Top Page > Browsing
 convergence in electric field Date: 2023/01/29 02:01 Name: maedeh   hi everyonei apply an external electric field perpendicular to a nanoribbon and I used :scf.Electric.Field  0.0 0.0 0.5*10^(-19)  # default=0.0 0.0 0.0 (GV/m)at the first step, I try to relax my system but it doesn't converge. The strength of the electric field is 0.5V/A˚ do i have to modify something else?
 Page: [1]

 Re: convergence in electric field ( No.1 ) Date: 2023/01/30 20:48 Name: Yung-Ting Lee Dear maedeh,(1) 0.5 V/angstrom = 0.5 *10^(-9) GV/angstrom = 0.5 *10^(-9+10) GV/m = 5.0 (GV/m).For applying an external electric field perpendicular to a nanoribbon, one may set a command line as below.scf.Electric.Field  0.0 0.0 5.0  # applying an external electric field in the z-axis (unit=GV/m)(2) The relaxation may not converge at less than 100 steps because electron density will be slightly changed due to an applied electric field.The number of relaxation steps could be larger than 100 to reach energy and force convergences.One may restart the relaxations or increase the number of steps to find the optimization structure with an applied electric field.On the other hand, sometimes, atoms will be driven to move due to a change in electron density by an external electric field.Therefore, fixing several atoms may be necessary in order to help the structure reach convergence easily. Otherwise, atoms move all the time in the unit cell.Best regards,Yung-Ting Lee Re: convergence in electric field ( No.2 ) Date: 2023/02/06 14:26 Name: maedeh  i did what you say and also I decrease the electric field strength but the system didn't converge the system after 119 step is:https://s2.uupload.ir/files/untitled_dk9n.png Re: convergence in electric field ( No.3 ) Date: 2023/02/06 14:34 Name: maedeh  this is my *.dat file:scf.Electric.Field  0.0 0.2 0.0  # default=0.0 0.0 0.0 (GV/m)Atoms.Number                      72Atoms.SpeciesAndCoordinates.Unit  FRACAtoms.UnitVectors.Unit            Angscf.XcType                    GGA-PBEscf.SpinPolarization          offscf.ElectronicTemperature    300.0scf.energycutoff              220.0scf.maxIter                  100scf.EigenvalueSolver          bandscf.Kgrid                    6  1  1scf.Mixing.Type              rmm-diiskscf.Init.Mixing.Weight        0.05scf.Min.Mixing.Weight        0.01scf.Max.Mixing.Weight        0.30scf.Mixing.History            25scf.Mixing.StartPulay        15scf.criterion                1.0e-7MD.Type                      rfc5MD.maxIter                    1000MD.TimeStep                  1.0MD.Opt.criterion              0.0003 Re: convergence in electric field ( No.4 ) Date: 2023/02/09 02:44 Name: Yung-Ting Lee Dear maedeh,(1) RFC5 is a method for optimizing cell vectors and atomic coordinates simultaneously.The "opt" for geometry optimization may be what you need.#############MD.Type  opt#############(2)For constrained relaxation, 1 and 0 stand for fixing and relaxing, respectively.The reference for "Constrained relaxation" can be found in " https://www.openmx-square.org/openmx_man3.8/node53.html ".In the input that you provided, all atoms are fixed. Due to "rfc5", I guess that only cell vectors change at iteration in your calculation.Following the input, I recommend the command lines listed below for fixing Sn atoms in the bottom only and relaxing other atoms.##########################Please update these command lines and try again.Best regards,Yung-Ting Lee Re: convergence in electric field ( No.5 ) Date: 2023/02/13 23:33 Name: maedeh  thank you

 Page: [1]

 Thread Title (must) Move the thread to the top Your Name (must) E-Mail (must) hidden display URL Password (used in modification of the submitted text) Comment (must) Save Cookie