Comment on "Representation and Prediction of Molecular Diffusivity of Nonelectrolyte Organic Compounds in Water at Infinite Dilution Using the Artificial Neural Network-Group Contribution Method [Gharagheizi et al., J. Chem. Eng. Data 2011, 56, 1741-1750]

In their article, Gharagheizi et al. [J. Chem. Eng. Data 2011, 56, 1741-1750] claim to develop a model to predict the molecular diffusivity of "nonelectrolyte organic compounds" in water using an artificial neural network-group contribution method. Many of these compounds have ionizable functionalities with pKa values that result in either substantial or effectively complete ionization in water (making them electrolytes, in contrast to the claims of Gharagheizi et al.). Consequently, the model developed and applied by Gharagheizi et al. does not computationally model the actual speciation(s) of each compound expected to be present under the experimental conditions for which the underlying data has been obtained and erroneously classifies many organic compounds as non-electrolytes