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] abib_v1_0.gz(14 Kbytes)
Manuscript Title: Computer analysis of Mossbauer spectra.
Authors: B.L. Chrisman, T.A. Tumolillo
Catalogue identifier: ABIB_v1_0
Distribution format: gz
Journal reference: Comput. Phys. Commun. 2(1971)322
Programming language: Fortran.
Computer: IBM 360-75.
RAM: 32K words
Word size: 32
Keywords: Nuclear physics, Least squares fitting, Lorentzian line shape, Mossbauer effect.
Classification: 17.3.

Nature of problem:
Experimental Mossbauer data are fitted to a sum of lorentzian lines. The program allows for several types of background: flat, sloping or parabolic. The parameters for the background are fitted simultaneously with the lorentzian parameters.

Solution method:
The least-squares fit is performed using a Gauss-Newton iteration procedure. Each lorentzian line is characterized by three parameters: intensity (given in counts), half width at half maximum (given in channel numbers) and a line position (given in channel number). The width and intensity parameters may be restricted to be multiples of one another. However, the line positions cannot be so restricted.

At present the program is dimensional such that the maximum number of parameters allowed is 50 and the maximum number of data points is 512. Either of these restrictions may be removed by increasing the appropriate dimensioned variables.

Unusual features:
By suitably choosing certain restrictions on the parameters a variety of physical models can be imposed on the data and the goodness of fit checked. By assigning certain values to control parameters the informational output of the program can be controlled in a convenient manner.

Running time:
The execution time is of course dependent on the number of parameters and the number of data points as well as on the initial estimates of all the parameters. Using the IBM 360-50 at Kanas State University with a G level compiler twelve lorentzian lines with a flat background (37 parameters) were fitted to 256 data points in approximately 5 min. Using the H level compiler this is reduced to 2.5 min. Using a Model 360-75 the execution time is less than 20 sec.