I
Name InitialIonicTemperature
Section DFTBPlusInterface
Type float
Default 0.0
If this variable is present, the ions will have initial velocities
velocities to the atoms following a Boltzmann distribution with
this temperature (in Kelvin). Used only if TDDynamics = Ehrenfest
and MoveIons = yes.
Name InitializeGPUBuffers
Section Execution::Accel
Type logical
Initialize new GPU buffers to zero on creation (use only for debugging, as it has a performance impact!).
Name InitialSpins
Section States
Type block
The spin character of the initial random guesses for the spinors can
be fixed by making use of this block. Note that this will not "fix" the
the spins during the calculation (this cannot be done in spinors mode, in
being able to change the spins is why the spinors mode exists in the first
place).
This block is meaningless and ignored if the run is not in spinors mode (SpinComponents = spinors).
The structure of the block is very simple: each column contains the desired $\left< S_x \right>, \left< S_y \right>, \left< S_z \right> $ for each spinor. If the calculation is for a periodic system and there is more than one k-point, the spins of all the k-points are the same.
For example, if we have two spinors, and we want one in the $S_x$ "down" state, and another one in the $S_x$ "up" state:
%InitialSpins
0.5 | 0.0 | 0.0
-0.5 | 0.0 | 0.0
%
WARNING: if the calculation is for a system described by pseudopotentials (as opposed to user-defined potentials or model systems), this option is meaningless since the random spinors are overwritten by the atomic orbitals.
This constraint must be fulfilled:
$ \left< S_x \right>^2 + \left< S_y \right>^2 + \left< S_z \right>^2 = \frac{1}{4} $
Name InstrumentFunctions
Section Execution::Debug
Type block
This input options controls which routines are going to be instrumented
for the tools selected using the Debug option.
%InstrumentFunctions
function_name | instrumentation_tool
%
Here is an example to better understand how this works:
%InstrumentFunctions
grid/grid.F90/grid_init_from_grid_stage_1 | verrou
%
NOTE: Currently only a single function can be instrumented!
Available instrumentation tools:
Options:
- verrou:
Verrou helps you look for floating-point round-off errors.
Name Interaction1D
Section Hamiltonian::XC
Type integer
Default interaction_soft_coulomb
When running in 1D, one has to soften the Coulomb interaction. This softening
is not unique, and several possibilities exist in the literature.
Options:
- interaction_exp_screened:
Exponentially screened Coulomb interaction.
See, e.g., M Casula, S Sorella, and G Senatore, Phys. Rev. B 74, 245427 (2006).
- interaction_soft_coulomb:
Soft Coulomb interaction of the form $1/\sqrt{x^2 + \alpha^2}$.
Name Interaction1DScreening
Section Hamiltonian::XC
Type float
Default 1.0
Defines the screening parameter $\alpha$ of the softened Coulomb interaction
when running in 1D.
Name Interactions
Section System
Type block
This input option controls the interactions between systems. It basically
allows to select which systems will interact with another system through
a given interaction type. The format of the block is the following:
%Namespace.Interactions
interaction_type | interaction_mode | …
%
Here is an example to better understand how this works:
%SystemA.Interactions
gravity | all_except | "SystemB"
%
This means that SystemA and all the systems that belong to the same namespace (i.e., all its subsystems) will interact through gravity with all interaction partners that are also able to interact through gravity, except with SystemB. Note that the opposite is not true so, although clearly unphysical, this will not prevent SystemB from feeling the gravity from SystemA (in Octopus the interactions are always one-sided).
NB: Each interaction type should only appear once in the block. Any further instances beyond the first will be ignored.
Available modes and interaction types:
Options:
- no_partners:
(interaction mode)
Do not interact with any partner.
- all_partners:
(interaction mode)
Interact with all available partners.
- only_partners:
(interaction mode)
Interact only with some specified partners. A list of partner names must
be given.
- all_except:
(interaction mode)
Interact with all available partners except with some specified
partners. A list of partner names to exclude must be given.
- gravity:
(interaction type)
Gravity interaction between two masses.
- lorentz_force:
(interaction type)
Lorentz force resulting from an EM field acting on a moving charge.
- coulomb_force:
(interaction type)
Coulomb force between two charged particles.
- linear_medium_to_em_field:
(interaction type)
Linear medium for propagation of EM fields.
- current_to_mxll_field:
(interaction type)
Drude dispersive linear medium for propagation of EM fields.
- maxwell_e_field:
(interaction type)
Electric field resulting from the Maxwell solver.
- maxwell_b_field:
(interaction type)
Magnetic field resulting from the Maxwell solver.
- maxwell_vector_potential:
(interaction type)
Vector potential resulting from the Maxwell solver.
- lennard_jones:
(interaction type)
Force resulting from a Lennard Jones potential between classical particles.
Name InteractionTiming
Section Time-Dependent::Propagation
Type integer
Default timing_exact
A parameter to determine if interactions should use the quantities
at the exact iteration or if retardation is allowed.
Options:
- timing_exact:
Only allow interactions at the exact iterations required by the algorithms behing executed
- timing_retarded:
Allow retarded interactions
Name InvertKSConvAbsDens
Section Calculation Modes::Invert KS
Type float
Default 1e-5
Absolute difference between the calculated and the target density in the KS
inversion. Has to be larger than the convergence of the density in the SCF run.
Name InvertKSGodbyMu
Section Calculation Modes::Invert KS
Type float
Default 1.0
prefactor for iterative KS inversion convergence scheme from Godby based on van Leeuwen scheme
Name InvertKSGodbyPower
Section Calculation Modes::Invert KS
Type float
Default 0.05
power to which density is elevated for iterative KS inversion convergence
scheme from Godby based on van Leeuwen scheme
Name InvertKSMaxIter
Section Calculation Modes::Invert KS
Type integer
Default 200
Selects how many iterations of inversion will be done in the iterative scheme
Name InvertKSmethod
Section Calculation Modes::Invert KS
Type integer
Default iter_godby
Selects whether the exact two-particle method or the iterative scheme
is used to invert the density to get the KS potential.
Options:
- two_particle:
Exact two-particle scheme.
- iterative:
Iterative scheme for $v_s$.
- iter_stella:
Iterative scheme for $v_s$ using Stella and Verstraete method.
- iter_godby:
Iterative scheme for $v_s$ using power method from Rex Godby.
Name InvertKSStellaAlpha
Section Calculation Modes::Invert KS
Type float
Default 0.05
prefactor term in iterative scheme from L Stella
Name InvertKSStellaBeta
Section Calculation Modes::Invert KS
Type float
Default 1.0
residual term in Stella iterative scheme to avoid 0 denominators
Name InvertKSTargetDensity
Section Calculation Modes::Invert KS
Type string
Default target_density.dat
Name of the file that contains the density used as the target in the
inversion of the KS equations.
Name InvertKSVerbosity
Section Calculation Modes::Invert KS
Type integer
Default 0
Selects what is output during the calculation of the KS potential.
Options:
- 0:
Only outputs the converged density and KS potential.
- 1:
Same as 0 but outputs the maximum difference to the target density in each
iteration in addition.
- 2:
Same as 1 but outputs the density and the KS potential in each iteration in
addition.
Name IonsConstantVelocity
Section Time-Dependent::Propagation
Type logical
Default no
(Experimental) If this variable is set to yes, the ions will
move with a constant velocity given by the initial
conditions. They will not be affected by any forces.
Name IonsTimeDependentDisplacements
Section Time-Dependent::Propagation
Type block
(Experimental) This variable allows you to specify a
time-dependent function describing the displacement of the ions
from their equilibrium position: $r(t) = r_0 + \Delta
r(t)$. Specify the displacements dx(t), dy(t), dz(t) as
follows, for some or all of the atoms:
%IonsTimeDependentDisplacements
atom_index | "dx(t)" | "dy(t)" | "dz(t)"
%
The displacement functions are time-dependent functions and should match one
of the function names given in the first column of the TDFunctions block.
If this block is set, the ions will not be affected by any forces.