Program babel

Namelist Input ↑

anisotropic_elasticity ↑

(type: logical , default: .false. , mandatory: no )
If anisotropic_elasticity=.true., the elastic constants are given for a crystal of any symmetry. You need to define all non null elements of the 6x6 matrix corresponding to elastic constants in Voigt notation.

See also: CVoigt , cubic_elasticity , hexagonal_elasticity


CVoigt ↑

(type: real, dim(6,6) , default: 0. , mandatory: no )
6x6 matrix defining the elastic constants in Voigt notation when anisotropic_elasticity=.true.. All non null elements need to be given. By defaults, the program assumes that the elastic constants are in GPa.

See also: anisotropic_elasticity , CVoigt_noise


cubic_elasticity ↑

(type: logical , default: .false. , mandatory: no )
If cubic_elasticity=.true., the elastic constants are given for a crystal of cubic symmetry. You need then to give the three constants C11, C12, and C44. Make sure the crystal is oriented in the cubic axes corresponding to the elastic constants, or use anisotropic_elasticity=.true. otherwise.

See also: C11, C12, C44, C13, C33 , anisotropic_elasticity , hexagonal_elasticity


hexagonal_elasticity ↑

(type: logical , default: .false. , mandatory: no )
If hexagonal_elasticity=.true., the elastic constants are given for a crystal of hexagonal symmetry. You need then to give the five constants C11, C12, C44, C13 and C33. Make sure the crystal is oriented so that the z coordinate corresponds to the 6-fold symmetry axis, or use anisotropic_elasticity=.true. otherwise.

See also: C11, C12, C44, C13, C33 , anisotropic_elasticity , cubic_elasticity


C11, C12, C44, C13, C33 ↑

(type: real , default: 0. , mandatory: no )
Elastic constants used for the special cases: - cubic_elasticity=.true.: C11, C12, C44 are used - hexagonal_elasticity=.true.: C11, C12, C44, C13 and C33 are used and C66=(C11-C12)/2 By defaults, the program assumes that the elastic constants are in GPa.

See also: cubic_elasticity , hexagonal_elasticity


CVoigt_noise ↑

(type: real , default: 0. , mandatory: no )
CVoigt_noise=x will add a random noise of maximal amplitude x to all elastic constants, keeping the elastic matrix positive definite. This is the relative amplitude and the absolute amplitude is obtained by multiplying with the norm of the elastic tensor. This may be necessary for creating line defects when the crystal possesses some symmetries, because of the Stroh formalism leading to undefined solutions in that cases. CVoigt_noise=1.d-4 will generally solve the problem without perturbing the solution. Do not use a noise smaller than 1.d-6. When adding a noise, you should check that the result (the elastic energy or the Stroh matrix for instance) does not depend too much on this noise by running several times the same calculation.


strain ↑

(type: logical , default: .false. , mandatory: no )
Set the variable strain to .true. if you want to apply a homogeneous strain to the crystal in addition to the strain created by the line defects. The homogeneous strain to apply is defined either by epsi and eStrain or by tau and sStrain.

See also: eStrain , epsi , tau , sStrain , induced_homogeneous_strain


eStrain ↑

(type: real, dim(3,3) , default: 0. , mandatory: no )
Tensor defining the homogeneous strain to apply when strain=.true. The applied strain is the product of this tensor with the scalar epsi.

See also: strain , epsi


epsi ↑

(type: real , default: 1. , mandatory: no )
Scalar defining the homogeneous strain to apply when strain=.true. The applied strain is the product of this scalar with the tensor eStrain(1:3,1:3).

See also: strain , eStrain


sStrain ↑

(type: real, dim(3,3) , default: 0. , mandatory: no )
Stress tensor used to define the homogeneous strain to apply when strain=.true. The applied strain is the product of the inverse elastic constants tensor with this stress tensor multiplied by the scalar tau.

See also: strain , tau


tau ↑

(type: real , default: 1. , mandatory: no )
Scalar used to define, with the stress tensor sStrain(1:3,1:3), the homogeneous strain to apply when strain=.true. The applied strain is the product of the inverse elastic constants tensor with the stress tensor multiplied this scalar. This needs to be given in the same units as the elastic constants (GPa).

See also: strain , sStrain


induced_homogeneous_strain ↑

(type: logical , default: .true. , mandatory: no )
When induced_homogeneous_strain=.true. , the homogeneous strain induced by the line defects is applied to the crystal so as to minimize the elastic energy existing in the simulation box. For a dislocation dipole or a dislocation loop, this strain corresponds to the plastic strain introduced in the crystal when creating the dipole or the loop.

See also: strain


imm ↑

(type: integer , default: 0 , mandatory: no )
Maximal number of atom in the structure. imm needs to be defined if the input structure is duplicated to dimension the arrays. Otherwise, the actual number of atoms, read from input structure file, is used.

See also: duplicate


duplicate ↑

(type: logical , default: .false. , mandatory: no )
If duplicate=.true., the crystal is duplicated. lat(1:3) is the number of times the crystal should be duplicated in each direction given by the vectors at(1:3,i) read in the structure file. If lat(i) is negative, then the opposite of the vector at(1:3,i) is used. The maximal number of atoms in the duplicated structure needs to be set with imm and the periodicity vectors have to be defined. When the structure is also rotated, the duplication is done first and then the rotation. The periodicity vectors used for duplication are thus those before rotation.

See also: imm , lat


lat ↑

(type: integer, dim(3) , default: 1 , mandatory: no )
Number of replicas in each direction when duplicating the structure files.

See also: duplicate


rotate ↑

(type: logical , default: .false. , mandatory: no )
If rotate=.true., the crystal is rotated using rotation matrix defined in rot(1:3,1:3). The program also rotates the definition of the line-defect and the elastic constants. If both rotation and translation are performed, the crystal is first rotated and then translated.

See also: rot


rot ↑

(type: real, dim(3,3) , default: identity matrix , mandatory: no )
Rotation matrix used when rotate=.true.

See also: rotate


symmetrize ↑

(type: logical , default: .false. , mandatory: no )
If symmetrize=.true., the atomic positions are symmetrized according to symmetry operations defined in file symFile.

See also: symFile


symFile ↑

(type: string , default: '' , mandatory: yes/no )
Xml file containing definition of the symmetry operations used to symmetrize the structure.

See also: symmetrize


translate ↑

(type: logical , default: .false. , mandatory: no )
If translate=.true., the crystal is translated with the vector uTranslate(1:3) The program also translates the points definig the line-defect centers. If both rotation and translation are performed, the crystal is first rotated and then translated.

See also: uTranslate


uTranslate ↑

(type: real, dim(3) , default: 0. , mandatory: no )
Translation vector used when translate=.true. If alat is defined, uTranslate(1:3) is multiplied by alat.

See also: translate , alat


fixGravity ↑

(type: logical , default: .false. , mandatory: no )
When fixGravity=.true., a solid displacement is added to atom final positions and displacements to keep fixed gravity center


xNoise ↑

(type: real , default: 0. , mandatory: no )
If positive, add to atomic positions of the output structure a noise which amplitude is equal to xNoise multiplied by alat. The gravity center is kept fixed.

See also: alat


alat ↑

(type: real , default: 1. , mandatory: no )
Length used to scale all distances in the present input file (uTranslate, rc, ...), including the line defect definitions in other namelists (cDiso, bDislo, ...). It is also used to scale xNoise. It has no effect on the atomic configuration read in input structure file.


xImages, yImages, zImages ↑

(type: logical , default: .false. , mandatory: no )
If xImages=.true. (yImages or zImages), periodic boundary conditions are assumed in the corresponding direction and periodic images are considered for all defined line-defects. The number of periodic images considered in each direction is controlled by nxImages (nyImages or nzImages).

See also: nxImages, nyImages, nzImages


nxImages, nyImages, nzImages ↑

(type: integer , default: 10 , mandatory: no )
Number of images to consider in each direction with periodic boundary conditions. The default value is large enough to build an initial configuration but should be increased for energy calculation. For a small unit-cell like the ones used in ab-initio calculation, you probably need to increase this value to 100. A way to check convergence is to run several simulations. As the point where periodicity is enforced is randomly chosen, different simulations will lead to different results for elastic energy. Nevertheless, differences should be small (less than 0.1 meV) and correspond to calculation convergence.

See also: xImages, yImages, zImages


clipAtom ↑

(type: logical , default: .false. , mandatory: no )
If clipAtom=.true., periodic boundary conditions are applied to atom coordinates to bring them back in the primitive unit cell. The periodicity vectors have to be defined.


clipDisplacement ↑

(type: logical , default: .false. , mandatory: no )
If clipDisplacement=.true., periodic boundary conditions are applied to atom displacements. The periodicity vectors have to be defined.


remove_cut ↑

(type: logical , default: .false. , mandatory: no )
If remove_cut=.true. (or add_cut=.true.), atoms located inside the cut associated with the creation of a single dislocation or a dislocation dipole are removed or inserted (Volterra process). This also works with dislocation loops.

See also: Dislo , DDipole , Loops


add_cut ↑

(type: logical , default: .false. , mandatory: no )
Same control parameter as remove_cut

See also: remove_cut


max_Euler ↑

(type: integer , default: 0 , mandatory: no )
EXPERIMENTAL: Eulerian or Lagrangian coordinates can be used. In Lagrangian coordinates, elastic fields (displacement, stress) are calculated at point x0 corresponding to the initial coordinates. In Eulerian coordinates, fields are calculated at point x = x0 + u(x), where u(x) is the displacement calculated at the final point. A self-consistency loop is used for Eulerian coordinates, with a maximal number of iterations given by max_Euler. Default value max_Euler=0 corresponds to Lagrangian coordinates and should be preferred.

See also: delta_Euler


delta_Euler ↑

(type: real , default: 1.d-5 , mandatory: no )
When Eulerian coordinates are used, delta_Euler controls the absolute convergence of the displacement. The self-consistency loop will stop when u(x) is smaller than (x-x0) (in absolute value).

See also: max_Euler


rc ↑

(type: real , default: 1. , mandatory: no )
Cutoff radius corresponding to the core of the line defects. This is useful for energy calculations and to define atoms inside the core (see out_core). If alat is defined, rc is multiplied by alat.

See also: out_core , alat


factorE ↑

(type: real , default: 0.00624150947961 , mandatory: no )
Conversion factor to transform stress * volume in energy. The default value assumes that elastic constants are in GPa and distances in Angstroms, so as to obtain energies in eV. This is used to calculate elastic energy and with out_Ebinding to calculate impurity binding energies

See also: out_Ebinding


inpFile ↑

(type: character string , default: '-' , mandatory: yes/no )
Name of the structure file used as input. If input='-', the structure is read from keyboard. The file formats are defined by inpXyz | inpCfg | inpGin | inpSiesta | inpNDM | inpLmpDump | inpLmpData | inpPoscar

See also: inpXyz, inpCfg, inpGin, inpSiesta, inpNDM, inpLmpDump, inpLmpData, inpPoscar


inpXyz, inpCfg, inpGin, inpSiesta, inpNDM, inpLmpDump, inpLmpData, inpPoscar ↑

(type: logical , default: .false. , mandatory: one should be set to .true. )
If inpXyz=.true. the input structure file is given in xyz format.

See also: Structure file formats , inpFile


nTypes ↑

(type: integer , default: 0 , mandatory: no )
Number of different atom types. This is necessary only if you want to define the label(:) property for each atom and this cannot be obtained from the input structure file. Otherwise, the number of different atom types should be set automatically after reading the input structure file.

See also: Structure file formats , label


label ↑

(type: character(len=5), dim(360) , default: (/ A, B, ... /) , mandatory: no )
Label corresponding to each atom type if necessary and it cannot be obtained from the input structure file.

See also: nTypes , Structure file formats


mass ↑

(type: real, dim(360) , default: 0. , mandatory: no )
Mass corresponding to each atom type if necessary and it cannot be obtained from the input structure file.

See also: nTypes , Structure file formats


at ↑

(type: real, dim(3,3) , default: 0. , mandatory: no )
at(1:3,1), at(1:3,2), at(1:3,3): 3 periodicity vectors The program tries first to read them in the input file and then in the structure file. If vectors are defined in both files, the ones defined in the structure file will be used. If alat is defined, at(1:3,1:3) is multiplied by alat.

See also: inpFile , alat


displacementFile ↑

(type: character string , default: '' , mandatory: no )
If it is defined, a displacement vector is read in this file for each atom and is added to atom position. The displacement can be multiplied by displacementFileFactor. The format of this file is xyz, as generated by programs babel or displacement. The displacement vector should be in columns 5-7, just after atom positions.

See also: displacementFileFactor , Structure file formats


displacementFileFactor ↑

(type: real , default: 1. , mandatory: no )
Factor to multiply displacement read in file displacementFile before adding it to atom positions.

See also: displacementFile


outFile ↑

(type: character string , default: '-' , mandatory: no )
Name of the structure file where to save the output structure which auxiliary properties set for each atom (displacement, stress, ...) If outFile="-", the output file is printed on screen. The file format is defined by outXyz | outCfg | outOnlyAtoms ...

See also: outXyz, outCfg, outOnlyAtoms, outGin, outSiesta, outLmpDump, outLmpData, outPoscar , initial , out_alat


outXyz, outCfg, outOnlyAtoms, outGin, outSiesta, outLmpDump, outLmpData, outPoscar ↑

(type: logical , default: .false. , mandatory: no )
Format of the structure file used for output.

See also: Structure file formats , outFile


initial ↑

(type: logical , default: .false. , mandatory: no )
If initial=.true., the coordinates of the input structure, i.e. before creation of the line defects, are used on output.

See also: outFile


out_alat ↑

(type: real , default: 1. , mandatory: no )
When the output structure is written with xyz or cfg file format, atomic positions and periodicity vectors can be scaled by a lattice parameter given in out_alat.

See also: outFile , Structure file formats


out_neighbours ↑

(type: logical , default: .false. , mandatory: no )
Print in output structure file the number of neighbours for each atom. The neighbourhood sphere is defined by the radius rNeigh.

See also: outFile , rNeigh , max_nNeigh


rNeigh ↑

(type: real , default: -1 , mandatory: yes if out_neighbours=.true. )
Radius of the neighbourhood sphere used to look for atom neighbours. If alat is defined, rNeigh is multiplied by alat.

See also: out_neighbours , alat


max_nNeigh ↑

(type: integer , default: 16 , mandatory: no )
Maximal number of neighbours for an atom. This is used to dimension arrays when atom neighbourhoods are computed (out_neigbours=.true.).

See also: out_neighbours


out_neighbours ↑

(type: logical , default: .false. , mandatory: no )
Print in output structure file the atom id.

See also: outFile


out_displacement ↑

(type: logical , default: .false. , mandatory: no )
If out_displacement=.true., the displacement defined on each atom is printed in output structure file.

See also: outFile


out_elasticStrain ↑

(type: logical , default: .false. , mandatory: no )
If out_elsaticStrain=.true., the elastic strain calculated on each atomic position is printed in output structure file.

See also: outFile


out_stress ↑

(type: logical , default: .false. , mandatory: no )
If out_stress=.true., the stress tensor calculated on each atomic position is printed in output structure file.

See also: outFile


out_pressure ↑

(type: logical , default: .false. , mandatory: no )
If out_pressure=.true., the pressure calculated on each atomic position is printed in output structure file. P = -1/3 Trace( stress )

See also: outFile , out_stress


out_VonMises ↑

(type: logical , default: .false. , mandatory: no )
If out_VonMises=.true., the Von-Misès equivalent shear stress calculated on each atomic position is printed in output structure file. P = -1/3 Trace( stress ) sVM = Sqrt( 3/2*( ( s11 + P )^2 + ( s22 + P )^2 + ( s33 + P )^2 + 2.d0*( s23^2 + s13^2 + s12^2 ) ) )

See also: outFile , out_stress


out_core ↑

(type: logical , default: .false. , mandatory: no )
If out_core=.true., print index 1 for atom belonging to defect core, i.e. closer than rc from a line defect.

See also: outFile , rc


out_Ebinding ↑

(type: logical , default: .false. , mandatory: no )
If out_Ebinding=.true., the binding energy with the elastic stress field and an impurity is calculated and printed on each atomic positions. The elastic dipole characterizing the impurity, pImpurity(:,:), needs to be defined. (~ Eshelby inclusion eigenstrain)

See also: outFile , pImpurity


pImpurity ↑

(type: real, dim(3,3) , default: 0. , mandatory: yes, if out_Ebinding=.true. )
Elastic dipole characterizing the impurity for binding energy calculation. This is equivalent to the eigenstrain for an Eshelby infinitesimal inclusion. It should be given in the same units as the output energies (eV by default).

See also: out_Ebinding


out_x, out_y, out_z ↑

(type: logical , default: .false. , mandatory: no )
If out_x=.true., print first atomic coordinates as auxiliary properties. (not really useful)

See also: outFile


verbosity ↑

(type: integer , default: 4 , mandatory: no )
Integer values defining amount of information on output unit. verbosity=0: no output verbosity=4: normal output verbosity>=10: debugging


debug ↑

(type: logical , default: .false , mandatory: no )
For debug purpose.

Namelist Dislo ↑

nd ↑

(type: integer , default: 0 , mandatory: yes )
Number of dislocations defined in the namelist.

See also: bDislo , lDislo , cDislo


bDislo ↑

(type: real, dim(3,200) , default: 0. , mandatory: yes )
bDislo(1:3,i) is the Burgers vector of the dislocation numbered i. If alat is defined in the namelist Input, bDislo(:,:) is multiplied by alat.

See also: alat


lDislo ↑

(type: real, dim(3,200) , default: 0. , mandatory: yes )
lDislo(1:3,i) is the line vector of the dislocation numbered i. Units of this vector are arbitrary as the program will normalize it.


cDislo ↑

(type: real, dim(3,200) , default: 0. , mandatory: yes )
cDislo(1:3,i) is the dislocation position, i.e. a point belonging to the dislocation line. If alat is defined in the namelist Input, cDislo(:,:) is multiplied by alat.

See also: alat


cutDislo ↑

(type: real, dim(3,200) , default: 0. , mandatory: no )
cutDislo(1:3,i) is used to define the infinite half plane where the dislocation creates a displacement discontinuity: cutDislo(1:3,i) is the direction orthogonal to the dislocation line belonging to this discontinuity half plane. This does not need to be defined in the input file as the program will do it for you: - if dislocations can be gathered in dipoles, the plane containing the two dislocation lines will be used as a discontinuity plane. - if dislocations are isolated, the glide plane will be used as a discontinuity plane. As the glide plane is well defined only when an edge component is present, an arbitrary plane is used for a screw dislocation.

Namelist DDipole ↑

nDDipole ↑

(type: integer , default: 0 , mandatory: yes )
Number of dislocation dipoles defined in the namelist.

See also: bDDipole , lDDipole , c1DDipole , c2DDipole


bDDipole ↑

(type: real, dim(3,200) , default: 0. , mandatory: yes )
bDDipole(1:3,i) is the Burgers vector of the dislocation dipole numbered i. The Burgers vector of the first dislocation belonging to this dipole is bDDipole(1:3,i) and the one of the second dislocation is -bDDipole(1:3,i). If alat is defined in the namelist Input, bDDipole(:,:) is multiplied by alat.

See also: alat


lDDipole ↑

(type: real, dim(3,200) , default: 0. , mandatory: yes )
lDDipole(1:3,i) is the line vector of the dislocation dipole numbered i. Units of this vector are arbitrary as the program will normalize it.


c1DDipole ↑

(type: real, dim(3,200) , default: 0. , mandatory: yes )
c1DDipole(1:3,i) is the position of the first dislocation belonging to the dipole numbered i. The Burgers vector of this dislocation is bDDipole(1:3,i). If alat is defined in the namelist Input, c1DDipole(:,:) is multiplied by alat.

See also: alat


c2DDipole ↑

(type: real, dim(3,200) , default: 0. , mandatory: yes )
c2DDipole(1:3,i) is the position of the second dislocation belonging to the dipole numbered i. The Burgers vector of this dislocation is -bDDipole(1:3,i). If alat is defined in the namelist Input, c2DDipole(:,:) is multiplied by alat.

See also: alat

Namelist Loops ↑

nLoop ↑

(type: integer , default: 0 , mandatory: yes )
Number of loops defined in the namelist.

See also: bLoop , iLoop , xLoop , cLoop , rLoop , nHabitLoop


bLoop ↑

(type: real, dim(3,20) , default: 0. , mandatory: yes )
bLoop(1:3,i) is the Burgers vector of the loop numbered i. If alat is defined in the namelist Input, bLoop(:,:) is multiplied by alat.

See also: alat


iLoop ↑

(type: integer, dim(100) , default: 0 , mandatory: yes )
Number of points used to define the perimeter of the loop numbered i.

See also: xLoop


xLoop ↑

(type: real, dim(3,100,20) , default: 0. , mandatory: no )
xLoop(1:3,n,i) is the position of the point n belonging to the perimeter of the loop numbered i. These points need to be defined in continuous order along the loop perimeter as they correspond to a discretization of the dislocation line defining the loop. Insted of defining by hand the different points on the perimeter, you can use rLoop, cLoop, and nHabitLoop to define a circular loop If alat is defined in the namelist Input, xLoop(:,:,:) is multiplied by alat.

See also: rLoop , cLoop , nHabitLoop , alat


cLoop ↑

(type: real, dim(3,20) , default: 0. , mandatory: no )
cLoop(1:3,i) is the position of the "center" of the loop numbered i. This variable is optional and is used to define the cut surface, i.e. the displacement discontinuity created by the loop, by considering triangles between this point and the segments composing the loop perimeter. If you do not define it, by default, it is taken as the last point defining the loop perimeter (centerLoop=.FALSE. by default) or as the gravity center of all the points used to discretize the loop perimeter (centerLoop=.TRUE.). cLoop(1:3,i) can also be used to draw a circular loop, defining its center cLoop(1:3,i), it radius rLoop(i) and the normal to its habit plane nHabitLoop(1:3,i) If alat is defined in the namelist Input, cLoop(:,:) is multiplied by alat.

See also: centerLoop , rLoop , nHabitLoop , alat


rLoop ↑

(type: real, dim(20) , default: -1. , mandatory: no )
rLoop(1:3,i) used to define the loop radius when drawing a circular loop. If alat is defined in the namelist Input, rLoop(:) is multiplied by alat.

See also: cLoop , nHabitLoop , alat


nHabitLoop ↑

(type: real, dim(3,20) , default: 0. , mandatory: no )
nHabitLoop(1:3,i) used to define the normal to the loop habit plane when drawing a circular loop.

See also: cLoop , rLoop


shiftInsertedLoop ↑

(type: real, dim(3,20) , default: 0. , mandatory: no )
shiftInsertedLoop(1:3,i) is the displacement to add to the atoms, which have been inserted during the Volterra procedure for an interstitial loop. This allows controling the stacking fault of the interstitial loop.

See also: add_cut , alat


xLoop_noise ↑

(type: real , default: 0. , mandatory: no )
Amplitude of the noise to be added to the positions of the points defining the loop perimeter.

See also: xLoop


centerLoop ↑

(type: logical , default: .FALSE. , mandatory: no )
If centerLoop=.TRUE., then a center cLoop(1:3,n) is defined for each loop n. This center is used to discretize the loop surface in triangles all sharing a common corner corresponding to this center. If centerLoop=.FALSE. (default), then the last point defining the loop perimeter is used for the center and this last point is removed. If you do not manage to obtain the correct atomic stacking inside the loop, you may use this option and play with the variable cLoop(:,:).

See also: cLoop


uLoop_fast ↑

(type: logical , default: .FALSE. , mandatory: no )
If uLoop_fast=.TRUE., then only the solid angle part is calculated for the displacement created by the dislocation loops. If not, the elastic contribution is also calculated, assuming isotropic elasticity with a Poisson coefficient equal to 1/2 Not really usefull, as the impact of CPU time is negligible.