scnc
/
quantum-espresso
mirror of https://gitlab.com/QEF/q-e.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Documentation updated (modenum, nat_todo, nrapp) 

git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@3367 c92efa57-630b-4861-b058-cf58834340f0
This commit is contained in:
giannozz 2006-09-07 17:17:00 +00:00
parent 2145c19e68
commit bfbf72840e
2 changed files with 124 additions and 116 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

235
Doc/INPUT_PH
View File

@ -1,116 +1,121 @@
c phonon input cards are as follows:
c
c card 1 title_ph (a line that is reprinted on output)
c
c namelist "inputph"
c
cc variable meaning default value
c
c amass(ntyp) atomic mass [amu] of each atomic type 0
c If not specified, masses are read from data file
c
c outdir scratch directory './'
c
c prefix prepended to input/output filenames 'pwscf'
c must be the same used in the calculation
c of unperturbed system
c
c niter_ph maximum number of iterations in a scf step 50
c
c tr2_ph threshold for selfconsistency 1e-10
c
c alpha_mix(niter) mixing factor (for each iteration) for alpha_mix(1)=0.7
c updating the scf potential:
c vnew(in) = alpha_mix*vold(out) + (1-alpha_mix)*vold(in)
c nmix_ph number of iterations used in potential mixing 4
c
c iverbosity iverbosity=0 short output, iverbosity=1 verbose output 0
c
c reduce_io reduce I/O to the strict minimum .false.
c
c max_seconds maximum allowed run time before the job stops smoothly 1.d7
c
c fildyn file where the dynamical matrix is written 'matdyn'
c
c fildrho file where the charge density responses are written ' '
c
c filelph file where electron-phonon matrix elements are written ' '
c
c fildvscf file where the the potential variation is written ' '
c (for later use in electron-phonon calculation)
c
c
c epsil if .true. in a q=0 calculation for a non metal the .false.
c macroscopic dielectric constant of the system is
c computed. Do not set epsil to .true. if you have a
c metallic system or q/=0: the code will complain and stop
c
c trans if .true. the phonons are computed .true.
c If trans.and.epsil effective charges are calculated
c
c lraman if .true. calculate nonresonant Raman coefficients .false.
c using second-order response as in:
c M. Lazzeri and F. Mauri, Phys. Rev. Lett. 90, 036401 (2003)
c Optional variables for Raman:
c eth_rps threshold for calculation of Pc R |psi> 1.0d-9
c eth_ns threshold for non-scf wavefunction calculation 1.0e-12
c dek delta_xk used for wavefunction derivation wrt k 1.0e-3
c
c recover if .true. restart from an interrupted run .false.
c
c elph if .true. electron-phonon lambda coeffs are computed .false.
c For metals only, requires gaussian smearing.
c If elph.and.trans, the lambdas are calculated in the same
c run, using the same k-point grid for phonons and lambdas
c If elph.and..not.trans, the lambdas are calculated using
c previously saved DeltaVscf in fildvscf, previously saved
c dynamical matrix, and the present punch file. This allows
c the use of a different (larger) k-point grid.
c
c zue if .true. in a q=0 calculation for a non metal the
c effective charges are computed from the phonon
c density responses. Note that if trans.and.epsil
c effective charges are calculated using a different
c algorithm. The results should be the same within
c numerical noise.
c
c fpol if .true. calculate dynamic polarizabilities .false.
c ( experimantal stage, see example33 for calculation
c of methane )
c
c lnscf If .TRUE. the run makes first a nscf calculation .false.
c
c ldisp If .TRUE. the run calculates phonons for a grid of .false.
c q-points specified by nq1, nq2, nq3 - for direct
c calculation of the entire phonon dispersion
c (ldisp and modenum are not compatible)
c
c nq1,nq2,nq3 parameters of the Monkhorst-Pack grid (no offset) used 0,0,0
c when ldisp=.true. Same meaning as for nk1, nk2, nk3
c in the input of pw.x
c
c maxirr maximum number of irreducible representation to be 0 i.e. all
c computed in a single run
c
c nrapp the representations to do 0
c (nrapp and nat_todo are not compatible)
c
c nat_todo number of atom to be displaced 0
c (nrapp and nat_todo are not compatible)
c
c modenum for single mode calculation -1
c If not set here, will be read from file .save
c (if modenum.ne.0 => nat_todo = 0; nrapp = 1; list(1) =
c modenum)
c
phonon input cards are as follows:
card 1 title_ph (a line that is reprinted on output)
namelist "inputph"
variable meaning default value
amass(ntyp) atomic mass [amu] of each atomic type 0
If not specified, masses are read from data file
outdir scratch directory './'
prefix prepended to input/output filenames 'pwscf'
must be the same used in the calculation
of unperturbed system
niter_ph maximum number of iterations in a scf step 50
tr2_ph threshold for selfconsistency 1e-10
alpha_mix(niter) mixing factor (for each iteration) for alpha_mix(1)=0.7
updating the scf potential:
vnew(in) = alpha_mix*vold(out) + (1-alpha_mix)*vold(in)
nmix_ph number of iterations used in potential mixing 4
iverbosity iverbosity=0 short output, iverbosity=1 verbose output 0
reduce_io reduce I/O to the strict minimum .false.
max_seconds maximum allowed run time before the job stops smoothly 1.d7
fildyn file where the dynamical matrix is written 'matdyn'
fildrho file where the charge density responses are written ' '
filelph file where electron-phonon matrix elements are written ' '
fildvscf file where the the potential variation is written ' '
(for later use in electron-phonon calculation)
epsil if .true. in a q=0 calculation for a non metal the .false.
macroscopic dielectric constant of the system is
computed. Do not set epsil to .true. if you have a
metallic system or q/=0: the code will complain and stop
trans if .true. the phonons are computed .true.
If trans.and.epsil effective charges are calculated
lraman if .true. calculate nonresonant Raman coefficients .false.
using second-order response as in:
M. Lazzeri and F. Mauri, Phys. Rev. Lett. 90, 036401 (2003)
Optional variables for Raman:
eth_rps threshold for calculation of Pc R |psi> 1.0d-9
eth_ns threshold for non-scf wavefunction calculation 1.0e-12
dek delta_xk used for wavefunction derivation wrt k 1.0e-3
recover if .true. restart from an interrupted run .false.
elph if .true. electron-phonon lambda coeffs are computed .false.
For metals only, requires gaussian smearing.
If elph.and.trans, the lambdas are calculated in the same
run, using the same k-point grid for phonons and lambdas
If elph.and..not.trans, the lambdas are calculated using
previously saved DeltaVscf in fildvscf, previously saved
dynamical matrix, and the present punch file. This allows
the use of a different (larger) k-point grid.
zue if .true. in a q=0 calculation for a non metal the
effective charges are computed from the phonon
density responses. Note that if trans.and.epsil
effective charges are calculated using a different
algorithm. The results should be the same within
numerical noise.
fpol if .true. calculate dynamic polarizabilities .false.
( experimantal stage, see example33 for calculation
of methane )
lnscf If .TRUE. the run makes first a nscf calculation .false.
ldisp If .TRUE. the run calculates phonons for a grid of .false.
q-points specified by nq1, nq2, nq3 - for direct
calculation of the entire phonon dispersion.
The pw.x data file should not be produced using
"calculation='phonon'" in this case
nq1,nq2,nq3 parameters of the Monkhorst-Pack grid (no offset) used 0,0,0
when ldisp=.true. Same meaning as for nk1, nk2, nk3
in the input of pw.x
nrapp choose the subset of irreducible representations (irreps)
for which the linear response calculation is performed:
"nrapp" irreps, specified in input (see below) are used.
Default: nrapp=0, i.e. use all irreps . nrapp must be <= 3*nat
Do not specify "nat_todo" together with "nrapp"
maxirr perform calculations only up to the first "maxirr" irreps.
Default: maxirr=0, i.e. use all irreps. maxirr must be <= 3*nat
Do not specify "nat_todo" or "nrapp" together with "maxirr"
nat_todo choose the subset of atoms to be used in the linear response
calculation: "nat_todo" atoms, specified in input (see below)
are displaced. Default: nat_todo=0, i.e. displace all atoms
Do not specify "nrapp" together with "nat_todo"
End of namelist "inputph"
card 2 xq(3) the phonon wavevector; must be equal to the one used
in the non-selfconsistent calculation (not read if
ldisp is true).
card 3 list(na), na = 1, nrapp
contains the list of indices of irreps used in the
calculation if "nrapp" is specified
OR:
card 3 atomo(na), na = 1, nat_todo
contains the list of indices of atoms used in the
calculation if "nat_todo" is specified
c End of namelist "inputph"
c
c card 2 xq(3) the phonon wavevector; must be equal to the one used
c in the non-selfconsistent calculation (not read if
c ldisp is true).
c
c card 3 list(na), na = 1, nrapp
c
c card 3 atomo(na), na = 1, nat_todo
c

5
Doc/INPUT_PW
View File

@ -874,7 +874,10 @@ press_cov_thr REAL ( default = 0.5D0 Kbar )
&PHONON ( only in calculation = 'phonon' )
modenum INTEGER ( default = 0 )
for single-mode phonon calculation
for single-mode phonon calculation : modenum is the index of the
irreducible representation (irrep) into which the reducible
representation formed by the 3*nat atomic displacements are
decomposed in order to perform the phonon calculation
xqq(3) REAL
q-point (units 2pi/a) for phonon calculation