Thorben Petersen

Digitalization and increasing the flexibility of production concepts offer the possibility to react to market challenges in the field of specialty chemicals. Shorter product lifetimes, increasing product individualization, and the resulting market volatility impose new requirements on plant operators. Novel concepts such as modular production plants and developments in digitalization (Industry 4.0) are able to assist the implementation of smart factories in specialty chemicals. These essential concepts will be presented in this Minireview.

Products on demand: Modular, intelligent production concepts provide suitable solutions for the challenges that plant operators are facing in dynamic markets that have increasing product individualization. Concepts such as batch-to-conti and the integration of fast configurable modules have already proven their potential. Thus, the way towards a smart factory has already been paved.

The history of the two main theories of chemical bonding, Valence Bond and Molecular Orbital theories, with their respective localized and delocalized visions of the electron pairs, is briefly recalled with an emphasis on their equivalence at comparable computational levels. The delocalized/localized alternative also exists in the very framework of MO theory, since localized MOs can be obtained through unitary transformations which leave Slater determinants invariant. In particular, the use of hybrid orbitals in chemical interpretations is thus perfectly justified. It is argued that a dual knowledge of both theories and their interconnections is of great help to get a complete understanding of the nature of bonding in molecules. Other means of switching from the delocalized to localized representations are also shown to be quite insightful in, e.g., hydrogen-bonding around water molecules, or conformational analysis of propene and many other unsaturated acyclic systems. It is shown that playing with such equivalent readings of the same wave function of dicarbon provides a very clear bonding picture for this enigmatic molecule, whereas the original untransformed wave function is ambiguous. The localized orbital representations of water, ethylene and dicarbon are shown to basically provide the same picture as the analysis of electron density maps as provided by the Maximum Probability Domains method and by the Dynamic Voronoi Metropolis Sampling method. Some current misconceptions are falsified, like the supposed existence of some unique "real" set of orbitals for a molecule, or the belief that the localized lone pair orbitals of water, the famous "rabbit-ear" hybrid orbitals, are in conflict with photoelectron spectroscopy.

“Where do we go from here?” is the underlying question regarding the future (perhaps foreseeable) developments in computational chemistry. Although this young discipline has already permeated practically all of chemistry, it is likely to become even more powerful with the rapid development of computational hard- and software.