Tutorial 1

Basic SIESTA execution: energy

optimization of H2 molecule

In this tutorial, we perform the energy optimization of a H2 molecule with SIESTA. Our
aim is to give a flavour of the general system descriptors included in the input file of a
SIESTA calculation, and a brief description of the self-consistent field method.

1.1 Execution of the program

Enter in the directory /T1.01/H2. This directory contains the files that SIESTA requieres
to optimize the energy of a H2 molecule; H2.fdf (input file) and H.psf (pseudopotential
file).
To execute the code, type in the console the following command,

SIESTA<H2.fdf>H2.txt

where SIESTA is the siesta executable, and H2.fdf and H2.txt are the input and
output files, respectively. This simple SIESTA execution will take a few seconds to
finish. After a successful run of the program, you should have several files in your
directory, see Figure 1.1.
SIESTA users must be aware that there are a series of steps that should be followed to
obtain a well converged calculation, see Figure 1.1 . The generation of a pseudopotential
is a complex process, and every pseudopotential must be thoroughly checked before
use. Tutorial 3 provides basic guidelines on the generation and test of pseudopotentials.
Because SIESTA falls into category of methods with atom-centered basis sets, the basis
set must also be optimized in order to perform calculations within reasonable time and

1
Figure 1.1: Schematic representation of the input and outputs files in a SIESTA calcu-
lation. The prefix H2 is specified by the parameter SystemLabel in the input file.

precision, see Tutorial 4. The specific parameters that define the SIESTA calculation
(k-points, meshcutoff, XC, spin, etc) must also be carefully chosen.

Figure 1.2: Series of steps that should be followed to obtain a well converged calculation

1.2 FDF file: description of program options

The input file contains parameters specifying both the system we want to study, and
the accuracy of the calculation. Whereas the former parameters are mandatory, other
parameters have default values assigned and, in principle, it is not necessary to explicitly
include them in the input file. In this section, we focus on the description of the general

2
system descriptors. However, it is important to stress out that the default values do not
always guarantee a converged calculation, see Tutorials 5, 6 and 7.

1.2.1 General system descriptors

1. Name and Label
SystemName is a string of one or several words containing a descriptive name of the
system (max. 150 characters). SystemLabel is a single word (max. 20 characters
without blanks) containing a nickname of the system, used to name output
files.
SystemName H2 ( default_value : blank )
SystemLabel H2 ( default_value : siesta )

2. Chemical species and Atoms

As said before, the parameters specifying the system have to be necessarily specified
in the input file. Those parameters are the number of different species and atoms
present in the unit cell, the list of the different species, atomic numbers, atom
positions and the label associated with each atomic specie.
NumberOfSpecies (integer): Number of different atomic species in the simulation.
Atoms of the same species, but with a different pseudopotential or basis set are
counted as different species.
NumberOfAtoms (integer): Number of atoms in the simulation.
ChemicalSpeciesLabel (data block): It specifies the different chemical species that
are present, assigning them a number for further identification. Siesta recognises
the different atoms by the given atomic number.

NumberOfSpecies 1
NumberOfAtoms 2
% block C h e m i c a l S p e c i es L a b e l
1 1 H
% endblock C h e m i c a l S p ec i e s L a b e l

The first number in a line in the data block ChemicalSpeciesLabel is the species
number. It is followed by the atomic number, and then by the desired label. This

3
 label will be used to identify corresponding files, namely, pseudopotential file, user
basis file, basis output file, and local pseudopotential output.

3. Cell and Atoms positions

The size of the cell itself is specified, using some combination of the options Lattice
Constant, LatticeVectors, LatticeParameters, and SuperCell. In this tutorial we
use the combination of the first two parameters.

Lattice Constant It is a real parameter that define the scale of the lattice vectors.
The value of this parameter is multiplied by each of the matrix LatticeVectors el-
ement.

LatticeVectors (block) It is a real matrix that contains (row) vectors that define
the lattice. Thus, we have full autonomy over the shape and size of cell. The cell
vectors are read in units of the lattice constant defined above. They are read as a
matrix CELL(ixyz,ivector), each vector being one line.

Notice that although the system considered in this tutorial is aperiodic (H2 molecule),
SIESTA still does use of periodic boundary conditions (PBC). Thus, it is needed
to introduce a vacuum region that should be large enough that periodic images
corresponding to adjacent replicas of the supercell do not interact significantly, see
Figure 1.3.

Figure 1.3: H2 molecule placed in the center of a large box vacuum

LatticeConstant 10.00 Ang

% block LatticeVectors
1.0 0.0 0.0
0.0 1.0 0.0

4
 0.0 0.0 1.0
% endblock LatticeVectors

Once that the size of the cell is specified, the positions of the hydrogen atoms
within the cells are given using the AtomicCoordinatesAndAtomicSpecies block.
In this block the atom index are also indicated. Notice that the hydrogen atoms
are placed in the center of the cell, see Figure 1.4.
% block A t o m i c C o o r d i n a t e s A n d A t o m i c S p e c i e s
4.5000 5.0000 5.0000 1
5.5000 5.0000 5.0000 1
% endblock A t o m i c C o o r d i n a t e s A n d A t o m i c S p e c i e s

Figure 1.4: Schematic illustration of the super-cell approach

1.3 Convergence of the SCF cycles

The self-consistent Kohn-Sham method is used to calculate energies and forces. This
method consists in defining an initial electron density to solve Kohn-Sham equations.
Then, from the solution of these equations a new electronic density is calculated. The
cycle is repeated until a self-consistent electron density is achieved, see Figure 1.5
We can inspect the output file H2.txt to analyse the electronic density convergence

5
 Figure 1.5: Self Consistent Method

of the H2 molecule. Type in the console the following command to check the energy of
the system in each of the SCF steps,

cat H2.txt|grep scf:

scf : iscf Eharris ( eV ) E_KS ( eV ) FreeEng ( eV ) dDmax Ef ( eV )

scf : 1 -34.1562 -30.0304 -30.0304 0.56242 -2.7224
scf : 2 -34.1379 -34.1291 -34.1291 0.02752 -2.2036
. . . .
. . . .
scf : 18 -34.1363 -34.1362 -34.1362 0.00007 -2.3057

6
 Notice that both total energy and electron density density variation (dDmax) are
minimal in the last SCF step, see Figure 1.6 and 1.7. The accuracy of the calculation of
energies and forces in a self-consistent-field loop can be tuned by specifying the number
of maximum and minimum number of SCF iterations per time step, mixing options, etc.
Those parameters are described in advances tutorials.

Figure 1.6: Electron density density variation (dDmax) as a function of SCF step

7
Figure 1.7: Total Energy as a function of SCF step