspin.o potex1.o potex2.o constrmag.o cl_shift.o relativistic.o LDApU.o paw_base.o metagga.o egrad.o pawsym.o pawfock.o pawlhf.o diis.o rhfatm.o hyperfine.o fock_ace.o paw.o mkpoints_full.o charge.o charge_gpu.o Lebedev-Laikov.o stockholder.o dipol.o solvation.o scpc.o pot.o pot_gpu.o tet.o dos.o elf.o hamil_rot.o chain.o dyna.o us.o core_rel.o aedens.o wavpre.o wavpre_noio_gpu.o broyden.o dynbr.o reader.o writer.o xml_writer.o brent.o stufak.o fileio.o vhdf5.o sphpro.o opergrid.o stepver.o chgloc.o fast_aug.o fock_multipole.o fock_gpu.o mkpoints_change.o subrot_cluster.o sym_grad.o mymath.o npt_dynamics.o subdftd3.o subdftd4.o internals.o dynconstr.o dimer_heyden.o dvvtrajectory.o vdwforcefield.o nmr.o pead.o k-proj.o subrot_gpu.o subrot_scf.o paircorrection.o rpa_force.o ml_interface.o force.o pwlhf.o gw_model.o optreal.o steep.o rmm-diis_gpu.o davidson_gpu.o david_inner.o root_find.o lcao_bare.o locproj.o electron_common.o electron_gpu.o rot.o electron_all.o shm.o pardens.o optics.o constr_cell_relax.o stm.o finite_diff.o elpol.o hamil_lr.o rmm-diis_lr.o subrot_lr.o lr_helper.o hamil_lrf.o elinear_response.o ilinear_response.o linear_optics.o setlocalpp.o wannier.o electron_OEP.o electron_lhf.o twoelectron4o.o gauss_quad.o m_unirnk.o minimax_ini.o minimax_dependence.o minimax_functions1D.o minimax_functions2D.o minimax_struct.o minimax_varpro.o minimax.o mlwf.o ratpol.o pade_fit.o screened_2e.o wave_cacher.o crpa.o chi_base.o wpot.o local_field.o ump2.o ump2kpar.o fcidump.o ump2no.o bse_te.o bse.o time_propagation.o acfdt.o afqmc.o main_gpu.o rpax.o chi.o acfdt_GG.o dmft.o GG_base.o greens_orbital.o lt_mp2.o rnd_orb_mp2.o greens_real_space.o chi_GG.o chi_super.o sydmat.o rmm-diis_mlr.o linear_response_NMR.o wannier_interpol.o wave_interpolate.o linear_response.o auger.o dmatrix.o phonon.o wannier_mats.o elphon.o core_con_mat.o embed.o extpot.o fftmpiw.o fftmpi_map.o fft3dlib.o fftw3d_gpu.o fftmpiw_gpu.o main.o -Llib -ldmy -Lparser -lparser /usr/local/apps/intel/2019/linux/mkl/lib/intel64/libmkl_scalapack_lp64.a -lmkl_blacs_openmpi_lp64 -Lparser -lparser -lstdc++ CUDA/lib/libCudaUtils_x86_64.a -L/usr/local/apps/cuda/10.2//lib64 -lnvToolsExt -lcudart -lcuda -lcufft -lcublas
solvation.o: In function pot_k_mp_vcorrection_': solvation.f90:(.text+0xb0de): undefined reference to m_sum_1_'
solvation.f90:(.text+0xb213): undefined reference to m_sum_1_' make[2]: *** [vasp] Error 1 make[2]: Leaving directory /nfs/hpc/apps/vasp/6.2.0/build/gpu'
cp: cannot stat ‘vasp’: No such file or directory
make[1]: *** [all] Error 1
make[1]: Leaving directory `/nfs/hpc/apps/vasp/6.2.0/build/gpu'

The non-GPU versions build without issue. Let me know what other information you need from me.

Thanks,

Rob

How to use NC_K variable correctly?

As shown in Fig.1, I find a variable NC_K, the critical of charge density in VASPsol OUTCAR file.
qq 20161028173435
Fig.1
It isn't reported in USAGE.md file. Whether does the NC_K variable represent cutoff charge density of the total solid-liquid system or cutoff charge density of empty space not occupied by the metal slab only? Could you introduce the NC_K variable in detail? And how to use NC_K correctly?
Many thanks!
Yuan

EC-IR simulation

Hi Prof. Hennig,
I’m a loyal user of VASPsol. It has been used to simulate PZCs of Au, Ag and Cu surfaces, and the double layer capacitance of Pt(111) surface with surface charge method (Tag NELECT in VASP). They agree well with experiment measurements.

Recently, I try using VASPsol with DFPT to simulate the electrochemical IR spectrum (EC-IR includes vibrational frequencies and IR intensity derived from Born effective charge) of CO adsorbed at metal surface but fail. The DFPT and implicit solvation model aren’t compatible, are they, why and what should I do to obtain the EC-IR based on VASPsol? Could you help me to solve these problem? Many thanks! The details of simulation are listed below.

As shown in Table 1, I use the dielectric constant of water (78.4) and 0.1 M ion strength to simulate the solvent layer. Based on JPCC2016 of Sautet group1, the size of solvent layer is set to 95 Å and tested in Fig.1.
Before using the implicit solvation model, I respectively used small displacement method (SD) and DFPT to calculate the frequencies of C-O stretching (first frequency) and Pt-C stretching (second frequency) in a vacuum system (Table 2). Both methods are equivalent. In the solvation system, their equivalence is lost, only SD method can obtain vibrational frequencies exactly corresponding to C-O stretching and Pt-C stretching, but DFPT not.

table1
fig1
table2

Reference

  1. S. N. Steinmann and P. Sautet, The Journal of Physical Chemistry C, 2016, 120, 5619-5623.

Best regards,
Yuan

No FERMI_SHIFT

Hi,

I just installed VASPsol (I'm using vasp5.4.4) it seems to be working; however, FERMI_SHIFT is not printed. I want to calculate the energy correction, Q V. The documentation, USAGE.md, says "the constant, FERMI_SHIFT, which is printed out to stdout, needs to be added to the Fermi level." I'm wondering if there is something wrong with my build or if there have been changes to VASP/VASPsol.

Thanks

Applied external potential on the adsorption and surface diffusion for metal electrodes.

From the paper arXiv:1601.03346. VASPsol can applied external potential on surface. But I can find very little introduction on how to do it.

In JDFTx there are parameters control the external potential. Does VASPsol has similar parameters?

fluid-cation K+ 0.1
fluid-anion F- 0.1
target-mu -0.163167 # U = 0.00 V wrt SHE (4.44 V)
#target-mu -0.232991 # U = +1.90 V wrt SHE (6.34 V)

How to calculate the grand potential of charged system with vaspsol?

It seem that the correct formula is
Ω = F+(N-N0)U
F = E(DFT) + (N0 - N)
FERMI_SHIFT
where N is the number of electrons of charged system, N0 is the number of electrons of neutral system. U is the work function.
However, dF/dN = E(fermi) - FERMI_SHIFT = E(fermi) + E(vac), Here E(fermi) and E(vac) are fermi energy and vacuum level, it is not -U (-U = E(fermi) - E(vac) ).

EC-IR simulation

Hi Prof. Hennig,
I’m a loyal user of VASPsol. It has been used to simulate PZCs of Au, Ag and Cu surfaces, and the double layer capacitance of Pt(111) surface with surface charge method (Tag NELECT in VASP). They agree well with experiment measurements.

Recently, I try using VASPsol with DFPT to simulate the electrochemical IR spectrum (EC-IR includes vibrational frequencies and IR intensity derived from Born effective charge) of CO adsorbed at metal surface but fail. The DFPT and implicit solvation model aren’t compatible, are they, why and what should I do to obtain the EC-IR based on VASPsol? Could you help me to solve these problem? Many thanks! The details of simulation are listed below.

As shown in Table 1, I use the dielectric constant of water (78.4) and 0.1 M ion strength to simulate the solvent layer. Based on JPCC2016 of Sautet group1, the size of solvent layer is set to 95 Å and tested in Fig.1.
Before using the implicit solvation model, I respectively used small displacement method (SD) and DFPT to calculate the frequencies of C-O stretching (first frequency) and Pt-C stretching (second frequency) in a vacuum system (Table 2). Both methods are equivalent. In the solvation system, their equivalence is lost, only SD method can obtain vibrational frequencies exactly corresponding to C-O stretching and Pt-C stretching, but DFPT not.

table1
fig1

table2

Reference

  1. S. N. Steinmann and P. Sautet, The Journal of Physical Chemistry C, 2016, 120, 5619-5623.

Best regards,
Yuan

VASP6 with VASPsol add-on - compile error

Has anyone compiled VASP6 with VASPsol? With basically the same compile script that I'm able to compile VASP5.4.4 with VASPsol does not work with VASP6. I'm getting the following error:

solvation.o: In function pot_k_mp_vcorrection_': solvation.f90:(.text+0x1152f): undefined reference to m_sum_s_'
solvation.f90:(.text+0x11698): undefined reference to `m_sum_s_'
make[2]: *** [vasp] Error 1

I'm using (I have also tried other versions of these compiler with the same issue):
intel/16.0
mkl/16.0
openmpi/1.10

vaspsol setting for the enhanced sampling

Dear developers,
I employed vaspsol to screen the charged surface on constrained MD simulation , which is set by LSOL = .True. Lambda_DK = 3.04. However, the mean force for the constraint is not changed although the surface charge is different. It's obvious that this doesn't make sense. I would to know if any setting is missed for the screen effect on MD simulation.
Thanks for your time.

VASPsol convergence when involving I (Iodine)

Hello,

During my calculations in using VASPsol, it seems that when involve I (Iodine), it is extremely to to converge. I wonder if this is related to the implementation of VASPsol? Thanks for your time.

Depth of the solvation medium

Dear Altruists

I have a rough metal surface on top of which I am considering the PCM solvation modeled by the VASPsol. I would like to know that is it possible to control the uniform depth or thickness of the PCM electrolyte on top of the metal surface by the VASPsol. If so, which tag should I consider for changing the depth or thickness of the electrolyte?

Thank you very much for your kind dedication.

Arindam Sannyal

Latest VASP version 5.4.1

The new version of VASP is out. We need to update the installation procedure and keep the patch files only for the previous VASP versions.

mgrid.f90 error

Hello,
I'm running into and error when trying to run a VASPsol gpu job with Vasp 6.2.0. I've followed the compilation method spelled out in #50 and everything seems to compile fine. However when I go to run a job I get to the first calculation step and retrieve the following error which then causes my job to terminate.

FATAL ERROR: data in PRESENT clause was not found on device 1: name=c host:0x5b6acd50
file:/home/sawilso6/bin/vasp_compiles/vasp_6/vasp.6.2.0_vaspSOL/build/std/mgrid.f90 rc_add line:1753

Thank you for your help.

Best,
Steve

Can JDFTx GLSSA13 parameters be directly imported into VASPsol?

My understanding is that VASPsol closely follows the PCM implementation used in JDFTx for their GLSSA13 model. Additionally, the parameters controlling VASPsol are largely analogous to parameters outlined in the JDFTx source code. Does this mean we can directly use parameters from JDFTx to simulate other solvents, such as methanol, using VASPsol? For example, JDFTx specifies the following parameters (translated to VASPsol input keywords) for methanol:

EB_K = 32.66
SIGMA_K = 0.6
NC_K = 6.5e-4
TAU = -5.2e-6

I realize you have said before that NC_K should not be modified in most cases. Is it then safe to use the remaining three parameters without modification?

Optimization of surface adsorbate with VASPsol

Dear all,

First of all i would like to thank you for the excellent work done with the VASPsol patch, it is such a useful tool.
I have a question concerning the optimization procedure of adsorbates on surfaces.
From the usage.md i know that the proper workflow requires a vacuum calculation and a following implicit solvent one, starting from the vac wavecar and vac contcar. In case of an SPE calculation everything is fine. But, for an optimization, what kind of calculation should i perform as second? Can i limit to an SPE on the vacuum relaxed structure? Or should i fully (nucleas+electrons) re-optimize the vacuum result including the implicit solvent ?

Best wishes,
Silvio

1. Free Energy or Energy calculated at T=0 K in the OUTCAR calculated using VASPsol.

I am an amateur in DFT calculations.
As far as I know, we calculate the ground state energy of the system inside the simulation cell at T=0 K by minimizing the electron density. But when we use VASPsol, you have reported in the paper the minimization of 'free energy' that involves the ion concentration term in linearized Possion-Boltzmann equation. This ion concentration term is dependent on temperature T. What is this temperature? What is the energy (energy(sigma->0)) found in the OUTCAR of VASPsol calculation?

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.