7. phaseshifts API

7.1. Package Contents

This chapter covers the main modules of the phaseshifts and provides some API documentation for those wishing to incorporate this package into their own projects.

7.2. Subpackages

The main sub packages are listed below:

  • phaseshifts.gui - includes all the necessary files for the graphical user interface.

  • phaseshifts.lib - contains the Fortran libphsh library and the python wrappings.

  • phaseshifts.doc - source documentation for the phaseshifts package.

  • phaseshifts.test - modules for testing the phaseshift package.

7.3. Submodules

7.3.1. phaseshifts.atorb

atorb.py

Provides convenience functions for generating input and calculating atomic charge densities for use with the Barbieri/Van Hove phase shift calculation package.

See:

http://www.icts.hkbu.edu.hk/surfstructinfo/SurfStrucInfo_files/leed/

Requires:

f2py (for libphsh fortran wrapper generation)

Note

To generate libphsh fortran wrappers (libphsh.pyd) for your platform then use ‘python setup.py’ in the lib directory of this package to install into your python distribution. Alternatively, use:

f2py -c -m libphsh libphsh.f

Windows users may have to add appropriate compiler switches, e.g.

# 32-bit
f2py -c -m libphsh --fcompiler=gfortran --compiler=mingw-32 libphsh.f

# 64-bit
f2py -c -m libphsh --fcompiler=gfortran --compiler=mingw-64 libphsh.f
class phaseshifts.atorb.Atorb(**kwargs)[source]

Bases: object

Notes

Original author: Eric Shirley

There are nr grid points, and distances are in Bohr radii \(a_0 \simeq 0.539 \mathrm{\AA}\)

\(r(i) = r_{min} \cdot (r_{max} / r_{min})^{(i/n_r)}\), \(i=1,2,3,...n_r-1,n_r\)

The orbitals are stored in phe(), first index goes \(1...n_r\), the second index is the orbital index (\(i...n_{el}\))

Look at the atomic files after printing this out to see everything… Suffice it to say, that the charge density at radius \(r(i)\) in units of electrons per cubic Bohr radius is given by:

\(\sum_{j-1}^{n_el}{occ(j) \cdot phe(i,j)^2 / (4.0\,\pi\,{r(i)^2)}}\)

Think of the phe functions as plotting the radial wave-functions as a function of radius on a logarithmic mesh…

The Dirac equation is solved for the orbitals, whereas their density is treated by setting \(phe(i,j)\) to Dirac’s \(\sqrt{F(i,j)^2 + G(i,j)^2}\) times the sign of \(G(i,j)\)…

So we are doing Dirac-Fock, except that we are not treating exchange exactly, in terms of working with major and minor components of the orbitals, and the phe’s give the CORRECT CHARGE DENSITY…

The above approximation ought to be very small for valence states, so you need not worry about it…

The Breit interaction has been neglected altogether…it should not have a huge effect on the charge density you are concerned with…

static calculate_Q_density(**kwargs)[source]
Parameters:
kwargs may be any of the following.
elementint or str, optional

Generate element atorb input file on the fly. Additional kwargs may be used to govern the structure of the input file - please use help(phaseshifts.Atorb.gen_input) for more information.

inputstr, optional

Specify atorb input file otherwise will use the class instance value.

output_dirstr, optional

Specify the output directory for the at_*.i file generated, otherwise the default current working directory is used.

Returns:
strfilename

Examples

>>> Atorb.calculate_Q_density(input='atorb_C.txt')
      18.008635    -33.678535
       4.451786    -36.654271
       1.569616    -37.283660
       0.424129    -37.355634
       0.116221    -37.359816
       0.047172    -37.360317
       0.021939    -37.360435
       0.010555    -37.360464
       0.005112    -37.360471
       0.002486    -37.360473
       0.001213    -37.360473
       0.000593    -37.360473
       0.000290    -37.360474
    N L M J S OCC.
    1   0 0  -1/2   1    2.0000        -11.493862
    2   0 0  -1/2   1    2.0000         -0.788618
    2   1 1  -1/2   1    0.6667         -0.133536
    2   1 1  -3/2   1    1.3333         -0.133311
 TOTAL ENERGY =      -37.360474  -1016.638262

>>> Atorb.calculate_Q_density(element='H')
       0.500007     -0.343752
       0.152392     -0.354939
       0.065889     -0.357254
       0.028751     -0.357644
       0.012732     -0.357703
       0.005743     -0.357711
       0.002641     -0.357712
       0.001236     -0.357713
       0.000587     -0.357713
       0.000282     -0.357713
 N L M J S OCC.
    1   0 0  -1/2   1    1.0000         -0.229756
 TOTAL ENERGY =       -0.357713     -9.733932
static gen_input(element, **kwargs)[source]
Parameters:
elementint or str

Either the atomic number, symbol or name for a given element

outputstr, optional

File string for atomic orbital output (default: ‘at_<symbol>.i’)

ngridint, optional

Number of points in radial grid (default: 1000)

relbool, optional

Specify whether to consider relativistic effects

filenamestr, optional

Name for generated input file (default: ‘atorb’)

headerstr, optional

Comment at beginning of input file

methodstr, optional

Exchange correlation method using either 0.0=Hartree-Fock, 1.0=LDA, -alpha = xalpha (default: 0.0)

relicfloat, optional

Relic value for calculation (default: 0)

mixing_SCFfloat, optional

Self consisting field value (default: 0.5)

tolerancefloat, optional

Eigenvalue tolerance (default: 0.0005)

echfloat, optional

(default: 100)

static get_quantum_info(shell)[source]
Returns:
tuple(int, int, list[float, float], list[float, float])

(n, l, j=[l-s, l+s], occ=[\(n^-_r\), \(n^+_r\)])

Notes

  • n is the principle quantum number (\(n > 0\)).

  • l is the azimuthal quantum number (\(0 \leq l \leq n-1\)).

  • s is the spin quantum number (\(s \pm \frac{1}{2}\)).

  • j is the total angular momentum quantum numbers for both \(l-s\) or \(l+s\), respectively.

  • \(n_r\) is the occupancy of the spin-split \(l-s\) and \(l+s\) levels, respectively.

static replace_core_config(electron_config)[source]
Parameters:
electron_configstr

String containing the electronic configuration of the given element.

Returns:
str

A substituted string where the nobel gas core has been replaced.

Examples

>>> Atorb.replace_core_config('[Ar] 4s2')
 '1s2 2s2 2p6 3s2 3p6 4s2'

>>> Atorb.replace_core_config('[Xe] 6s2 5d1')
 '1s2 2s2 2p6 3s2 3p6 3d10 4s2 4p6 5s2 4d10 5p6 6s2 5d1'
phaseshifts.atorb.get_electron_config(element_obj)[source]

Obtain the electronic orbital configuration for the given element_obj.

phaseshifts.atorb.get_element(element: str, backend: str | None = None) → object[source]

Obtain an element object for querying information using backend.

7.3.2. phaseshifts.conphas

conphas.py

Provides a native python version of the conphas (phsh3) FORTRAN program by W. Moritz, which is distributed as part of the SATLEED code (see “Barbieri/Van Hove phase shift calculation package” section) and can be found at: http://www.icts.hkbu.edu.hk/surfstructinfo/SurfStrucInfo_files/ leed/leedpack.html

The Conphas() class also provides a number of convenience functions (see docstrings below).

7.3.2.1. Examples

>>> from os.path import join
>>> from phaseshifts.conphas import Conphas
>>> con = Conphas(output_file=join('testing', 'leedph_py.d'), lmax=10)
>>> con.set_input_files([join('testing', 'ph1')])
>>> con.set_format('cleed')
>>> con.calculate()
class phaseshifts.conphas.Conphas(input_files=[], output_file=[], formatting=None, lmax=10, **kwargs)[source]

Bases: object

Class Conphas

Notes

This work is based on the original conphas (phsh3) FORTRAN program by W. Moritz, which is distributed as part of the SATLEED code (see “Barbieri/Van Hove phase shift calculation package” section) and can be found at: http://www.icts.hkbu.edu.hk/surfstructinfo/SurfStrucInfo_files/ leed/leedpack.html

__fix_path(file_path)

Fix escaped characters in filepath

__set_data(data=None)
calculate()[source]

Calculates continuous phase shifts from input file(s).

Examples

>>> con = Conphas(output_file=r'testing\leedph_py.d', lmax=10)
>>> con.set_input_files([r'testing\ph1'])
>>> con.set_format('cleed')
>>> con.calculate()
 L = 0
 jump between 25.0 eV and 30.0 eV; IFAK = -1
 L = 1
 jump between 65.0 eV and 70.0 eV; IFAK = -1
 L = 2
 jump between 20.0 eV and 25.0 eV; IFAK = 1
 jump between 80.0 eV and 85.0 eV; IFAK = 0
 L = 3
 L = 4
 jump between 275.0 eV and 280.0 eV; IFAK = 1
 L = 5
 L = 6
 L = 7
 L = 8
 L = 9
 L = 10
load_data(filename)[source]

Load (discontinuous) phase shift data from file

Parameters:
filestr

Path to phase shift file.

Returns:
tuple: (double, double, int, int, ndarray)

(initial_energy, energy_step, n_phases, lmf, data)

Notes

  • initial_energy is the starting energy of the phase shifts.

  • energy_step is the change in energy between consecutive values.

  • n_phases is the number of phase shifts contained in the file.

  • lmf is the maximum azimuthal quantum number considered.

  • data is a (2 x n_phases) array containing the phase shift data.

read_datafile(filename)[source]

Read in discontinuous phase shift file

Parameters:
filenamestr

The path to the discontinuous phase shift file

set_format(formatting=None)[source]

Set appropriate format from available options

Parameters:
formatstr, optional

The format identifier for different packages; can be ‘cleed’ or None.

set_input_files(input_files=[])[source]

set list of input filenames

set_lmax(lmax)[source]

Set max orbital angular momentum (azimuthal quantum number)

Parameters:
lmaxint

Maximum azimuthal quantum number to be considered in calculations.

set_output_file(output_file)[source]

set output filename

static split_phasout(filename, output_filenames=[])[source]

split phasout input file into separate files

7.3.3. phaseshifts.elements

Properties of the chemical elements.

Each chemical element is represented as an object instance. Physicochemical and descriptive properties of the elements are stored as instance attributes.

Author:

Christoph Gohlke

Version:

2013.03.18

7.3.3.1. Requirements

7.3.3.2. References

  1. http://physics.nist.gov/PhysRefData/Compositions/

  2. http://physics.nist.gov/PhysRefData/IonEnergy/tblNew.html

  3. http://en.wikipedia.org/wiki/%(element.name)s

  4. http://www.miranda.org/~jkominek/elements/elements.db

7.3.3.3. Examples

>>> from elements import ELEMENTS
>>> len(ELEMENTS)
109
>>> str(ELEMENTS[109])
'Meitnerium'
>>> ele = ELEMENTS['C']
>>> ele.number, ele.symbol, ele.name, ele.eleconfig
(6, 'C', 'Carbon', '[He] 2s2 2p2')
>>> ele.eleconfig_dict
{(1, 's'): 2, (2, 'p'): 2, (2, 's'): 2}
>>> sum(ele.mass for ele in ELEMENTS)
14659.1115599
>>> for ele in ELEMENTS:
...     ele.validate()
...     ele = eval(repr(ele))

7.3.4. phaseshifts.leed

Provides CLEED validator and Converter classes.

The CLEED_validator() class provides a method for checking the input files for errors, whereas the Converter.import_CLEED() method allows importing CLEED input files as a MTZ_model class

class phaseshifts.leed.CLEED_validator[source]

Bases: object

Class for validation of CLEED input files

static is_CLEED_file(filename)[source]

Determine if file is a CLEED input file

Returns:
True

if a valid filename ending in any of .bul, .inp, .bsr, .bmin

False

otherwise

validate(filename, aoi=False)[source]

Checks CLEED input file for errors

Parameters:
filenamestr

Path to input file. Should be *.bul , *.ctr, *.inp or *.bmin

aoibool

Check for angle of incidence parameters

class phaseshifts.leed.CSearch(model_name, leed_command=None)[source]

Bases: object

class for csearch related data exchange

_getResults()[source]
getIteration(iteration)[source]
getRFactor(iteration)[source]
getTimeStamp(iteration)[source]
setModel(name)[source]

sets the model name

class phaseshifts.leed.Converter[source]

Bases: object

Convert different input into phaseshift compatible input

static import_CLEED(filename, **kwargs)[source]

Imports CLEED input file and converts model to muffin-tin input.

It assumes the following:

  • the basis vectors \(a_1\), \(a_2\), & \(a_3\) are \(x\),:math:y,:math:z cartezian coordinates

  • if no \(a_3\) is found, the maximum \(z\) distance between atoms multiplied by four is given.

  • the unitcell is converted from cartezian to fractional coordinates

  • atom sites are converted from Angstrom to Bohr units

  • additional info from the phase shift filename is provided by splitting the ‘_’ chars:

    1. First string segment is element or symbol, e.g. Ni

    2. Second string segment is the oxidation (valence), e.g. +2

  • lines with rm: provide the radii dictionary of the individual phase shifts, whereas lines starting with lmax: provide the lmax dictionary for individual phase shifts. Additionally, the valency can be given in lines starting with ox:. Both bulk and surface input files will be searched for these.

  • if no rm: found for that species, the atomic radius is used for zero valence, otherwise the covalent radius is used.

  • if no lmax values are found for the specific phase shift, then the global value will be used instead.

  • if no valence values are found for the specific phase shift, then the guessed oxidation value from the phase shift filename is used instead. However, if the oxidation state cannot be parsed using the filename then a default oxidation state of zero is used.

Additional information can, however, be provided using ‘phs:’ at the start of a line within the input file and may have the following formats:

  1. “phs: c <float> nh <int> nform <int> exchange <float>”

  2. “phs: <phase_shift> valence <float> radius <float>”

The identifiers exchange, nform, valence and radius may be abbreviated to exc, nf, val and rad, respectively.

Information given in this manner overrides any other input.

Parameters:
filenamestr

Path to input file.

Returns:
phaseshifts.model.MTZ_model
Raises:
IOErrorfilename invalid
ValueErrorbad formatting of input

7.3.5. phaseshifts.model

model.py

Provides convenience functions for generating input and calculating atomic charge densities for use with the Barbieri/Van Hove phase shift calculation package.

class phaseshifts.model.Atom(element, coordinates=[0.0, 0.0, 0.0], **kwargs)[source]

Bases: object

Atom class for input into cluster model for muffin-tin potential calculations.

set_coordinates(coordinates)[source]
set_mufftin_radius(radius)[source]

Sets the muffin-tin radius of the atom in Angstroms.

set_valence(valency)[source]

Sets the valency of the atom

exception phaseshifts.model.CoordinatesError(msg)[source]

Bases: Exception

Coordinate exception to raise and log duplicate coordinates.

class phaseshifts.model.MTZ_model(unitcell, atoms, **kwargs)[source]

Bases: Model

Muffin-tin potential Model subclass for producing input file for muffin-tin calculations in the Barbieri/Van Hove phase shift calculation package.

calculate_MTZ(mtz_string='', **kwargs)[source]
Parameters:
atomic_filestr

The path to the atomic input file. If this is omitted the default is generate one using the MTZ_model.gen_atomic() method.

cluster_filestr

The path to the cluster input file which may be a bulk or slab model.

slabint or bool

Determines whether the MTZ calculation is for a slab model (True). The default is a bulk calculation.

outputdict

Dictionary output of ‘mtz’ - muffin-tin potential & ‘output_file’ - the path to the MTZ output file.

Returns:
output_fileslist(str)

Paths to the MTZ output file.

create_atorbs(**kwargs)[source]
Returns:
output_filesdict

Dictionary list of atorb*.i input files for the Atorb class to calculate the charge density from.

gen_atomic(**kwargs)[source]
Parameters:
input_dirstr

Input directory where at*.i files are kept.

input_filestuple

List of input files to generate atomic input file from.

output_filestr

The filename of the resulting atomic*.i output file, which is simply a superimposed set of the radial charge densities from the individual input files.

Returns:
output_filestr

Returns the output file path string.

Raises:
IOErrorexception

If either input directory or files do not exist.

Notes

If ‘input_files’ is not given then the default list of input files are inferred from the list of atoms in the model.

gen_input(**kwargs)[source]
Returns:
filename on success
Raises:
CoordinatesErrorexception

if the model atoms have duplicate coordinates and the ‘pos_check’ kwarg is set to True.

get_MTZ(filename)[source]

Retrieves muffin-tin potential from file

get_elements()[source]

Return the unique elements in model

load_from_file(filename)[source]
Parameters:
filenamestr

The path of the input file (e.g. cluster*.i or slab.i)

Raises:
IOErrorexception

If the file cannot be read.

TypeErrorexception

If a input line cannot be parsed correctly.

set_exchange(alpha)[source]

Sets the alpha exchange term for muffin-tin calculation

set_nform(nform)[source]

Sets form of muffin-tin calculation

Parameters:
nformint or str

This governs the type of calculation, where nform can be:

1. “cav” or 0 - use Cavendish method

2. “wil” or 1 - use William’s method

3. “rel” or 2 - Relativistic calculations

set_nh(nh)[source]

Sets the nh muffin-tin zero estimation parameter

set_slab_c(c)[source]

Examples

For Re the bulk c distance is 2.76Å, whereas a possible slab c distance could be ~10Å.

class phaseshifts.model.Model(unitcell, atoms, **kwargs)[source]

Bases: object

Generic model class.

_nineq_atoms()[source]
Returns:
nineq_atoms, element_dicttuple
nineq_atomsThe estimated number of inequivalent atoms based on

the valence and radius of each atom.

element_dicta dictionary of each element in the atom list where

each element contains an atom dictionary of ‘nineq_atoms’, ‘n_atoms’ and a complete ‘atom_list’

add_atom(element, position, **kwargs)[source]

Append an Atom instance to the model

Parameters:
elementstr or int

Either an element name, symbol or atomic number.

positionlist(float) or ndarray

(1x3) array of the fractional coordinates of the atom within the unit cell in terms of the lattice vector a.

check_coordinates()[source]

Check for duplicate coordinates of different atoms in model.

Raises:
CoordinateErrorexception

If duplicate positions found.

set_atoms(atoms)[source]

Set the atoms for the model.

Parameters:
atomslist(Atoms)

Array of Atom instances. Entries in the list which are not of type Atom will be ignored.

Raises:
TypeErrorexception

If atoms parameter is not a list.

set_unitcell(unitcell)[source]

Set the unitcell for the model

Parameters:
unitcellUnitcell

Instance of Unitcell class to set to model.

Raises:
TypeErrorexception

If unitcell parameter is not a Unitcell.

class phaseshifts.model.Unitcell(a, c, matrix_3x3, **kwargs)[source]

Bases: object

Unitcell class

set_a(a)[source]
Parameters:
a: float

The magnitude of the in-plane lattice vector in Angstroms

Notes

To retrieve a in terms of Angstroms use ‘unitcell.a’, whereas the internal parameter ‘unitcell._a’ converts a into Bohr radii (1 Bohr = 0.529Å), which is used for the muffin-tin potential calculations in libphsh (CAVPOT subroutine).

set_alpha(alpha)[source]
set_beta(beta)[source]
set_c(c)[source]
Parameters:
cfloat

The magnitude of the in-plane lattice vector in Angstroms

Notes

To retrieve c in terms of Angstroms use ‘unitcell.c’, whereas the internal parameter ‘unitcell._c’ converts c into Bohr radii (1 Bohr = 0.529Å), which is used for the muffin-tin potential calculations in libphsh (CAVPOT subroutine).

set_gamma(gamma)[source]
set_vectors(m3x3)[source]

7.3.6. phaseshifts.phsh

phsh.py - quickly generate phase shifts

phsh provides convenience functions to create phase shifts files suitable for input into LEED-IV programs such as SATLEED and CLEED.

7.3.6.1. Examples

phsh.py -i *.inp -b *.bul -f CLEED -S phase_dir