Figure 2
A two-part algorithm for the estimation of the composition and concentration of an unknown mixture of \(N\) analytes with known thermodynamic parameters. The unknown mixture is injected at temperature \(T\) for which corresponding kinetic parameters may be calculated using the Eyring equation. The first part of the algorithm surveys the dissociation phase of the sensorgrams to provide a first estimate of the fractions \({F}_{i}\) of each analyte. The starting point for this optimization routine is set to \(1/N\). A first estimate of the total concentration \({C}_{TOT}\) can then be found via linear regression. The second part of the algorithm considers the whole sensorgrams to provide a final estimate of \({F}_{i}\) and \({C}_{TOT}\). Note that, if sensorgrams of more than one total concentration are available, the corresponding dilution rates (\(D\left(s\right)\)) must be known.
