Actually, I have recently published a paper on melting behavior of some graphene quantum dots containing 6 and 10 atoms, but I had to write on my own more than 800 program lines in bash and fortran for such small-sized structures to get my desired results. And it is based on such an experience that I believe QE is not that sophisticated in DFMD calculations of large, non-solid, molecular structures. Even if you are a highly-experienced programmer, you cannot get the desired results conclusively as the number of atoms increases to form a solution.
Regards, Ashkan *Ashkan Shekaari* Plasma Physics Research Center Science and Research Branch I A U, 14778-93855 Tehran, Iran. On Sat, May 6, 2017 at 1:00 PM, ashkan shekaari <[email protected]> wrote: > Dear Mattioli, > > Water is not a crystal and therefore it cannot be treated as a periodic > structure. > The aforementioned example provides only a naive picture of water as an > aqueous compartment leading to a crude approximation of the given system as > well. Moreover, QE is not best suited for DFMD simulations of large, > nonperiodic, liquid systems. > > Regards, > Ashkan > > *Ashkan Shekaari* > Plasma Physics Research Center > Science and Research Branch > I A U, 14778-93855 Tehran, Iran. > > On Sat, May 6, 2017 at 12:21 PM, Giuseppe Mattioli < > [email protected]> wrote: > >> >> Dear David and Ashkan >> >> > You cannot simulate the system as periodic. You have to use a large >> super >> > cell with vacuum spaces along the three spatial directions. Moreover, >> you >> > can simulate the solution only via importing a large number of atomic >> > positions of the order of 25000 at least, which does not seem to be >> > feasible by QE even in case of using random generators unless you have a >> > supercomputer facility. >> >> This comments are misleading, if not wrong in some cases. >> First of all you can for sure perform ab initio molecular dynamics >> (aimd) simulations of a water solution in a periodic box, and you do >> not need a huge number of atoms. One of the cp.x examples provides >> starting points, see here >> /Your_Path_To_Espresso_6.1/Examples/CPV/example04 >> >> >> Does anyone know of examples, benchmarks, or recommendations? Would >> the >> >> X.blyp-van.ak pseudopotentials be appropriate? Any guidance or advice >> >> about parameter settings for this application would be much appreciated >> >> Yes, you can use ultrasoft pseudopotentials such as X.blyp-van.ak. >> They used to be old and well-tested vanderbilt ultrasoft >> pseudopotentials generated many years ago by Axel Kohlmeyer (ak). But >> if you are not familiar with this terminology (ultrasoft, >> norm-conserving, van, mt, rrkj, paw, ...) you should start with some >> tutorial on pseudopotentials, because you must be sure to use >> converged plane-wave and density cutoffs for a given set of >> pseudopotentials. >> >> There is a very large number of options that must be set in the cp.x >> input. I cannot review them here. But there are also tons of >> literature on aimd simulations of water solutions, and you will easily >> find something that will help to choice a lot of parameters (box >> dimensions, NVT, NVP, NVE dynamics, thermostats, ...) >> >> > Indeed, QE is not best suited for MD simulations and I strongly >> recommend >> > the gromacs package. >> >> Of course Ashkan might be right on a couple of points: if you have >> very big molecules weakly interacting with the solvent, and you are >> interested in the morphology of the solute only, then you have to >> perform very long md simulations and to use a large number of water >> molecules, and this might be costly and time-consuming if performed at >> an aimd level. If you want to study proton exchanges, reactions, >> dipole dynamics, ... you need aimd. You surely know whether your >> scientific task strictly requires aimd rather than model-potential md. >> >> HTH >> Giuseppe >> >> Quoting ashkan shekaari <[email protected]>: >> >> > Indeed, QE is not best suited for MD simulations and I strongly >> recommend >> > the gromacs package. >> >> > You cannot simulate the system as periodic. You have to use a large >> super >> > cell with vacuum spaces along the three spatial directions. Moreover, >> you >> > can simulate the solution only via importing a large number of atomic >> > positions of the order of 25000 at least, which does not seem to be >> > feasible by QE even in case of using random generators unless you have a >> > supercomputer facility. >> > >> > On May 6, 2017 9:09 AM, "D J Anick" <[email protected]> wrote: >> > >> >> Hello plane wavers, >> >> >> >> I am interested in using QE for a molecular dynamics simulation of an >> >> aqueous solution containing a solute, modeling it as a 3-D periodic >> cell. >> >> Principal questions would be about solvation shell geometries, >> distribution >> >> of configurations adopted by the solute, and H-bond duration / >> stability. >> >> >> >> Does anyone know of examples, benchmarks, or recommendations? Would >> the >> >> X.blyp-van.ak pseudopotentials be appropriate? Any guidance or advice >> >> about parameter settings for this application would be much >> appreciated. >> >> >> >> Thank you in advance, >> >> David Anick >> >> david.anick###rcn.com >> >> _______________________________________________ >> >> Pw_forum mailing list >> >> [email protected] >> >> http://pwscf.org/mailman/listinfo/pw_forum >> >> >> >> >> -- >> ******************************************************** >> - Article premier - Les hommes naissent et demeurent >> libres et égaux en droits. Les distinctions sociales >> ne peuvent être fondées que sur l'utilité commune >> - Article 2 - Le but de toute association politique >> est la conservation des droits naturels et >> imprescriptibles de l'homme. Ces droits sont la liberté, >> la propriété, la sûreté et la résistance à l'oppression. >> ******************************************************** >> >> Giuseppe Mattioli >> CNR - ISTITUTO DI STRUTTURA DELLA MATERIA >> v. Salaria Km 29,300 - C.P. 10 >> I 00015 - Monterotondo Stazione (RM) >> Tel + 39 06 90672836 - Fax +39 06 90672316 >> E-mail: <[email protected]> >> http://www.ism.cnr.it/en/staff/giuseppe-mattioli/ >> ResearcherID: F-6308-2012 >> >> >> _______________________________________________ >> Pw_forum mailing list >> [email protected] >> http://pwscf.org/mailman/listinfo/pw_forum >> > >
_______________________________________________ Pw_forum mailing list [email protected] http://pwscf.org/mailman/listinfo/pw_forum
