The Atoms and Feff Window

When you import crystal data from an atoms.inp or CIF file, three things happen:

1. A new ATOMS and FEFF window is created for interacting with the structural data and the various controls are set to values taken from the atoms.inp or CIF file,

2. A message is written to the status bar in the Main window.

3. An entry is placed in the FEFF list on the main window.

You can also import a feff.inp file directly. This is discussed in the next section

This new window looks like this. In this example, crystal data for anatase TiO₂ have been imported from an atoms.inp file.

At the top of the window is a tool bar with four buttons. The first of these is used to change the name of this FEFF calculation. Among other things, this is the label used in the FEFF list on the Main window. The second button is used to discard the FEFF calculation and this window. The final two buttons open a web browser and take you either to the FEFF document or to this page in the ARTEMIS document.

There are a series of tabs across the top. These will contain different stages of the structural calculation. Here we will examine the ATOMS tab. The other tabs will be examined in the following sections.

---

 

Crystal data

The crystal cell data – including lattice constants and angles, the space group symbol, and the elements of the shift vector – are placed in text boxes for easy editing. The coordinates of the unique sites are listed in the grid at the bottom of the window. The absorber is chosen by clicking one of the boxes on the Core column.

Remember that FEFF considers numbers with 5 digits of precision after the decimal point. 0.333 is not the same thing as 0.33333. You may, however, enter things like 1/3 and avoid the precision issue entirely.

As a new feature compared to earlier versions of ATOMS, there are two radial distances. The cluster size determines the extent of the cluster expanded into the feff.inp. This should usually be set to something rather large, 9 Å is often a good default. This probably (but not always!) assures that the cluster in the feff.inp file is adequately large to include all unique potentials and has all atom types sufficiently well bounded that the muffin tin potentials are likely to be be computed reasonably well.

The second distance will set the value of RMAX in the feff.inp file. In general, you do not want this to be much larger than the extent of the data you intend to analyze. 5 Å or 6 Å is usually the largest sensible value for longest path. The reason for this is that the pathfinder part of FEFF has been rewritten for this version of ARTEMIS. While the new pathfinder implementation offers a number of useful new features, it is substantially slower than FEFF's native pathfinder. In any case, there is no benefit to computing paths that you will never use in your fit.

The absorption edge for the calculation is chosen from the menu to the left of the lattice constant area. This is usually determined from the input data, but may need to be explicitly selected. If not specified in the atoms.inp file, the edge will be set to K for element lighter than Ce (Z=58), and to L_III for heavier elements.

The style menu is another new feature in this version of Atoms. It is used to set how the list of unique potentials is determined from the elements in the atoms list. The choices are

elements

Each unique element species is assigned a potential number.

tags

Each unique tag is assigned a potential number.

sites

Each crystallographic site is assigned a potential number.

Remember that FEFF only allows for 7 unique potentials other than the absorber. The tags and sites options can often result in more than 7 potnatials, which will result in an unrunnable feff.inp file. Specifying unique potentials by tags is a good way of differentiating between dissimilar atoms of the same species. For example, in an oxygenyl species, it is often useful to give the axial oxygen atoms a different potential from the remaining oxygens by using the tags option.

Here is an example of an atoms.inp file for sodium uranyl acetate, which contains two very short axial oxygen atoms double bonded to the uranium atoms at about 1.8 Å and a number of equatorial oxygen atoms at a much longer distance. The axial and equatorial oxygen positions are distinguished by their tags and will given separate unique potentials when using the tags style.

The assignment of potential indeces is explained in detail and with examples in the extended explanations chapter.

title        Templeton et al.
title        Redetermination and Absolute configuration of Sodium Uranyl(VI) triacetate.
title        Acta Cryst 1985 C41 1439-1441
space = P 21 3
a =   10.6890        b =      10.6890        c =      10.6890
alpha =       90.0   beta =   90.0   gamma =  90.0
core =       U       edge =  L3
atoms
! elem   x          y          z       tag        occ
  U     0.42940    0.42940    0.42940  U             1.00000
  Na    0.82860    0.82860    0.82860  Na            1.00000
  O     0.33430    0.33430    0.33430  Oax           1.00000
  O     0.52420    0.52420    0.52420  Oax           1.00000
  O     0.38340    0.29450    0.61100  Oeq           1.00000
  O     0.54640    0.24430    0.50070  Oeq           1.00000
  C     0.47860    0.22600    0.59500  C             1.00000
  C     0.50880    0.12400    0.68620  C             1.00000

---

 

Polarization

A FEFF calculation considering linear polarization can be triggered by setting one or more non-zero values for the polarization vector.

This vector sets the value of the POLARIZATION keyword in the resulting feff.inp. The value written in the feff.inp file is the value that will be used in the pathfinder and when computing the path contributions. That is, if you edit the POLARIZATION in the feff.inp file, the edited value will take precedence over the value specified here.

FEFF's ELLIPTICITY keyword is not supported at this time. That means the trick of modeling “polarization in the plane” is not yet supported by ARTEMIS.

---

 

Atoms tool bar

The toolbar across the top of the ATOMS tab offers several functions.

Clicking the open button will post the standard file selection dialog for importing a new atoms.inp or CIF file. This is more useful in the stand-along version of ATOMS than in ARTEMIS where the crystal data file imported in other ways. Right clicking this button will post the recent files dialog populated with recently imported atoms.inp, feff.inp, and CIF files.

The save button will prompt you for a filename for an output atoms.inp saving the current state of the tab.

Clicking the export button will post the dialog on the right, which offers several different kinds of output files based on the crystal data.

• The “Feff6” and “Feff8” options will write input files for FEFF6 and FEFF8.

• The “Atoms” option write the same file as the save button.

• The “P1” option writes the crystal data to an atoms.inp file using the P 1 space group and with a fully decorated unit cell. The “Spacegroup” option writes a file that fully describes the space group.

• The “Absorption” option writes a file containing some calculations based on tables of X-ray absorption coefficients.

• “XYZ” and “Alchemy” are formats that are commonly understood by molecule rendoring software.

• “Overfull” is an “XYZ” file with the contents of the unit cell in Cartesian coordinates and with all atoms near a cell wall replicated near the opposite cell wall. The purpose of this output type is generate nice figures of unit cells with decorations on all the corners, sides, and edges.

The clear button is used to clear all data from all controls on the ATOMS tab.

The run button is pressed to convert the crystal data into input data for FEFF. Pressing this displays the next tab.

The aggregate button is discussed in detail in a later section