Comparison of the linear Van den Berg/Ru i transformation and a non-linear fit of the Langmuir isotherm applied to Cu speciation data in the estuarine environment

Copper complexation measurements in estuarine organic-rich samples with differential pulse anodic stripping voltammetry (DPASV) and differential pulse cathodic stripping voltammetry (DPCSV) show a relatively large scatter in the data, due to a high organic matter content. Two methods for the estimation of the complexation parameters according to the Langmuir isotherm are compared: the linear transformation of Van den Berg/R zíc and the non-linear method according to Wilkinson. For DPASV measurements in estuarine samples the values of the conditional stability constant (K′), estimated with the linearization method are systematically lower than those estimated with the Wilkinson method. The Van den Berg/Ru i linearization is more susceptible to outliers or data points that do not follow the model of the Langmuir isotherm. For DPCSV measurements, where the data followed the Langmuir isotherm more closely, this difference could not be demonstrated. The non-linear method is to be preferred because this method is more suited to the error structure of the data, which is of constant absolute magnitude. Moreover it offers the advantage that the standard error of the estimated parameters can be calculated consistently.