Elsevier Science Home
Computer Physics Communications Program Library
Full text online from Science Direct
Programs in Physics & Physical Chemistry
CPC Home

[Licence| Download | New Version Template] abqa_v1_0.gz(40 Kbytes)
Manuscript Title: Versatile program for analysis of Mossbauer spectra.
Authors: M.F. Bent, B.I. Persson, D.G. Agresti
Program title: MOSSBAUER DATA LEAST-SQUARES FIT
Catalogue identifier: ABQA_v1_0
Distribution format: gz
Journal reference: Comput. Phys. Commun. 1(1969)67
Programming language: Fortran.
Computer: IBM 360/75.
Operating system: SYSTEM 360/0S.
RAM: 40K words
Word size: 32
Peripherals: graph plotter.
Keywords: Nuclear physics, Mossbauer effect, Hyperfine interaction, Velocity spectra, Least-squares fitting, Simultaneous fitting of, Complementary spectra.
Classification: 17.3.

Nature of problem:
The program is designed to facilitate the analysis of experimental Mossbauer spectra in terms of hyperfine interactions.

Solution method:
A linearized least-squares procedure is used to adjust to the experimental data appropriate theoretical functions. An experimental spectrum is analyzed, not in terms of individual components, but rather a theoretical function is formulated using physical parameters which are adjusted in the fitting procedure. Since these parameters are usually not strongly correlated, this minimizes convergence difficulties without recourse to non-linear least-squares procedures, thereby saving computer time.

Restrictions:
The program may be used to fit data resulting from the overlap of source and absorber spectra, both of which may be split by hyperfine interactions. Spectra involving combined quadrupole and magnetic inter- action can be analyzed if the quantization axes for the two interactions are parallel. The program can be readily adapted to most experimental situations.

Unusual features:
Data from complementary experiments can be simultaneously fitted in a single least-squares procedure. When data obtained under properly chosen experimental conditions are analyzed in such a combined fit, the correlations between the parameters are reduced. This increases the likelihood that the fit will converge and improve the precision in the determination of the parameters. A wide range of physical interpretations can conveniently be tried for a given experimental spectrum, so that systematic errors can be estimated in a quantitative way.

Running time:
On the IBM 360/75 at California Institute of Technology the loading and assembly of the program require about 30 seconds. The execution time is depending in the amount of data, the complexity of the functions to be fitted to it, and the accuracy of the initial estimates of values of parameters to be determined by the fit. To fit a function composed of 10 superimposed Lorentzians to 200 data points may require about 10 seconds of execution time.