Difference between revisions of "Explanations for Si sc pulse input files"

Revision as of 11:27, 14 June 2017

Unit system

Hartree atomic units are used in this calculation by default.

&calculation

 &calculation
   calc_mode = 'GS_RT'
 /

The variable calc_mode is set to be 'GS_RT' mode, which corresponds to execute the ground state (GS) and real-time (RT) calculation with single calculation task.

&control

 &control
  sysname = 'Si'
 /

The variable sysname is set to be 'Si', which is used as the filename prefix of the outputs.

&system

 &system
   iperiodic = 3
   al = 10.26d0,10.26d0,10.26d0
   isym = 8 
   crystal_structure = 'diamond'
   nstate = 32
   nelec = 32
   nelem = 1
   natom = 8
 /

iperiodic = 3 indicates that three dimensional periodic boundary condition (bulk crystal) is assumed. al = 10.26d0, 10.26d0, 10.26d0 specifies the lattice constans of the unit cell crystaline. The variable isym indicates the symmetry in the unit cell. Considering the bulk silicon crystal with the applied electric field parallel to the one lattice axis, isym = 8 is preferred to speed up the calculation. For more infomation, see Symmetry group of crystaline. crystal_structure = 'diamond' indicate the crystal structure of the considered material. nstate = 32 indicates the number of Kohn-Sham orbitals to be solved. nelec = 32 indicate the number of valence electrons in the system. nelem = 1 and natom = 8 indicate the number of elements and the number of atoms in the system, respectively.

&pseudo

 &pseudo
   iZatom(1)=14
   pseudo_file(1) = './Si_rps.dat'
   Lloc_ps(1)=2
 /

iZatom(1) = 14 indicates the atomic number of the element 1. pseudo_file(1) = 'Si_rps.dat' indicates the filename of the pseudopotential of element 1. Lloc_ps(1) = 1 indicate the angular momentum of the pseudopotential that will be treated as local.

&functional

 &functional
   xc ='TBmBJ'
   cval = 1d0
 /

xc ='TBmBJ' specifies the type of exchange correlation potential. The TBmBJ indicates a meta-generalized gradient approximation proposed by Tran and Blaha Phys. Rev. Lett. 102, 226401 (2009). cval specifies the additional parameter of the TBmBJ potential. In the case of the silicon, cval = 1d0 is prefered to reproduce the experimentally mesured optical constants.

&rgrid

 &rgrid
   num_rgrid = 12,12,12
 /

num_rgrid=12,12,12 specifies number of the real space grids for single crystal calculation.

&kgrid

 &kgrid
   num_kgrid = 4,4,4
 /

num_kgrid=4,4,4 specifies number of the k-space grids for single crystal calculation.

&tgrid

 &tgrid
  nt=3000
  dt=0.16  
 /

dt=0.16 sets the time step of the time-evolution calculations. Nt=3000 indicates the number of time steps in the calculation.

&propagation

 &propagation
   propagator='etrs'
 /

propagation='etrs' specifies the numerical mathod of the time evolution of the wave function.

&scf

 &scf
   ncg = 5
   nscf = 120
 /

ncg = 5 indicates the number of conjucate gradient step in the single scf calculation, and nscf = 120 specifies the number of the SCF step.

&emfield

 &emfield
   trans_longi = 'tr'
   ae_shape1 = 'Acos2'
   rlaser_int1 = 1d14
   pulse_tw1 = 441.195136248d0
   omega1 = 0.05696145187d0
   epdir_re1 = 0.,0.,1.
 /

ae_shape1 = 'Acos2' specifies the pulse shape of the electric field, having cos-square envelope. laser_int1 specifies maximum intensity of the applied electric field in unit of W/cm^2. epdir_re1 = 0.d0,0.d0,1.d0 identifies the unit vector of polarization direction. Specifying the real part of the polarization vector by 'epdir_re1', linear polarization is assumed. Using both the real ('epdir_re1') and imaginary ('epdir_im1') parts of the polarization vector, circularly (and general ellipsoidally) polarized pulses may also be described.

omega1 specifies photon energy (frequency multiplied with hbar). pulse_tw1 sets the pulse duration. Note that it is not FWHM but a full duration of the sine-square envelope. phi_cep1 specifies the carrier envelope phase of the pulse. It is possible to take two pulses simultaneously to simulate pump-probe experiments, adding information for two pulses. The time delay can be indicated using the variable 't1_t2'.