Fabrication of monodisperse silica spheres using statistical design of experiments

Designing bioinspired green nanosilicas using statistical and machine learning approaches

“…A powerful advantage is the possibility to construct linear regression models to establish empirical relationships for prediction of product responses as a function of synthesis factors. 25 Full and fractional factorial designs have been successfully used previously to screen the synthesis of Stöber silica, [30][31][32] SBA-15, 33 and silica via dissolution precipitation. 34 Whereas factorial designs are regarded as resource-efficient for identification of significant synthesis factors, they often do not contain a sufficiently large number of treatments for response modelling with more precise second-order regression polynomials.…”