Next: Programs for Manipulation of Up: McPhase USERS MANUAL Previous: Fitting is an art: Contents Index

Programs for Specific Tasks in the World of McPhase

This section contains some further programs which do some important manipulations in magnetism, ...

addj file1.j file2.j:

adds exchange parameters of file2.j to the parameters in file1.j - output is written to stdout

anisotropy:
program to calculate the magnetic anistropy by doing a mcphas or single ion calculation for different external magnetic field directions and evaluating the expectation value of the magnetic moment.

     usage: anisotropy -h
           anisotropy  T H xn yn zn nofsteps [-r sipffilename Hxc1 Hxc2 ... Hxcnofcomponents]

     -h           : this (help) message
      T           : temperature in Kelvin
      H           : absolute value of the external magnetic field (T)
      xn,yn,zn    : direction normal to plane, in which the anisotropy
                    should be calculated ... e.g. if you want to
                    calculate the anisotropy in the xy plane, then
                    enter xn yn zn = 0 0 1
      nofsteps    : number of steps to be calculated 

    option:
    -r sipffilename: filename of single ion parameter file
                      Hxc1,Hxc2,... are the exchange field components (meV)
                     (exchange field is kept constant, external magnetic
                     field is rotated in the anisotropy calculation)

    output files:

    ./results/anisotropy.out  contains anisotropy information

cfsplit:

Calculates via group theory, the multiplicity of crystal field split levels for particular point symmetries. The command is: "cfsplit $\langle$ PTGP $\rangle$ $\langle$ J $\rangle$ " where $\langle$ PTGP $\rangle$ is the name of the point group (both Hermman-Maguin and Schoenflies notation are understood), and $\langle$ J $\rangle$ is the half-integer value of the total angular momentum J of the lowest multiplet. E.g. "cfsplit Oh 2" gives the familiar cubic Eg/T2g splitting (last line of output). See also: symhmltn which calculates the allowed two-ion exchange terms for a point group.

cpsingleion 10 100 1 file.levels.cef [options]:

By runnning singleion some file.levels.cef is created in folder results. The cpsingleion program may then be used. It calculates the specific heat in the temperature interval 10-100 K with a step width of 1 K. Alternatively a comparison to experimental data can be made by cpso1ion 1 2 cpexp.dat, where the temperatures are given in column 1 and the experimental specific heat in column 2 of file cpexp.dat. The calculated specific heat is compared to the experimental data and a standard deviation sta is calculated and output is written to stdout. Other quantities can be calculated using the options: -s (calculate entropy (J/molK) instead of cp), -f (calculate free energy (J/mol) instead of cp),-u (calculate magnetic energy (J/mol) instead of cp), -z (calculate partition sum instead of cp)

cpso1ion 10 100 1 [options]:

same as cpsingleion but for output of program ic1ion (no file.levels.cef required, takes values from so1ion.out) .

cpic1ion 10 100 1 [options]:

same as cpsingleion but for output of program ic1ion (no file.levels.cef required, takes values from ic1ion.out) .

cpic1ion 10 100 1 [options]:

same as cpsingleion but for output of program ic1ion (no file.levels.cef required, takes values from icf1ion.out) .

epsdebye Tmax dT Tdebye scale [d1 d2 datafile]:

calculates the phonon induced strain $\epsilon$ using the debye model according to the following formula: $\epsilon=S*T*D(\Theta_{D}/T)$ with $D(z)=\frac{3}{z^3}\int_0^z \frac{x^3 dx}{e^x-1}$ Range is from zero to Tmax in stepwidths dT unless a datafile is given. If a datafile is given, with data column d1 and d2,the strain is calculated for T-values of data column d1 and epsilon is compared to data in column d2 - a standard deviation sta is calculated as a sum of squared deviations. As output the datafile is given, an additional is column added containing the calculated strain epsilon. The datafile has to be sorted according to descending T values !!! output is written to stdout.

extendunitcell n1 n2 n3

program to extend crystallographic unit cell n times in r1 (or r2,r3) direction, meaning take mcphas.j, mcphas.tst and mcdiff.in and generate an extended description of the unit cell n1xr1,n2xr2,n3xr3 put result into extend.j, extend.tst and extend.in

fermicol col T filename

calculates the Fermifunction from energy in column col.

col

column containing energy values (eV) relative to EF

T

temperature (K)

filename

file name

fitfermi T EF fwhm min max filename

fits a (Gaussian convoluted) Fermi function to data in file

T

Temperature (K)

EF

initial value of Fermi Energy (eV)

fwhm

initial vlaue for resolution (eV) (if less than zero the fwhm is not fitted and set to |fwhm|)

min max

energy range of fit (may be less than range of experimental data points)

filename

filename (col 1 is energy in eV and col 2 intensity)

The fermifunction is defined as $f(E)=b+l(E-EF)+[d+k(E-EF)]/(\exp((E-EF)/kT)+1)$ , the function is convoluted with a Gaussian function of given fwhm and the result is compared to experimental data.
output: files can be found in directory results, filename.fit is created with fitted function and parameter values

formfactor *.sipf

program to calculate the neutron magnetic formfactor from the formfactor coefficients in the single ion parameter file *.sipf. If a radial wave function is given in *.sipf then the formfactorand the expectation values of the spherical Bessel functions are evaluated by integration with this radial wave function (see appendix J).

jjj2j:

transforms file of mcphas.jjj format to mcphas.j format - output is written to stdout

makenn 23.3:

Program to calculate neighbors of atoms given a crystal structure. Note that in order to use makenn you have to set up a working mcphas.j file with the crystal structure. The program makenn takes mcphas.j and creates all neighbours within a sphere of distance 23.3Å, for every neighbour the classical dipole interaction is calculated and is stored in file makenn.j. If the exchange parameters (and neighbour positions) are not known for your system, you can use this module to generate a list of nearest neighbours and exchange parameters. Currently implemented are not only dipolar interactions, but also exchange interactions via the Bethe-Slater curve or the RKKY model.

option -rkky A(meV) kf(1/A)

calculates the rkky interaction according to

option -rkky3d A(meV) ka(1/A) kb(1/A) kc(1/A)

calculates the rkky interaction according to $J(R)=A cos(2 \kappa)/(2 \kappa)^3$ with $\kappa^2=k_a^2 R_a^2 + k_b^2 R_b^2 + % k_c^2 R_c^2$

option -rkkz A(meV) kf(1/A)

calculates the rkky interaction according to

option -rkkz3d A(meV) ka(1/A) kb(1/A) kc(1/A)

calculates the rkky interaction according to $J(R)=A (sin(2 \kappa)- 2 \kappa cos(2 \kappa))/(2 \kappa)^4$ with $\kappa^2=k_a^2 % R_a^2 + k_b^2 R_b^2 + k_c^2 R_c^2$

option -kaneyoshi A(meV) D(A) alpha

calculates the Kaneyoshi parametrisation for the Bethe-Slater curve: $J(R)= A [-(R/D)^2+(R/D)^4]exp[-\alpha (R/D)^2]$ with corresponding to the orbital radius

option -kaneyoshi3d A(meV) Da(A) Db(A) Dc(A) alpha

calculates the Kaneyoshi parametrisation for the Bethe-Slater curve: $J(R)= A [-\rho^2+\rho^4]exp[-\alpha \rho^2]$ with $\rho^2=R_a^2/D_a^2+R_b^2/D_b^2+R_c^2/D_c^2$

option -d

puts to the last column the distance of the neighbours (A)

The neigbours of each atom are also stored in seperate files resultsmakenn.a*.pc, which can be used with the program pointc to evaluate the pointcharge model and calculate crystal field paramaters.

pointc Ce3+ 0.2 4 1 5.3

calculates Crystal field Parameters from Point Charges ... meaning calculate Stevens Parameters Blms and Wybourne Parameters Llms for one point charge of +0.2 $\vert e\vert$ in distance x=4 Åy=1 Åz=5.3 Åfrom a Ce $^{3+}$ ion. Alternative Usage: pointc Ce3+ filename ... meaning read several charges+coordinates from file, file format: column 1=charge, column 2-4 = x y z coordinate (note, (makenn creates useful files for this option from the crystal structure). results are written to stdout (including radial matrix elements and Stevens factors)
Note: if an ion is not implemented, it's parameters can be entered in a single ion property file and pointc is started as pointc file.sipf 0.2 4 1 5.3
the single ion property file must then contain the following information (# denotes comments):
#the name of the ion IONTYPE=Ce3+ #stevens parameters (optional, necessary for output of Blm) ALPHA=-0.0571429 BETA=0.00634921 GAMMA=0 # the radial matrix elements RN=<r^N> in units of a0^N (a0=0.5292 A) R2=1.309 R4=3.964 R6=23.31 # alternatively the radial wave function can be given: # radial wave function parameters R_Np,XIp(r)= r^(Np-1) . exp(-xi r) . (2 XIp)^(Np+0.5) / %%@ sqrt(2Np!) # values tabulated in clementi & roetti Atomic data and nuclear data tables 14 (1974) %%@ 177-478 # Co2+ is isoelectronic to Fe+, looking at page 422 of Clemente & Roetti # the 3D radial wave function is expanded as R(r)=sum_p C_p R_Np,XIp(r) N1=3 XI1=4.95296 C1=0.36301 N2=3 XI2=12.2963 C2=0.02707 N3=3 XI3=7.03565 C3=0.14777 N4=3 XI4=2.74850 C4=0.49771 N5=3 XI5=1.69027 C5=0.11388 # if the above parameters are given the radial wave function is output to file %%@ radwavfun.dat

radwavfunc file.sipf:

program to evaluate the radial wave function given by th parametrisation in file.sipf.

powdercell2j file:

used to create mcphas.j type file from output of powdercell,output is written to mcphas.j. Example of input file:
No name crystal coordinates cartesian coordinates x y z x y z ------------------------------------------------------------------ 1 Sr1 0.3644 0.0000 0.2500 1.0962 -4.1497 -2.7991 ...

rotateBlm

Rotates a set of crystal field parameters for Stevens equivalent operators by an azimuthal angle fi about the original z axis and a polar angle theta about the new y axis. A right hand axis system is assumed and a positive rotation is one which advances a right-hand screw in a positive direction along the axis. The calculations are done by means of matrix multiplication based on the method of Buckmaster (phys. stat. sol. a, vol 13, pp 9, 1972) and Rudowicz (J. Phys: Solid State Phys., vol 18, pp 1415, 1985). usage: $0 [-h] [--help] [-i input_file] [--input input_file] [-o output_file] [--output output_file] [-v] [--verbose] [-th theta] [-fi fi] [CF parameters] -h : this (help) message -i in_file : input CF parameters file in cfield or mcphase formats -o out_file : output CF parameters file in mcphase format -v : verbose mode. Will print out input parameters as read. -th : polar angle theta in degrees -fi : azimuthal angle fi in degrees if -i is omitted, the program will assume the input CF parameters are given on the command line in the format: Bkq=x.xx,Bkq=x.xx, etc. e.g. $0 B20=0.21,B40=0.0005,B60=0.051,B66=0.626 negative q parameters such as B_2^{-2}, are specified as: B22S, with an 'S' at the end, as per the McPhase convention. you may also specify the ion type by a dding another parameter after the CF parameters: e.g. $0 B20=0.21,B40=0.5 Pr3+ if -o is omitted, the program prints the parameters to standard output.

setup_jqfit[-h] [-help] h k l:

program to setup a fit of exchange parameters in order to reproduce an experimental propagation vector.
-h : print help message hkl : Miller indices of propagation vector required input files: mcphas.j (+ single ion paramter files) : structural information including all magnetic atoms output files: mcdisp.par : contains propagation vector and list of other hkl to be probed mcdisp.mf : required input file for mcdisp calcsta : required input file for simannfit and searchspace calcsta.pl.forfit: file with fitparameters for Bethe slater, RKKY fits fit.bat : batch to start the fit

After running this program you can start immediately a fit of exchange parameters. Edit calcsta.pl.forfit and fit.bat to fine tune the fit according to your needs. During fitting a value of sta indicates, that the maximum of is at the propagation vector tau. How much it is below one depends on the magnitude of for the competing wavevectors in the list inmcdisp.par.

setup_mcdiff_in T Ha Hb Hc:

program to setup mcdiff.in with information on spinconfiguration to be used by program mcdiff. Note, you must have done a mcphas calculation to stabilise a magnetic structure at the desired Temperature/Field. setup_mcdiff_in reads the results of this calculation from results/mcphas.mf and generates an input file mcdiff.in

-h : this (help) message T : Temperature (K) Ha,Hb,Hc : Magnetic Field (T) required input files: results/mcphas.sps : result of a mcphas calculation output files: mcdiff.in : required input file for mcdiff - after running this program you can start mcdiff to do the calculation magnetic diffraction pattern

setup_mcdisp_mf T Ha Hb Hc:

program to setupmcdisp.mf with information on meanfields to be used by program mcdisp. Note, you must have done a mcphas calculation to stabilise a magnetic structure at the desired Temperature/Field. setup_mcdisp_mf reads the results of this calculation from results/mcphas.mf and puts the meanfields into mcdisp.mf.

-h : this (help) message T : Temperature (K) Ha,Hb,Hc : Magnetic Field (T) required input files: results/mcphas.mf : result of a mcphas calculation output files: mcdisp.mf : required input file for mcdisp - after running this program you can start mcdisp to do the calculation of dispersion of excitations or diffuse scattering

setup_mcphasjforfit [-h]:

program to setup a fit of exchange parameters by creating mcphas.j.forfit from mcphas.j

-h : print help message required input files: mcphas.j (+ single ion parameter files) : structural information including all magnetic atoms output files: mcphas.j.forfit : all interaction parameters are substituted with parJxxx[0.0,-1e0,1e0,0,1e-6] - after running this program you must setup a file calcsta to calculate the standard deviation and then you can start a fit by simannfit or searchspace

singleion [option] T[K] Hexta[T] Hextb[T] Hextc[T] Hxc1 Hxc2 Hxc3 ... Hxcnofcomponents

single ion - display single ion expectations values ... and transition energies.
Hext ..... external field in Tesla Hxc... exchange (molecular) field in meV
singleion reads mcphas.j and the singleion parameter files quoted therein and calculatesenergies, eigenstates, expectation values for the given temperature, external magnetic field Hext and exchange field Hxc (the interaction constants given in mcphas.j are ignored).
for each single ion property file the following files are generated:
results/file.sipf.levels.cef .. energy levels and eigenstates and <I> results/file.sipf.trs ......... transition energies,matrix elements and (powder) neutron intensities results/_file.sipf ......... parameters as read by singleion options: -nt ......... by default only 5 transition energies are output, if you want more, start e.g. with option -nt 7 to output 7 transition energies -pinit 0.1 .. consider only transitions with population of initial state > 0.1 -ninit 3 ... consider only transitions from the 3 lowest eigenstates -maxE 30 ... consider only transitions with energy lower than 30 meV -r ion.sipf . do not read mcphas-j but only the single ion parameter file ion.sipf -M ......... calculate expectation values and transition matrix elements for magnetic moment M instead of I -S ......... calculate expectation values and transition matrix elements for spin S -L ......... calculate expectation values and transition matrix elements for orbital momentum L note: for calculating T or H dependencies you can put single ion in a loop and pipe the result into a file .... linux: for B in $(seq 0 0.1 14); do singleion 2 $B 0 0 0 0 0; done > results/fielddep.dat .... windows command line: for /L %B in (0,1,14) do singleion 2 %B 0 0 0 0 0 >> results\fielddep.dat .... windows batch file (needed for noninteger numbers): @echo off && setlocal ENABLEDELAYEDEXPANSION for /L %%I in (0,2,140) do ( set /A W=%%I/10 && set /A "f = %%I %% 10" set B=!w!.!f! @echo on && singleion 2 0 0 !B! 0 0 0 && @echo off ) endlocal && @echo on

symhmltn:

Calculates via group theory, the allowed two-ion multipolar exchange terms between two ions whose center has some particular point symmetry. Syntax is: "symhmltn $\langle$ PTGP $\rangle$ $\langle$ l $\rangle$ " where $\langle$ PTGP $\rangle$ is the name of the point group (both Hermman-Maguin and Schoenflies notation are understood), and $\langle$ l $\rangle$ is the integer multipolar order (l=1 for dipolar exchange, l=2 for quadrupolar, etc.). E.g. "symhmltn Oh 1" gives the familiar Heisenberg exchange terms. See also: cfsplit which calculates the multiplicities of the crystal field split levels of a particular point group.

Next: Programs for Manipulation of Up: McPhase USERS MANUAL Previous: Fitting is an art: Contents Index

martin rotter 2013-09-19