|
NETZSCH
SEPARATION OF PEAKS
is a program for evaluation of different types of data. With
it, the experimental data are fitted as an additive
superposition of peaks, where each single peak can be
represented with the following types of profiles:
GAUSSIAN,
CAUCHY,
PSEUDO-VOIGT
(weighted mixture of Gaussian and Cauchy
profile),
FRASER-SUZUKI (asymmetric Gaussian),
PEARSON -7
(monotonic transition from Gaussian to Cauchy
profile).
Fraser-Suzuki profile
with various values of asymmetry
Each single peak is characterized by three or four
parameters:
- position,
- amplitude,
- half width,
- asymmetry (for Fraser-Suzuki profile only),
- slimness (for Pearson profile only).
For correction with a Baseline, the program offers
TSCHEBYSCHEFF polynomials to a maximum of the 3rd
degree.
The model is fit to the
experimental data by means of a hybrid GAUSS-NEWTON
procedure, where a Marquardt method is combined with a step
length optimization. Two equally remarkable features of the
algorithm are speed and stability.
The solution can be limited to a reasonable range
through:
- free choice of lower and upper limits for each parameter
and
- selection of Parameter Options constant, equal and
designation.
ASCII Files serve
as a general interface for the transfer of data.
Results of the
evaluation are:
- the optimal parameters and their standard errors,
- statistical characteristic values such as the correlation
coefficient
and the average fit error,
- graphic representation of the curve fit including the
separated
single peaks,
- calculation of the areas of the single peaks and their
percentage
of the total peak.
Program application
areas are:
- thermal analysis (Differential Scanning Calorimetry
and
Thermogravimetric Analysis),
- UV/VIS spectroscopy,
- NIR and IR spectroscopy,
- NMR spectroscopy,
- X-ray small-angle scattering of polymers
(for profile analysis with a Pseudo-Voigt profile,
the
Gauss component indicates the degree of
crystallinity).
|
