General description
In studying any kind of spectrum -- energy, mass, time-of-flight etc. -- of particles, the first task is usually to obtain a reliable quantitative description of the observed features. Nonlinear least-squares optimization methods are commonly used for this purpose. In this approach, called "curve fitting", one or several analytical shapes are used to model the observed spectrum and their parameters are adjusted iteratively by the least-squares optimization procedure, until the best possible match to the spectrum is obtained. In photoabsorption and electron spectra, for example, the peaks are often described by Voigt profiles, and peak positions, intensities and widths are obtained as a result of curve fitting.
There is a number of scientific data analysis programs that include curve-fitting capabilities, from very simple single gaussian fits to more complex linked-parameter multipeak routines. However, there are always restrictions in what the software can do and often some critical aspect is not implemented.
This curve-fitting package has been designed to provide maximum flexibility and expandability, while still keeping the user interface simple and manageable. On the oter hand, the user has a variety of ways to impose restrictions on the input parameters. Particularly important is the implemented powerful parameter linking arithmetics, since smart linking of the fit parameters is often the key to accurate and meaningful result. These macros have been developed for analyzing electron energy and photoabsorption spectra, but its features can be useful also for other spectroscopic applications.
This macro package is developed for Igor Pro -- a scientific data analysis environment for both PC and Mac. To use this package, Igor Pro must be installed and functional. The package itself is compatible with both PC's and Mac's.
Two optimization algorithms - Levenberg-Marquardt and Simplex - are available.
Several pre-defined lineshapes together with easily modifiable user-defined shapes that can be in either analytical or numerical (Igor wave) form.
Linear, constant, or user-defined background that can be optimized together with the peaks.
All input parameters can be either fixed or optimizable.
All parameters of a peak can be linked to the relevant parameters of other peaks. There is a number of ways to link parameters, the most complex involving the energies or intensities of three other peaks and an arbitrary constant. [Read more about linking]
Preparation of the initial fitting parameters can be done either directly in the parameter table or using a graphical mouse-driven interface. Both methods can be used simultaneously.
Fitting of several spectra simultaneously is supported.
A new table is generated for the fitting results.
Fit parameters can be saved to file and retrieved, making it easy to fit a number of similar spectra.
Labeling and automated tagging of peaks, making the result easy to read and follow.
Statistical error analysis based on analytical and Monte-Carlo methods.
New! Modelling of vibrational progressions by optimising the parameters of the Morse potential energy curves.
To find out more, download the SPANCF manual that is part of the package (requires Igor Pro to open).
In order to obtain a copy of these macros, contact Edwin Kukk: edwin_dot_kukk_at_utu_dot_fi (replace _dot_ and _at_ with the respective symbols)
