A Multivariate Correlation Analysis Method Identifying Key Synthetic Parameters of Nickel Nanocatalysts Active in H ₂ Production From Ammonia‐Borane Solvolysis

The control of the properties of functional nanoparticles (NPs) requires rational adjustment of the parameters in their synthesis. However, the systematic study of all the experimental parameters for a given protocol generates data that can be intricate to handle and interpret. We address herein the development of a methodological approach to optimizing synthesis/structure of functional nanoobjects. Principal component analysis and correlation matrix evidence the prevalent synthetic parameters and their possible correlation. We develop 2D-size plots visual graphic tool from statistical treatment of NPs by microscopy. We applied these tools in a textbook case dedicated to the synthesis of colloidal Ni nanocatalysts used for H₂ delivery from the solvolysis of ammonia-borane (AB) under mild conditions. As a result of the multivariate correlation analysis approach, we established the key role of the precursor nature in the synthesis, that mainly governed the size, dispersity and crystallinity of the formed NPs. At different examination scale, the clustering analysis of subpopulation of formed NPs or the electronic topology analysis of the molecular precursors were achieved. The difference in nanocatalysts performances is correlated to differences in crystallinity, which in turn comes from the nature of the precursors used. The methodological approach proposed here, focused at colloidal Ni NPs used for H₂ delivery, reinforces the role of data science for the development and optimization of nanomaterials.