In order to determine the ζ-potential of TiO2 aqueous suspensions three approaches were utilized. Electrophoresis, conductivity and the combination of both groups of data were used in models formulated imposing the thin double layer condition (κa>10). Some discrepancies among these results were found and a method to assess the accuracy of each one was elaborated. This method consisted in the evaluation of the Gibbs free PR619 increment (thermodynamics method) taking place during the charging process at the solid–liquid interface, and the study of the normalized yield stress/squared ζ-potential relationship (rheological method). A plot of the Gibbs free energy increment for pH 4 and pH 6 vs. κa showed a very similar behavior regardless of the method employed to evaluate the ζ-potential. A relationship between the yield stress and the squared ζ-potential, which is based upon the colloidal forces/microstructure connection, was also employed to solve this lack of criteria to decide which the actual ζ-potential was. A plot of the normalized yield stress vs. the squared ζ-potential provided a linear relationship, according to the theoretical predictions, whatever the ζ-potential utilized. From the slope of these plots the interparticle surface separation distance was calculated and from the intercept with the x-axis the ζcrit was evaluated. Both values were compared with those reported by other authors. The results were not conclusive but, since anomalous surface conductance can occur, a combination of data from electrophoresis and conductivity measurements would probably provide the more advisable ζ-potential value.