10. How to run WIEN2k for selected samples

Three test cases are provided in the WIEN2k package. They contain the two starting files case.struct and case.inst and all the output so that you can compare your results with them.

The test cases are the following (where the names correspond to what was called CASE in the rest of this User's Guide)

TiC
Fccni
TiO2

We recommend to run these test cases (in a different directory) and compare the output to the provided one. All test cases are setup such that the CPU-time remains small (seconds). For real production runs the value of RKMAX in case.in1 must be increased and a better (denser) k-mesh should be used.

In addition we provide a subdirectory example_struct_files were various more complicated struct files can be found.

1. TiC

The TiC example is described in detail in chapter 3 (Quickstart).

2. Fcc Nickel (spin polarized)

Ferromagnetic Nickel is a test case for a spin-polarized calculation. Ni has the atomic configuration $1s^2$, $2s^2$, $2p^6$, $3s^2$, $3p^6$, $3d^8$, $4s^2$ or [Ar] $3d^8$, $4s^2$. We treat the $1s$, $2s$, $2p$ and $3s$ as core states, and $3p$ (as local orbital), $3d$, $4s$ and $4p$ are handled as valence states. In a spin-polarized calculation the file structure and the sequence of programs is different from the non-spin-polarized case (see 4.5.2).

Create a new session and its corresponding directory. Generate the structure with the following data (we can use a large sphere as you will see from the output of nn):

Title	fcc Ni
Lattice	F
a	6.7 bohr
b	6.7 bohr
c	6.7 bohr
$\alpha,\beta,\gamma$	90
Atom	Ni, enter position (0,0,0) and RMT = 2.3

Initialize the calculation using the default RKmax and use 3000 k-points (a ferromagnetic metal needs many k-points to yield reasonably converged magnetic moments). Allow for spin-polarization.

Start the scf cycle (runsp_lapw). At the bottom of the converged scf-file (Fccni.scf) you find the magnetic moments in the interstital region, inside the sphere and the total moment per cell (only the latter is an ``observable'', the others depend on the sphere size).

:MMINT: MAGNETIC MOMENT IN INTERSTITIAL =   -0.03266
:MMI01: MAGNETIC MOMENT IN SPHERE  1    =    0.65679
:MMTOT: TOTAL MAGNETIC MOMENT IN CELL   =    0.62413

3. Rutile ($TiO_2$)

This example shows you how to ``optimize internal parameters''.

Create a new session and its corresponding directory. Generate the structure with the following data (we use a smaller O sphere because Ti-d states are harder to converge then O-p):

Title	TiO2
Spacegroup	$P4_2/mnm$ (136)
a	8.682 bohr
b	8.682 bohr
c	5.592 bohr
$\alpha,\beta,\gamma$	90
Atom	Ti, enter position (0,0,0) and RMT = 2.0
Atom	O, enter position (0.3,0.3,0) and RMT = 1.6

StructGen $^{\mbox{\textsc{TM}}}$should automatically add the equivalent positions.

Initialize the calculation using RKmax=6.5 in tio2.in1_st and use 100 k-points and a ``shift`` in kgen.

If you have a parallel machine, create a .machine file (possibly with 9 (or 3) processors since we have 9 k-points, for details see 5.4). You can use ``Execution $\rightarrow$ Run scf'', activate the ``parallel'' button'' and ``start scf'' in w2web, or the UNIX command

cp $WIENROOT/SRC_templates/.machine .

Then start the structure minimization in w2web using ``Execution $\rightarrow$ mini.positions''. This will generate TiO2.inM, which you should edit and change the ``stepsize'' from 5.0 to 3.0. For execution you can either use the defaults or put

run_lapw -I -p -fc 1.0

as job (for parallel execution).

Alternatively you can use

cp $WIENROOT/SRC_templates/case.inM TiO2.inM
emacs TiO2.inM (and change all 5.0 to 3.0)
min_lapw -j ``run_lapw -I -fc 1 -p''

This will start an scf cycle, call the program min, which generates a new struct file using the calculated forces, and continues with the next scf cycle. It will continue until the forces are below 1 mRy/bohr (TiO2.inM) and the final results are not ``saved'' automatically but can be found in the ``current'' calculation.

You can watch the minimization using the file TiO2.scf_mini, which contains the final iteration of each geometry step. If the forces in this file oscillate from plus to minus and seem to diverge, or if they change very little, you can edit TiO2.inM (reduce or increase the stepsize), and remove TiO2.tmpM (contains the ``history'' of the minimization and is used to calculate the velocities of the moving atoms).

The final structural parameter of the O-atom should be close to x=0.304, which compares well with the experimental x=0.305.