5.3 Molecular understanding and control of the mineralization process
While the basic processes of metal mineralization are qualitatively understood, mechanistic details regarding the reaction mechanisms of metal reduction and adsorption remain elusive. Current studies on metal mineralization using TMV, BSMV and their corresponding VLPs as biotemplates focused on characterization of synthesized nanostructures using different techniques such as transmission electron microscopy (TEM) [24], Fourier transform infrared spectroscopy (FTIR), and X-ray scattering analysis [19, 59]. These methods reveal the structure of the final synthesized structure and inform hypotheses of how metal mineralization has occurred. However, a more effective approach to understand how metal mineralization takes place is to perform in-situ FTIR [90]. This allows direct observation of the reaction progress enabling determination of the mechanism of mineralization. It is also possible to observe how changes in reaction conditions such as pH, temperature, and concentrations of precursors, reducing agents and biotemplate would affect metal mineralization including particle size, particle size homogeneity, and the type of metal nanostructures mineralized on the surface. This would require an extensive design of experiments to systematically evaluate the effect of each parameter and their interactions. Machine learning algorithms such as artificial neural networks have already been applied to these rich datasets to create predictive models for nanoparticle synthesis as a function of processing parameters [91]. Similarly, neural networks first used to predict the binding of metallic ion cofactors to enzymes could be extended to VLPs to predict and model metal-biotemplate interactions as a function of engineered CP protein sequence [92]. Such computational tools would greatly accelerate biotemplate engineering efforts and optimize deposition processes for metallic nanomaterial synthesis.