BasIreps

A program for calculating irreducible representation of

little groups and basis functions of polar and axial vector
properties.

Juan Rodríguez-Carvajal ( LLB 2004)

Package information
Actually we are distributing two executable files bellowing to BasIreps distribution.
The first is the console version of BasIreps (basireps.exe) and the second is the GUI
program for the console version (gbasireps.exe). GBasIreps is able to generate the input
control file for BasIreps and run the program in a silent mode showing automatically
the final files generated by BasIreps.

Both are included on the FullProf Suite package.

Running the program

For a rapid view of BasIreps, run the GUI GBasIreps by clicking on it, or invoking it
from a console or from the program EdPCR.

You may run directly BasIreps from a console as follows:

Local_Prompt> basireps My_file <cr>

My_file is the code of the input file summarizing the instructions for running BasIreps.
The complete name, by default, is "My_file.smb"; however any extension can be used
and it must be given in general cases.

To access the help file of BasIreps from GBasIreps the environment variable
FULLPROF has to be defined. It points to a directory where the executable files have to
be stored.

These notes are given to facilitate the use of the program. At present there is no detailed
manual for BasIreps. For someone with sufficient skills in group theory the output files
are self explanatory.

The program

The program BasIreps calculates the irreducible representations (irreps) of the so called
"little groups" from which the full irreducible representations of space groups can be
calculated using the induction formula. The method used in BasIreps for calculating the
irreps is based in the procedure used in the program KAREP by E. Hovestreydt, I.
Aroyo et al, J.Appl.Cryst. 25, 544 (1992). The method of Zak used in KAREP has been
implemented in a special subroutine within BasIreps.

After calculating the irreps corresponding to a given space group and propagation vector
the program calculates the basis functions corresponding to vectorial properties (atom
displacements or magnetic moments) of atoms in crystalline solids.

This program calculates non-normalized basis functions of the irreducible

representations of the little group Gk for atom properties in a crystal. In particular the
calculations can be performed for atomic displacements (or phonon modes), as polar
vectors, and for magnetic moments as axial vectors. In general the
displacement/magnetic moment of atom j in cell L may be written as a Fourier series of
the form:

m jL = ∑{S kj exp( −2π i k ⋅ R L )}

k

k and RL are vectors referred to the reciprocal and direct crystallographic basis
respectively. The vectors m jL and Skj have the same units and are referred to a basis
of unit vectors along the direct crystallographic cell basis. The vectors S kj are the
Fourier components of the magnetic moments (or displacements) m jL .

Taking into account the symmetry, the vectors S kj may be written as linear
combinations of the so-called basis functions of the irreducible representations of the
propagation vector group Gk . The number of free parameters in a displacive phase
transition, or in a magnetic structure, is less than N = 3 × n × O {k} . The number 3 comes
from the three components of S kj , n is the total number of atoms in a primitive cell and
O {k} is the number of propagation vectors. The number of independent free
parameters (order parameters) can be calculated from group theory. In general the
vector S kj may be written as:

{
S kj = ∑ Ca ,m Va ,m ( k ,ν j )}
a ,m

Where Ca ,m are the coefficients (that may be real or pure imaginary) of the linear
combination, and Va ,m ( k ,ν j ) are constant vectors referred to the basis of the direct
cell. The labels are the following ( k ,ν ) is for the particular propagation vector and the
representation called Γν . The index a varies from 1 up to the dimension of the
irreducible representation: a=1,2,...dim( Γν ). The index m varies from 1 to the number
of times the irreducible representation Γν is contained in the global mechanic/magnetic
reducible representation constructed by working with the symmetry operators acting on
atoms coordinates and components of polar/axial vectors (dimension 3 × n). See, for
instance, the book by Y.A. Izyumov, V.E. Naish and R.P. Ozerov (1991) for details.

The program BasIreps calculates the vectors Va ,m ( k ,ν j ) in a non-normalized form.

However, the final form of the basis functions is written as simple as possible. An effort
in future versions will be performed to recognize particular combinations of common
numbers.

The program BasIreps is written in a subset of the Fortran 95 language called F. It is

based in the Crystallographic Fortran Modules Library (CrysFML), by Juan Rodriguez-
Carvajal and Javier Gonzalez-Platas.

Input file
The file is written in free format. Each line starts with a keyword and a list of values.
Whatever line starting with "!" or "#" (without quotes) in the first column are
considered as a comment.

The minimum input for running the program is constituted by three lines with the
keywords TITLE, SPGR (or GEN) and KVEC. The keywords are case sensitive, so
only upper-case keywords are allowed.

An example of minimal input file for BasIreps is:

TITLE Minimum content of *.smb file

SPGR P n m a
KVEC 0.5000 0.0000 0.0000 X

Another examples is:

TITLE Minimum content of *.smb file

GEN x+1/2,y,-z ; x,-y,z ; -x,-y,-z
KVEC 0.0000 0.2500 0.0000 U

The values associated with the above keywords are the following:

TITLE
Whatever string of characters for identification purposes

SPGR
A string containing the Hermann-Mauguin symbol of the space group.
The different symmetry directions have to be separated by a blank character. The
program interprets the space group symbol (Hermann-Mauguin or Hall notations are
allowed) and generates an output with symmetry information about the provided space
group.

GEN
A string containing the Jones faithful representation of the generators. The character
";" (without quotes) is used as separator for the different generators.
KVEC
Components of the k-vector (three real numbers) followed by one or two letter for
labelling the k-vector (e.g. X, M, LD, UV ...). The components should be given with
respect to the reciprocal basis of the conventional unit cell, as is commonly used in
diffraction physics and crystallography.

With this minimal input the program calculates only the small representations of the
propagation vector group (little group). To calculate basis functions the program needs
the following keywords: BASIR and ATOM

BASIR
This keyword should be followed by two words.

The first one can be AXIAL or POLAR and the second must be CEL if one wants the
program to put all generated atoms within the reference unit cell (all fractional
coordinates >= 0 and < 1.0). AXIAL is adequate for the analysis of magnetic
structures, and POLAR is useful for studying phonons (normal vibration modes) or
structural phase transitions.

ATOM
This keyword introduces the information about one atom in the unit cell. Two labels are
needed for the atoms. The second one should be the name of the element. This is not
used internally but it is convenient for compatibility with other programs reading the
same file. The two labels are followed by three fractional coordinates. The program
generates the rest of atoms in the unit cell.

ATOM Cu Cu 0.2341 0.5000 0.00000

SUBL
This keyword is an alternative to ATOM for the case the numbering of sublattices is to
be controlled by the user. This keyword must be followed in the same line by the label of
the atom and the number of sublattices, NSUB, corresponding to the site.
The coordinates of the NSUB atoms must be given in the following NSUB lines starting
at least from the 5th column. An arbitrary label may be put in the first 4 columns. The
program will use the common label of the atoms adding and underscore "_" followed by
the number of the sublattice.
For instance, if in the input file there are the lines
......
SUBL Tb 4
At 0.00000 0.00000 0.50000
At 0.25000 0.75000 0.75000
At 0.50000 0.00000 0.00000
At 0.25000 0.25000 0.25000
.......

the program will generate the labels Tb_1, Tb_2, Tb_3 and Tb_4 for numbering the
atoms.
Examples
A complete example of input file to be used by BasIreps is:

TITLE Minimum content of *.smb file

SPGR C m c a
KVEC 0.0000 0.0000 1.0000 Z
BASIR AXIAL CEL
ATOM Fe Fe 0.0000 0.0000 0.47510
ATOM Co Fe 0.2341 0.5000 0.00000

Another example in which the sublattices are explicitly given in a particular order is the
following:

TITLE Test Basireps

SPGR I 41/a m d
KVEC 0.5000 0.5000 0.5000 X
BASIR AXIAL
SUBL Tb 4
At 0.00000 0.00000 0.50000
At 0.25000 0.75000 0.75000
At 0.50000 0.00000 0.00000
At 0.25000 0.25000 0.25000
SUBL Fe 4
At 0.00000 0.00000 0.00000
At 0.50000 0.00000 0.50000
At 0.75000 0.25000 0.75000
At 0.75000 0.75000 0.25000

GBasIreps
The GUI to BasIreps has the following important fields that are used to generate input,
"codefile.smb", and output files (after running BasIreps from the corresponding menu
or button): "codefile.bsr" and "codefile.fp"

Code of files
Common code of files for the run

Working Directory
Complete path name of the working directory. This may be selected using the Browse
button.

Title
A character string used as a title for the run

SpaceGroup or generators
A character string with different meanings. The user may input different things:

• A number for the space group between 1 and 230

• The Hermann-Mauguin symbol of the space group using one space for
separating symmetry directions. Example: I 41/a m d
 • The Hall symbol of the space group using also one space for separating
symmetry directions. Example: -I 4bd 2
• List of space group generators in Jone's faithful notation and separated by
semicolons.
Example: x,-y,z+1/2 ; -x,y,z ; -x,-y,-z

K-vector
Components of the propagation vector in reciprocal lattice units. The reciprocal basis
vectors are defined, through the well known dual relations, using direct basis vector of
the conventional unit cell. This is the common practice in diffraction physics and
crystallography.

Warning: The components of k-vectors in most textbooks of group theory are given
with respect to the reciprocal basis of the primitive direct basis vectors. For primitive
Bravais lattices there is no problem.

Brillouin Zone Label

Whatever character used for identification purposes. Up to two letters!

Example: GG

Polar vector/Axial vector

Checking one of the two alternatives will select the behaviour of the symmetry operators
upon acting on vectors.

Atoms in unit cell

If this box is checked the program will put all generated atoms within the reference cell
all coordinates values (v) verifying 0.0 <= v < 1.0. This is only operative if the box
"Explicit Sublattices" is not checked.

Number of atoms, Explicit Sublattices and Atom positions

In the case the box "Explicit Sublattices" IS NOT CHECKED

The "Number of atoms" coincides with the number of sites (NS). The program will
generate automatically all sublattices and, eventually, modify the number of sites if the
propagation vector group cannot connect part of the sublattices generated by the full
space group. The "Symbol" of the atom should have a maximum of four characters.
Usually the chemical symbol is enough. The fractional coordinates are given in the
fields "x/a", "x/b", "x/c". The GUI generates NS ATOM lines.

In the case the box "Explicit Sublattices" IS CHECKED

The "Number of atoms" is really the total number of atoms in the unit cell. In this case
the name of the atoms cannot be arbitrarily given. For a particular site they must share
the same atom code and the different sublattices are distinguished by an added
underscore "_" followed by the number of the sublattice. The GUI generates a list of
SUBL block of lines, deduced from the "Symbol". For example, to generate the SUBL
blocks in the example of SUBL keyword above we have to write Number of atoms = 8,
check the "Explicit Sublattices" box and provide the list:
 Tb_1 0.00000 0.00000 0.50000
Tb_2 0.25000 0.75000 0.75000
Tb_3 0.50000 0.00000 0.00000
Tb_4 0.25000 0.25000 0.25000
Fe_1 0.00000 0.00000 0.00000
Fe_2 0.50000 0.00000 0.50000
Fe_3 0.75000 0.25000 0.75000
Fe_4 0.75000 0.75000 0.25000

BE CAREFUL! if the provided sublattices do not correspond are not related by

operators in the propagation vector group (little group) the obtained results are wrong!
If you are not sure it is safer to use the first atom of sites and let the program generate
the sublattices.

Output files
The program generates two output files of extension '.bsr' and '.fp'. The first file
contains the full information generated by BasIreps. The second file contains
information in a format suitable to be pasted in the PCR file for FullProf.

As stated above, the content of the output files is self-explanatory.

Sometimes, mostly when complex multidimensional representations are concerned, the

number of calculated basis functions is greater than what is expected. That may be due
to propagation errors in the calculation of the rank of a general complex matrix. If this
situation appears for a particular case the user should inspect the obtained basis
functions and try to extract the linear independent basis functions from the provided set.

From the version 3.0, a formal writing of the Fourier coefficients S kj is output in the
file of extension *.bsr. Symbols for the coefficients Ca ,m are used in the following
order:

"u","v","w","p","q","r","a","b","c","d","e","f","g","l","m","n","s","t",
"U","V","W","P","Q","R","A","B","C","D","E","F","G","L","M","N","S","T"

If particular real numbers appear in the components of the basis functions (this may be
due, for instance, to incommensurate propagation vectors) the program tries to
recognize them and in the writing of the Fourier coefficients these special numbers
appear in symbolic form. The numbers are given the following names in the order of
appearance:

"r0","r1","r2","r3","r4","r5","r6","r7","r8","r9",
"p0","p1","p2","p3","p4","p5","p6","p7","p8","p9",
"q0","q1","q2","q3","q4","q5","q6","q7","q8","q9",
"s0","s1","s2","s3","s4","s5","s6","s7","s8","s9",
"t0","t1","t2","t3","t4","t5","t6","t7","t8","t9"

The values of these numbers are given in separate lines with more precision than that
used for writing the basis vectors Va ,m ( k ,ν j ) .
An example of output concerning this last point is given below. The calculation is
performed for the following case:
 TITLE Test Basireps
SPGR P 4/n m m
KVEC 0.5000 0.2341 0.0000 X
BASIR AXIAL CEL
ATOM Mn Mn 0.2500 0.2500 0.1320

.....

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
=>Basis functions of Representation IRrep( 1) of dimension 2 contained 3 times in GAMMA
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

SYMM x,y,z -x+1,y+1/2,-z+1

Atoms: Mn_1 Mn_2
1:Re ( 1.00 0.00 0.00) ( 0.74 0.00 0.00)
Im ( 0.00 0.00 0.00) (-0.67 0.00 0.00)
2:Re ( 0.00 1.00 0.00) ( 0.00-0.74 0.00)
Im ( 0.00 0.00 0.00) ( 0.00 0.67 0.00)
3:Re ( 0.00 0.00 1.00) ( 0.00 0.00 0.74)
Im ( 0.00 0.00 0.00) ( 0.00 0.00-0.67)
4:Re (-0.74 0.00 0.00) ( 0.10 0.00 0.00)
Im ( 0.67 0.00 0.00) (-1.00 0.00 0.00)
5:Re ( 0.00 0.74 0.00) ( 0.00 0.10 0.00)
Im ( 0.00-0.67 0.00) ( 0.00-1.00 0.00)
6:Re ( 0.00 0.00 0.74) ( 0.00 0.00-0.10)
Im ( 0.00 0.00-0.67) ( 0.00 0.00 1.00)

----- LINEAR COMBINATIONS of Basis Functions: coefficients u,v,w,p,q ....

General expressions of the Fourier coefficients Sk(i) i=1,2,...nat

SYMM x,y,z Atom: Mn_1 0.2500 0.2500 0.1320

Sk(1): (u-r0.p,v+r0.q,w+r0.r)+i.r1.(p,-q,-r)

SYMM -x+1,y+1/2,-z+1 Atom: Mn_2 0.7500 0.7500 0.8680

Sk(2): (r0.u+r2.p,-r0.v+r2.q,r0.w-r2.r)+i.(-r1.u-r3.p,r1.v-r3.q,-r1.w+r3.r)

Values of real constants r0,r1,...

r0 = 0.741531 r1 = 0.670919 r2 = 0.099737 r3 = 0.995014

Check combinations of values by pairs: usually these real constants are related to k-
vector. They can constitute real and/or imaginary parts of exp{2.pi.i.K.T }, being T a
non-primitive translation of a symmetry operator. In many simple cases r0=cos(2.pi.k.t)
and r1=sin(2.pi.k.t), etc ...

.....

One can see that the constant r2 is written as "0.10" in the basis vectors and the
constant r3 does not appear because it is approximated as "1.00" in the format used
for writing the numerical vectors.
The meaning of these constants are related, as suggested in the comment of the output,
to the propagation vector k=(0.5000, 0.2341, 0.0000). The complex numbers r0+i.r1
and r2+i.r3 are:
r0+i.r1 = exp{pi.i.ky} and r2+i.r3 = exp{2.pi.i.ky}

This suggests a further simplification of the general Fourier coefficients that can be
worked out by the user:

Sk(1)=(u-r0.p,v+r0.q,w+r0.r)+i.r1.(p,-q,-r)=(u,v,w)+r0.(-p,q,r)+i.r1.(p,-q,-r)
Sk(1) = (u,v,w) + r0.(-p,q,r)-i.r1.(-p,q,r) = (u,v,w) + (r0-ir1).(-p,q,r)
Sk(1) = (u,v,w) + (-p,q,r) .exp{-pi.i.ky}

Sk(2) = (r0.u+r2.p,-r0.v+r2.q,r0.w-r2.r)+i.(-r1.u-r3.p,r1.v-r3.q,-r1.w+r3.r)
Sk(2) = r0 (u,-v,w) + r2(p,q,-r) + i.r1.(-u,v,-w) +i.r3 (-p,-q,r)
Sk(2) = (r0-i.r1).(u,-v,w)+(r2-ir3).(p,q,-r)
Sk(2) = (u,-v,w).exp{-pi.i.ky}+(p,q,-r).exp{-2.pi.i.ky}
Disclaimer
The program BasIreps is distributed in the hope that it will be useful, but WITHOUT
ANY WARRANTY of being free of internal errors. In no event will the author be liable
to you for damages, including any general, special, incidental or consequential damages
arising out of the use or inability to use the program (including but not limited to loss of
data or data being rendered inaccurate or losses sustained by you or third parties or a
failure of the program to operate with any other programs).

Recommended references:

• International Tables for Crystallography, Volume A: Space Group Symmetry.

Edited by Theo Hahn, published by the IUCr, D. Reidel Publishing Company,
1983.

• Isotropy subgroups of the 230 crystallographic space groups

Harold T. Stokes and Dorian M. Hatch, Word Scientific, 1988, ISBN 9971-50-
772-2.

• Commensurate and Incommensurate Phase Transitions

Jerzy Kocinski, ELSEVIER, 1990. ISBN 0-444-98775-4
This is the third volume of a series called Phase Transition Phenomena.

• Representation of the Crystallographic Space Groups. Irreducible

Representations, Induced Representations and Corepresentations. 2nd Edition
O.V. Kovalev, Ed. by Harold T. Stokes and Dorian M. Hatch, Gordon and
Breach Science Publishers, 1993. ISBN 2-88124-934-5

• The mathematical theory of symmetry in solids

C.J. Bradley and A.P. Cracknell, Clarendon Press (Oxford University Press, Ely
House, London W.1), 1972.

For magnetic structure symmetry analysis see the following references:

• E. F. Bertaut, Acta Cryst. A24, 217 (1968).

• E. F. Bertaut, Spin Configurations in Ionic Structures: Theory and Practice, in

Magnetism, vol 3, Ed. G.T. Rado and H. Suhl, Academic Press, 1963.

• Y.A. Izyumov, V.E. Naish, J Magn.Magn.Mat. 12,239 (1979); Y.A. Izyumov,

V.E. Naish and V.N. Syromiatnikov J Magn.Magn.Mat. 12, 249 (1979); Y.A.
Izyumov, V.E. Naish and S.B. Petrov, J Magn.Magn.Mat. 13, 267 (1979); Y.A.
Izyumov, V.E. Naish and S.B. Petrov, J Magn.Magn.Mat. 13, 275 (1979); Y.A.
Izyumov, J Magn.Magn.Mat. 21, 33 (1980).

• Y.A. Izyumov, V.E. Naish and R.P. Ozerov,"Neutron diffraction of magnetic

materials", Consultants Bureau, Plenum Publishing Corporation, New York
(1991)