Curve Fit Command

Curve Fitting allows you to superimpose Gaussian, Lorentzian, or Asymmetric Gaussian-Lorentzian peak shapes on data from an XPS multiplex region. The parameters of these peaks are adjusted until the resulting summation of the peaks (the artificial, or synthetic curve) best approximates the experimental data. In this way, one can derive the individual peaks that make up a given data set and characterize the chemical state of the data.

Curve fitting can be performed independently for each region of a multiplex. Results can be displayed on the graph of the curve fit region and copied to the clipboard for use in other applications.

Curve fitting is inherently complex. If you are not familiar with the theoretical and applied aspects of curve fitting as it applies to XPS data, you may wish to consult the various references available.

Although you can only curve-fit a multiplex region, you can curve fit data from XPS surveys and depth profiles by converting this data to multiplexes. See the following for more information:

Creating a Multiplex From a Survey

Creating a Multiplex from a Depth Profile

General Curve Fitting Procedure

Following is a brief description of the steps involved in curve fitting:

  1. Prepare your data for curve fitting. Any data massage and transformation functions must be done prior to curve fitting.

  2. Select a baseline for the data, and subtract the background.

  3. Select the band type, default limits, and maximum number of iterations for the curve fit.

  4. Add bands (peaks) that best describe the inherent peaks whose summation will result in the closest match to the experimental data.

  5. Adjust the band parameters to obtain a close fit. In essence, you are now manually curve fitting the data and using visual cues to determine the goodness of fit.

  6. Instruct the computer to perform a curve fit, thereby optimizing the parameters of each band for the closest fit.

  7. Analyze the results and modify the parameters of the individual bands, limits, locking, and iterations, if necessary.

Preparing Data for Curve Fitting

Additional Curve Fitting Topics