Rational description and modelling of the separation of nanotubes from solid nanoparticles in centrifugation processes

Abstract Centrifugation of liquid dispersions is a typical technique being employed to separate single-walled carbon nanotubes (SWCNTs) from big aggregates and impurities. Shining light on the underlying mechanism contributes to a more efficient preparation of corresponding suspensions or inks with improved properties. In this work, we model the sedimentation process by applying fundamental equations and a mathematical algorithm for the centrifugation routine, which takes into account centrifugation parameters and the geometry of nanoparticles. The model is fed with experimental data, namely, the centrifugation mass yield, the SWCNT purity index, metal contents, and nanoparticle sizes measured for pristine, thermally-oxidized, and chemically functionalized SWCNT samples. Particle aggregation, including SWCNT bundling, and changes due to physicochemical pretreatments are taken into account. It is shown that the complete experimental data set can be quantitatively well understood in terms of the particle sedimentation coefficient. The proposed model thus provides a rational concept to optimize centrifugation protocols for the separation of mixtures containing nanoparticles of different shapes.