Difficulties and flaws in performing accurate determinations of zeta potentials of metal nanoparticles in complex solutions-Four case studies.

The zeta potential (ZP) is a parameter commonly used to characterize metal nanoparticles (NPs) in solution. Such determinations are for example performed in nanotoxicology since the ZP influences e.g. the interaction between cells and different biomolecules. Four case studies on different metal NPs (Cu and Zn NPs, and citrate capped Ag NPs) are presented in this study in order to provide guidance on how to accurately interpret and report ZP data. Solutions of high ionic strength (150 mM NaCl) induce a higher extent of particle agglomeration (elucidated with Ag NPs) when compared with conditions in 10 mM NaCl, which further complicates the prediction of the ZP due to e.g. sedimentation and broadening of the zeta potential distribution. The particle size is seldom included specifically in the standard ways of determining ZP (Hückel and Smoluchowski approximations). However corrections are possible when considering approximations of the Henry function. This was seen to improve the analysis of NPs, since there are cases when both the Hückel and the Smulochowski approximations are invalid. In biomolecule-containing cell media (BEGM), the signal from e.g. proteins may interfere with the measured ZP of the NPs. The intensity distribution of the ZP of both the blank solution and the solution containing NPs should hence be presented in addition to the mean value. Due to an increased ionic strength for dissolving of metal NPs (exemplified by Zn NPs), the released metal ions must be considered when interpreting the zeta potential measurements. In this work the effect was however negligible, as the particle size was several hundred nm, conditions that made the Smoluchowski approximation valid despite an increased ionic strength. However, at low ionic strengths (mM range) and small-sized NPs (tens of nm), the effect of released metal ions can influence the choice of model for determining the zeta potential. Sonication of particle dispersions influences not only the extent of metal release but also the outermost surface oxide composition, which often results in an increased ZP. Surface compositional changes were illustrated for sonicated and non-sonicated Cu NPs. In all, it can be concluded that accurate measurements and interpretations are possible in most cases by collecting and reporting complementary data on characteristics such as particle size, ZP distributions, blank sample information, and particle oxide composition.