PF Chrom v5 Tutorials

Chromatography Peak Modeling

In the first tutorial, four IC data sets, from the same standard of six cation components, are fitted. One is from the mid 1990s and the other three are from significantly different times on a modern instrument.

The basics are covered: the retention scale and the dead-time transforms, sectioning, baseline correction, raw data visualization, signal/noise on samples, and fitting a generalized chromatographic model with an instrument response function (IRF). There is also a deconvolution preprocessing and subsequent fitting of a closed form model. The tutorial addresses the advantages and disadvantages of fitting an IRF convolution model vs. subtracting an IRF prior to fitting a closed form model.

Instrument Response Function Deconvolution and Fitting

In this third tutorial, the IRF instrument response function is deconvolved using the averaged estimates from convolution fits and the data are subsequently fitted with a closed form model. This highlights the procedure that may be the main one used for managing routine analyses on a given instrument and column. The tutorial also includes the estimation of the IRF directly using a genetic algorithm and highlights the advantages and disadvantages between fitting an IRF-bearing convolution model versus first deconvolving a known IRF and fitting a closed form model.

HPLC Gradient Peaks – Fits Which Model the Gradient

The second HPLC tutorial introduces a new fitting technology, deconvolution fitting, which models the gradient’s presence within a peak. Two different deconvolution models are used in this tutorial to produce exceptional fits which internally unwind the gradient to generate chromatographic parameter estimates which represent the peaks that would have been seen in an isocratic elution. This technology allows the gradient to be estimated on a peak by peak basis and the Explore option is used to graphically view parameter relationships across the different gradient peaks fitted.

Chromatographic Experiments

In this second tutorial, sixteen different IC data sets with different concentrations and additive levels are fitted in full experiment to study the effects of these two variables. The tutorial extensively covers raw data visualization and the Explore option for the visualization of fitted relationships of parameters, moments, and peak properties.

In this tutorial, you will see in minutes what took us a month or more to accomplish with existing software at the time PF Chrom v5 was being developed.


HPLC Gradient Peaks – Direct Closed Form Fits

In this first of the HPLC gradient peak tutorials, the IRF is determined for an isocratic peak prior to the gradient and the genetic algorithm is then used with the HPLC gradient peaks to see how much of this IRF survives the gradient. For these gradient peaks, a twice-generalized closed form model is used in a single step to fit the gradient peaks, including an estimate of the fourth moment compression occurring during the elution. All of the options for directly fitting the gradient shape are covered.

HPLC Gradient Peaks – Fitting Unwound Data (Tutorial)

In this third HPLC gradient tutorial, fits are made of standards and production data where the gradient which has been modeled is unwound in a convolution step and then fitted. The standards are fitted to an astonishing 1 ppm unaccounted variance error. This tutorial uses production data to compare the three different approaches for fitting gradient peaks and highlights the advantages and drawbacks of each method.

Fitting Hidden Peaks

PF Chrom’s algorithms have a thirty year history of indentifying and esitmating hidden peaks within data. With the new segment fit option, peaks are separated into baseline resolved segments and all peaks within a segment, including hidden peaks, can share parameters. This tutorial will highlight the use of the Residuals and Second Derivative method for automatically finding hidden peaks with the use of this new segment option. The tutorial will also discuss graphical and numerical adjustments of peaks as well as setting common estimates across peaks.

Fitting Preparative (Overload) Peaks

In this tutorial a progression of increasingly higher concentration preparative peaks are fit to overload models. The deconvolution estimates the chromatographic peak that would be seen if the column was capable of fully managing the concentration without overload. The tutorial covers the many ways to parametrically estimate the magnitude of overload.

Two-State Experimental Models

In this tutorial two-state zero distortion density models are evaluated. These are models which assume two different zero-distortion densities occur in the separation.

HPLC Column Health and Overload

In this tutorial, peak data from three columns in different states of health are evaluated. A principal peak, in different states of overload, are evaluated using one of PF Chrom’s overload models. The tutorial is designed to highlight the visualization, fitting, and deconvolution of the peaks as a function of column health.


UDF Fitting – Experimental ZDDs,IRFs

PF Chrom introduces instrument response function (IRF) and higher moment zero distortion density (ZDD) fitting of chromratographic peaks. This tutorial illustrates user function models for exploring different ZDDs and using integrals to fit experimental IRFs. In this tutorial, the main generalized HVL models are reproduced in user functions.

Chromatography Peak Modeling
Chromatographic Experiments
IRF Deconvolution and Fitting
HPLC Gradient Peaks – Direct Closed Form Fits
HPLC Gradient Peaks – Fits Which Model the Gradient
HPLC Gradient Peaks – Fitting Unwound Data
Fitting Hidden Peaks
HPLC Column Health and Overload
Fitting Preparative (Overload) Peaks
User Function Fitting – Experimental ZDDs,IRFs
Two-State Experimental Models