lev00
^^^^^

19.09.2003  Version 2.0 
^^^^^^^^^^^^^^^^^^^^^^^

(1) some small bugs corrected.

(2) Added support for SIESTA (with some help from Adam Foster):

- lev00 can now fully support charge and spin density options
- due to a problem in setting users k-points to SIESTA, there
  is still no support for DOS and PDOS.

To compile lev00, you need param.inc file. The get one, run

get_param_siesta

(which is in TOOLS/SIESTA_Tools and should be compiled only once). It
will read the final geometry file *.XV and the density file *.RHO to
construct param.inc file. Then, as usual, run lev00.comp to get lev00
in your working directory.

22.01.2004 Version 2.1
^^^^^^^^^^^^^^^^^^^^^^^

(1) do_param.f: do_param writes ISPIN variable to the param.inc file.

(2) lev00 (several files): spin 1,2 for spin-polarised calculations 
    is recognised by lev00 when doing various DOS; no need now to
    copy band.out.1 (or .2) into band.out for each spin; when going
    into any of the DOS options, lev00 asks which spin if ISPIN=2.
    In the case of a spin-polarised calculation any DOS is without
    the spin prefactor of 2.

(3) lev00.f: structure of the main routine has been simplified considerably
    and is now more transparent; some CASTEP/CETEP options are not visible
    if VASP or SIESTA are on and vice versa.

(4) when searching for a string consisting of 2 or 3 words, the string
    is broken down into words; this means that number of spaces
    between the words does not matter anymore. This is useful if
    e.g. the number of spaces was changed in the output file of the PW code
    which lev00 needs (as e.g. happenned between versions 4.4. and 4.6
    of vasp)  

1.02.2004 Version 2.2
^^^^^^^^^^^^^^^^^^^^^^

(1) a bug in lev00.f: a call to prep_plot() in potion MAP was missing
(2) added option 44 in the plane menu: after the plane is chosen
(i.e. the normal, central point and the dimensions), one can choose
another plane parallel to the existing one. Note that if the plane was
chosen using 3 points methos, you will have to choose the same plane again by
specifying the normal only (it is diplayed in the menu), so that the
3-point  method becomes off.


30.3.2004 - 15.04.2004  Version 2.3
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(1) a bug in density.f related to reading density of siesta (thanks to Ross)

(2) a bug in input00.f in setting up the first k-point while reading
siesta input (actually, is not used)

(3) menus in plot_add.f, simulate.f and density.f changed: 
 - to allow character inputs instead of only numbers; 
 - 'An' to choose the way coordinates are given; 
 - 'Co' is added in all menus options related to density to view atomic
   positions in fractional/Cartesian format
 - 'Q' is used to quit 
This all is in accord with the general implementation of menus in tetr.
Still, other menus will have to be modified accordingly - later!

(4) atomic species are now asked every time lev00 is started and the
coordinates are read in
(changes in: lev00.f, lev00.inc, mendeleev.inc, input00.f)

(5) density reading routines (vasp, siesta) have been moved from density.f
to a separate file read_density.f

(6) a bug in a routine reading siesta density has been corrected; it
affected only the units of the density array grid()
(the file affected: read_density.f)

(7) insignificant changes in density.f 

(8) A MAJOR NEW DEVELOPMENT: a new option added in the density menu (#10) 
allowing to write input files to gOpenMol to plot "orbitals" out of the 
density; two files are produced (both are needed): 

 - gOpenMol.cube containing the density and atomic positions in the
specific format; all atomic positions are printed (they are not actually used 
by gOpenMol which is a bit of a puzzle); the density is printed on a grid
within a box (the density box) to be specified; the box may run across 
several cells or overlap with several adjacent cells, etc.

 - gOpenMol.xyz which contains atomic positions (again!); two options
   are available (the XY option in the menu):
      - all cell atoms
      - only atoms within the density box (a larger box can also be chosen
        using the option 3 in the menu; the latter appears if the appropriate
         XY menu option has been chosen)

Files affected: density.f

What you need to do with gOpenMol:
 - read the gOpenMol.xyz file by going to: 
        File -> Import -> Coord -> Browse -> Apply
 - read in the gOpenMol.cube file by going to:
        Run -> gCube2plt/g94cub2pl (cube) -> Browse -> Apply
   which should produce an output in the window to be openned finished by 
   "Job done ..."; it produces a file gOpenMol.plt
 - then plot by reading the gOpenMol.plt:
        Plot -> Contour -> Browse -> Import File 
You will find the min and max values of the density; choose the density value
in the boxes underneath the min/max info and click Apply: the density will be
shown alongside the atomic positions in the same window; you will be able to 
rotate the image.

To make the density transparent, click on Details opposite to the chosen 
density level and change the Opacity level, then click Apply. It is possible 
to plot several density levels in the same way by filling in several density
levels and then changing their opacities.

To see atoms better, use: 
        View -> Atom Type 
and then the option Licorice with e.g. Cyl.rad = 0.1 and Sph. rad. = 0.3.
To have different radii for different species, use CPK as well.
See the gOpenMol manual for more instructions.
gOpenMol is freely available at: http://www.csc.fi/gopenmol/


8.10.2004  Version 2.4
^^^^^^^^^^^^^^^^^^^^^^
1. cosmetic changes in get_all_from_SIESTA  (au_A,A_au)
   (file: input00.f)
2. similar cosmatic changes in siesta_dens 
   (file: read_density.f)
3. multiplication factor for the density used in line/plane options (multcon) 
   has been changed from 100 to 1, since the density is given in actual units
   (A^-3) 
   (files: menu.f, plot_add.f)
4. an option [Ap] is added to the plane menu to show (X,Y) positions of all atoms
   with respect to the local coordinate system fixed with the plane including
   the out-of-plane distance
   (files: density.f, plot_add.f)
5. the title line in the CONTCAR file is also printed on the screen for
   convenience to have the order of species 
   (file: input00.f)

24.11.2004  Version 2.41
^^^^^^^^^^^^^^^^^^^^^^^^
1. some minor changes to comply with Intel fortran-90 compiler
   [files: almost all]
2. make00.lev changed
3. new file tools_strings.f created that contains all the strings routines. A trick 
   in those to port an variable lenths string has been removed to comply with fortran
   rules.
   [file: inpu00.f]
4. PROCAR file with projected DOS is now working with option LORBIT=11 of VASP. Still, 
   only total p and d contributions are worked out by lev00, although all components of
   those are available in PROCAR. Can easily be taken into account as well, but later.
   [file: input00.f]

10.1.2005 Version 2.42
^^^^^^^^^^^^^^^^^^^^^^

1. another format for reading k-points from the OUTCAR file implemented as the
one existed before did not seem to appear in the OUTCAR file  in some cases
   [file: input00.f]


19.1.2005 Version 2.43
^^^^^^^^^^^^^^^^^^^^^^

1. can read unformatted siesta RHO (electron/spin density) files as well; detects 
   automatically if the file is formatted or unformatted. [Note: may not work
   if the unformatted RHO file was produced on another platform.]
   [files: read_density.f, SIESTA_Tools/get_param_siesta.f]
2. a debugging printout removed from vasp_dens() routine.
   [file: read_density.f]

2.2.2005 Version 2.44
^^^^^^^^^^^^^^^^^^^^^

1. Makefile make00.lev updated.
2. XY option in gOpenMol menu of the dnesity menu: default changed to keeping
   atoms inside the box.
   [file: density.f]
3. a bug in for_gOpenMol() routine (wrong calculation of center()) corrected.
   [file: density.f]
4. a very subtle bug (thanks to Chris!) in for_gOpenMol() routine, that
   resulted in a shift of the countour surfaces with respect to the geometry, 
   corrected.
   [file: density.f]

24.7.2005 Version 2.45
^^^^^^^^^^^^^^^^^^^^^^

1. the main menu corrected
   [file: lev00.f]]
2. English in several menus and some unimportant bugs (in calling plot() routine)
   [file: menu.f, prep_dos.f, prep_disp.f, plot_add.f, lev_coulomb,f, dipole.f]
3. several little bugs in the plane menu (affected only the menu look) fixed
   [file: plot_add.f]
4. some cosmetic changes in charge, dipole and quadrupole routines menus
   [files: dipole.f, plot_add.f]
5. a tiny uninportant bug in calling Plot_Explr() routine fixed
   [file: plot_add.f]
6. option 7 of the density menu is linked to CASTEP only
   [file: density.f]
7. removed 'island' message (given if option 4 of the main menu) for VASP and
   SIESTA
   [file: input00.f]
8. option 4 of the main menu is directly linked to CASTEP/CETEP only; in the
   case of VASP, the file name is asked (CHGCAR, PARCHG or LOCPOT) before
   reading in
   [file: lev00.f, density.f, input00.f]
9. manipulation option rewritten from scratch; includes SIESTA as well this time
   [files: lev00.f, manip_dens.f, density.f, read_density.f, make00.lev, tools_strings.f]
10. option 4 -> 4  removed (writing partial density)
   [file: prep_pl0.f]
11. option 8 of the main menu (shift density) was moved to the density menu as
   option 12, which allowed to generalise it for any code
   [files: lev00.f, density.f, make00.lev]
   [file removed: shift_charge.f]

7.01.2007 Version 2.46
^^^^^^^^^^^^^^^^^^^^^^^

1. NEW: get_pram_siesta has been modified to read in NBANDS and NKPTS from siesta DOS run 
   (if the new option BandPoints in on), as well as band energies into band.out file,  
   to allow for DOS support in lev00
   [file: get_param_siesta.f]
2. NEW: corresponding changes have been done in lev00 routines as well to support total DOS
   calculation with siesta: presently supported menu options 0 and 3 of DOS (smearing 
   tests and total DOS)
   [files: input00.f, lev00.f]
3. character*(*) construction introduced in all routines to handle variable string lengths
   [files: tools_strings.f, density.f, dipole.f, input00.f, lev00.f, manip_dens.f, 
           plot_add.f, simulate.f] 
4. NEW: it is possible to set the upper limit to the energy (x axis) when previewing and
   of the DOS graphs with gnuplot
   [files: prep_dos.f, prep_disp.f, plot_add.f]
5. an old bug in working out physical bands, when the smearing is on, fixed
   [file: dos_add.f]

10.02.2007 Version 2.46.1
^^^^^^^^^^^^^^^^^^^^^^^^^

1. Bug fixes in reading unformatted siesta density
   [files: manip_dens.f, read_density.f]
2. Bug fixed in get_param_siesta: stopped if bands not found; now fixed
   [file: SIESTA_Tools/get_param_siesta.f]
3. New design for the hat (when lev00 starts), similar to that of TETR.
   tmake script creates build.hat file which is attached to hat.f
   [files: lev00.f, hat.f, make00.lev, tmake]

12.03.2007 Version 2.46.2
^^^^^^^^^^^^^^^^^^^^^^^^^
1. Format statements in the DOS routines were changed to allow for up to 99999
   bands to be displayed
   [files: lev00.f, prep_dos.f]
2. get_param_siesta modified to read the main output file for the grid and
   spin if the RHO file is not available; also, one small bug fixed
   [file: get_param_siesta.f]

2.04.2007 Version 2.46.3
^^^^^^^^^^^^^^^^^^^^^^^^^

1. a bug in bastr.f corrected (interestingly, this one was fixed a long time
   ago in tetr!)
   [file: bastr.f]

16.10.07 Version 3.0
^^^^^^^^^^^^^^^^^^^^
This is MAJOR DEVELOPMENT: it is now f90 code with memory allocation, so that there is
no need to run do_param and get_param_siesta to recompile the code for the specific
system. It is now only required a single compilation during installation.
The codes  do_param and get_param_siesta are now completely redundant together
with the lev00.comp script!

NEW features: it

1. combines do_param and get_param_siesta routines with lev00 into a single package
2. all array boundaries are worked out automatically from the VASP or SIESTA output 
   and/or input files, where appropriate, and
3. all arrays depending on the system size paramaters are allocated and deallocated
   as appropriate.
4. other f90 features are used including modules instead of common blocks.
5. CASTEP and CETEP capabilities have been removed as redundant.
6. some cleaning up of the routines.
7. some little changes of the menu options around the code.
8. all (except that in the bastr.f90 routine) 'stop' commands have been
   removed so the the run could be safely contiued even if something goes wrong
   (e.g. a file does not exist or incompatible).

At this stage I tried to keep the code unchanged whenever possible to ease its
debugging and testing. Therefore, no new major features in this version. 

   [files: all; new files added ]

17.10.07 Version 3.01
^^^^^^^^^^^^^^^^^^^^^
1. bug fixes
2. detailed account of reading SIESTA input is given on the screen for
convenience
   [files: read_siesta_input.f90, tools_strings.f90, hat.f90, hould.f90,
           lev00.f90, mainmenu.f90, read_vasp_geom.f90, seed.f90, 
           lev_coulmb.f90, get_param_siesta.f90]

31.10.07 Version 3.02
^^^^^^^^^^^^^^^^^^^^^
1. negative VASP densities can also be read in, density is not set up to zero if
   negative (to support density differences)
   [file: read_density.f90]

10.11.07 Version 3.03
^^^^^^^^^^^^^^^^^^^^^
1. a less stringent procedure for reading vasp input files implemented to allow some 
   variations in the vasp output in OUTCAR; in particular, after a non-scf run, band
   energies are not given, so that lev00 proceeds without the DOS option; logical 
   variable (yesen) is intorduced to distinguish between the existence of k-points 
   (yeskp) and the KS energies.
   [files: mainmenu.f90, lev00.f90, read_vasp_geom.f90, do_param.f90, get_param_siesta.f90,
           read_siesta_input.f90, kpoints.f90]

4.12.07 Version 3.04
^^^^^^^^^^^^^^^^^^^^^

1. a check is performed on consistency of the information in te brill.dat file with the
   number of k-points from the initialisation (NKPTS); DOS will not be calculated if 
   inconsistency is found.
   [files: prep_dos.f90, prep_disp.f90]
2. a bug in reading unformatted siesta RHO files fixed
   [file: read_density.f90]

28.02.08 Version 3.05
^^^^^^^^^^^^^^^^^^^^^
1. when reading species names from the OUTCAR file, only the first two characters
   before the underscore character are kept; the rest is ignored - no error message
   is given anymore
   [file: do_param.f90]
2. Species La is added.
   [file: mendeleev.f90]
3. format of the charge density is changed when saved in the vasp format to improve
   transferability
   [files: read_density.f90, write_dens.f90, manip_dens.f90] 
4. format improved when writing number of grid points in calculating charges in spheres
   [file: plot_add.f90]

10.04.08 Version 3.06
^^^^^^^^^^^^^^^^^^^^^
1. a bug in the routine reading siesta density for manipulation fixed
  [file:  read_density.f90]
2. a bug in a write statement
  [file: get_param_siesta.f90]
3. a bug in writing siesta density fixed
  [file: write_dens.f90]
4. a bug in defining A_au, au_A fixed
  [file:  read_density.f90]

09.06.08 Version 3.07
^^^^^^^^^^^^^^^^^^^^^

1. density mainpulation routine bug: the density is saved much too late; fixed
  [file: manip_dens.f90]
2. a check on the species when reading vasp density files for manipulation is
   handled with more freedom - only warnings are given, but one may proceed in
   the calculation even if the files have different species or their numbers;
   files are considered inconsistent if only lattice vectors and the grid are
   different; no more check is performed against the geometry file read in 
   initially when staring lev00
  [file: read_density.f90]

01.07.08 Version 3.08
^^^^^^^^^^^^^^^^^^^^^
1. bug fixes 
  [files: menu.f90, read_density.f90]

27.11.08 Version 3.09
^^^^^^^^^^^^^^^^^^^^^
1. bug fixes in the routines working on the DOS
  [files: prep_dos.f90, prep_disp.f90, dos_add.f90]
2. To avoid a crap produced sometimes by gnuplot when it sees the double precision numbers,
   we write to the file for gnuplot when working on the DOS in a simpler format.
  [files: prep_dos.f90, prep_disp.f90, tools.f90]

28.11.08 Version 3.10
^^^^^^^^^^^^^^^^^^^^^
1. spin components of PARCHG and LOCPOT files (if exist) can also be read in the same manner
   as of CHGCAR (in the case of VASP)
  [files: mainmenu.f90, density.f90, read_density.f90]
  
30.04.09 Version 3.11
^^^^^^^^^^^^^^^^^^^^^
1. NEW: projected DOS for SIESTA. One can read in siesta PDOS files and plot them.
   Up to 20 PDOS curves can be plotted alongside the total DOS, for each PDOS curve
   a sum over a subset of atoms is done. One can choose different subsets for each PDOS.
   TTag routine from tetr (updated and suplied with an extra feature of choosing atoms
   by species, option [Sp]) has been implemented for selecting atoms.
  [files: ttag.f90, read_siesta_pdos.f90, device.f90, mainmenu.f90, tools_strings.f90, 
          plotting.f90,Makefile]
2. some minor bugs (mainly in writing on the screen)
  [files: read_siesta_input.f90, lev00.f90]

06.05.09 Version 3.12
^^^^^^^^^^^^^^^^^^^^^

1. NEW: information about the Fermi energy and HOMO/LUMO energies is displayed from the
   siesta output and EIG files in the PDOS option. This addition is not yet complete,
   as spin-polarised calculation is not yet handled.
   [files: read_siesta_input.f90, get_param_siesta.f90]
2. An option TD added to the siesta PD menu to show or not the total DOS.
   [file: read_siesta_pdos.f90, plotting.f90]
3. NEW: in the case of SIESTA, XV file is not longer needed, only fdf file is now used
   for all necessary information concerning arrays dimensions, geometry, etc.
   [files: get_param_siesta.f90, read_siesta_input.f90]

08.05.09 Version 3.13
^^^^^^^^^^^^^^^^^^^^^
1. NEW: total DOS can be plotted from siesta [seed].EIG file as well.
   [files: mainmenu.f90, read_siesta_pdos.f90, plotting.f90, 
           read_siesta_input.f90, siesta_eig.f90, Makefile]
2. A bug fixed.
   [file: read_siesta_pdos.f90]
3. Rearrangement of siesta PDOS menus.
   [file: read_siesta_pdos.f90]

20.05.09 Version 3.14
^^^^^^^^^^^^^^^^^^^^^

1. NEW: manipulation option has been taken out of the density menu and moved into the very
   first menu. This allows one to manipulate densities without reading in all the geometry, etc.
   files. This is especially useful for generating a density difference file.
   [files: mainmenu.f90, code.f90, read_density.f90, manip_dens.f90]
2. seed string is initialised before it is read in from a siesta input file; strangely, not doing so
   caused a problem on some platforms.
   [file: get_param_siesta.f90]
 
22.05.09 Version 3.15
^^^^^^^^^^^^^^^^^^^^^
1. NEW: a possibility added to the manipulation menu to check density files even without
   manipulating them,.
   [files: manip_dens.f90, read_density.f90]

28.05.09 Version 3.16
^^^^^^^^^^^^^^^^^^^^^
1. A bug (since version 3.12) fixed (atomic positions read in from SIESTA were wrongly multiplied by
   a conversion factor au_A).
   [file: read_siesta_input.f90]

18.06.09 Version 3.17
^^^^^^^^^^^^^^^^^^^^^
1. A bug in reading lattice vectors from siesta density file fixed.
   [files: manip_dens.f90, read_density.f90]

21.12.09 Version 3.18
^^^^^^^^^^^^^^^^^^^^^

1. Reading of RWIGS from the output of newer versions of vasp
2. Working out the number of species is done differently due to a different
   pseudopotential information vasp may produce
3. Errors in reading CHGCAR0-like files of VASP are made specific to help to identify them.
   [files: do_param.f90, read_vasp_geom.f90, read_density.f90]

10.01.10 Vewrsion 3.19
^^^^^^^^^^^^^^^^^^^^^^
1. Allocation of Z_atom was moved at the beginning of the routine density() since some compilers
   give an error after hitting 4 in the type of density menu for VASP.
   [file: density.f90]

15.03.10 Vewrsion 3.20
^^^^^^^^^^^^^^^^^^^^^^
1. A bug in calling routines find_string fixed (rewind at the beginning was missing)
   [files: get_param_siesta.f90]

29.05.10 Vewrsion 3.21
^^^^^^^^^^^^^^^^^^^^^^
1. differences in the format of OUTCAR, CHGCAR and PARCHG in vasp version 5.2 with the previous
   versions were addressed. 
   [files: do_param.f90, density.f90, read_density.f90]
2. Format of on-screen statements when reporting the reading from a file changed ("... done 10%", etc.)
   [files: read_density.f90, manip_dens.f90, simulate.f90, write_dens.f90]
3. New species added - the whole Mendellev's Table, all 112 of them!
   [files: read_vasp_geom.f90, mendeleev.f90]

06.11.10 Vewrsion 3.22
^^^^^^^^^^^^^^^^^^^^^^
1. A bug in a format statement fixed.
   [file: density.f90]

06.07.11 Version 3.23
^^^^^^^^^^^^^^^^^^^^^^
1. Bugs in calculating total and projected DOS for siesta from the EIG and PDOS files were fixed.
   [files: siesta_eig.f90, read_siesta_pdos.f90]
2. NEW: in calculating total and projected DOS from siesta EIG and PDOS files it is now
   possible to choose spin in the case of a spin-polarised calculation
   [files: mainmenu.f90, siesta_eig.f90, read_siesta_pdos.f90]
3. In addition, all the menus related to DOS are better presented (revamped)
   [files: get_param_siesta.f90, mainmenu.f90]
4. It is indicated in the case of the tetrahedron method for the DOS if there is no spread
   of eigenvalues (energies) wrt to the k-points
   [file: dos.add.f90]

10.07.11 Version 3.24
^^^^^^^^^^^^^^^^^^^^^^
1. Atomic number and species are displayed in the Tasks menu of PD in the case of vasp,
   i.e. whne showing all projected DOS tasks, one can see the actual atoms numbers and their
   species.
   [files: dos_inc.f90, prep_dos.f90, read_vasp_psi2.f90]
2. NEW: Instead of calculating PDOS in the PD menu for a single task (e.g. s-projection on a particular atom),
   it is now possible to choose various sets of atoms and calculate either t,s,p or d -type PDOS
   on all these atoms, i.e. the sum of the elementary PDOS. The whole menu is revamped completely. 
   This is achieved with option CT. By repeatedly invoking this option, a new subset of PDOS tasks can be added,
   i.e. a complex list of tasks can be created if needed. A single task can be chosen/added as well using the old
   option 1.
   [files: prep_dos.f90, dos_add.f90, read_vasp_psi2.f90, dos_inc.f90, prep_disp.f90, choose_tasks.f90, Makefile ]

11.07.11 Version 3.25
^^^^^^^^^^^^^^^^^^^^^^
1. Relaxation flags are read in for any siesta input and printed initially with the atomic coordinates.
   [files: atoms.f90, read_siesta_input.f90, lev00.f90, read_vasp_geom.f90]
2. NEW: an option  Ck is added in the main menu for siesta to check the last atomic forces and whether the relaxation
   successfuly completed.
   [files: mainmenu.f90

20.07.11 Version 3.26
^^^^^^^^^^^^^^^^^^^^^^
1. An inconsistency with spins fixed (jspin was made to be either 1 or 2, but never 0)
   [file: mainmenu.f90]
2. A bug in checking siesta forces (option Ck): in some cases siesta prints forces differently, and that was now taken
   into account.
   [file: mainmenu.f90]

10.09.11 Version 3.27
^^^^^^^^^^^^^^^^^^^^^^
1. Bug fixed in DOS/PDOS routines (a little mess with allocations/deallocations)
   [files: prep_dos.f90, read_vasp_psi2.f90, do_tar_improved ]

26.10.11 Version 3.28
^^^^^^^^^^^^^^^^^^^^^^
1. Vasp 5.2 (or even 4.0 ?) may print out in OUTCAR eigenvalues at every iteration which before would result in
   read in only the very first set (not yet converged). This is now fixed: it searches for the very last iteration
   and only then it read the KS eigenvalues for each k-point.
   [file: do_param.f90]
2. Option 1 in the PDOS menu ("Choose/Add a single PDOS task") was not actually working (likely, it was not finished
   when new features ST and CT were added). Now fixed: using this option one can choose a single task or append to the
   existing list (e.g. after 1 or CT). 
   [file: prep_dos.f90]
3. A bug fixed related to the tasks and going to a new PICTURE.
   [file: prep_dos.f90]
4. px,py, pz contributions for PDOS from the vasp PROCAR file were read in a wrong order, now fixed; no effect this
   had on the P projection as only the sum of them was used anyway.
   [file: read_vasp_pis2.f90]

12.11.11 Version 3.29
^^^^^^^^^^^^^^^^^^^^^^
1. Gnuplot 4.4 (the latest at this time) has a different syntax, correspondingly the gnuplot feature did not work
   with the new version. This has been fixed in this version of lev00. Hopefully, backward compatibility is also
   ensured.
   [file: plotting.f90]

04.02.12 Version 3.30
^^^^^^^^^^^^^^^^^^^^^^
1. A missing command for 3D gnuplot plots was added (this one is needed for the Gnuplot 4.4)
   [file: plotting.f90]
2. Some cleaning of variables, etc.
   [files:lev_coulomb.f90, plot_add.f90, device.f90, prep_dos.f90 ]
3. NEW: Tersoff-Hamann routine (option TH in the D menu) added for working with STM images and spectroscopy. 
   It was so far only tested for VASP. It may work with SIESTA as well, but I need to check the SIESTA LDOS
   file first.
   This is how the STM image (spectroscopy) to be calculated:
      (i) run VASP normally to relax the structure and get CHGCAR; 
     (ii) run VASP again with more k-points and the following options in INCAR:

ISTART =1        [restart (0,1,2)]
PREC=HIGH
NSW = 0
ICHARG=11        [keeps the density fixed (DOS or BAND options)]
LPARD=.TRUE.     [for the local DOS]
MBMOD = -3       [in EINT the energy window is given wrt E_Fermi]
EINT = -1.0 0.0  [energy window: in this case all occ states 1 eV below the Fermi]

    (iii) The run above should create PARCHG file which lev00 can read
     (iv) run lev00 (you need OUTCAR, CONTCAR, PARCHG): 
          V -> OUTCAR -> CONTCAR -> D -> P -> TH
      (v) in the TH menu one can do either spectroscopy in up to 9 lateral points
          or a 2D/3D plot of the constant LDOS topography
    
     [files: stm_TH.f90, plotting.f90, density.f90]
4. In a number of places numerical menu items were changed into letters.
   [file: density.f90]

07.02.12 Version 3.31
^^^^^^^^^^^^^^^^^^^^^^
1. NEW: 8 options in total are provided for the 3D plots, some of them are really nice!
   [files: menu.f90, stm_TH.f90, plotting.f90]
2. It has been specifically checked that constant LDOS image can also be created from PARCHG file
   using gOpenmol menu and the value of the current chosen beforehand with the TH menu. This way it is possible
   also to have atoms displayed as well under the image.
          
09.02.12 Version 3.32
^^^^^^^^^^^^^^^^^^^^^^
1. Added an oprion El to the TH menu to enable/disable empty lines in the ldos.dat file for the LDOS
   [file: stm_TH.f90]
2. 'set size ratio 1' command added to test.gnu for 3D plots
   [file: plotting.f90]

27.02.12 Version 3.33
^^^^^^^^^^^^^^^^^^^^^^
1. wrong titles for X and Y axes in the STM spectroscopy; now swapped, so that they are correct.
   [file: stm_TH.f90]
2. wrong title for Z axis in 3D plots in STM routine; fixed.
   [file: stm_TH.f90]
3. error trap in inputing LDOS value in the STM menu (option I) has been put in
   [file: stm_TH.f90]
4. difiver was missing in the format in the PLot3D routine, now fixed
   [file: plotting.f90]