Quantum chemistry: the dielectric chain model imagined by Jean-Louis Revelle and Daniel Rinaldi celebrates its 50th anniversary

• Manuel Ruiz Lopez, what was the impact of this post at the time of its release??

Let us first recall that the majority of theoretical chemists at the time were focused on studying small, isolated molecules. The available computational means did not really make it possible to approach the electronic computation of complex systems. Thus, at first, the publication of this model essentially attracted the attention of a few theorists who were interested in the effects of the solvent, which had hitherto been taken into account in a very primitive way (for example, by explicitly considering a single molecule of solvent reacting with the solute). ). But it should also be noted that the article was written in French (with an English abstract), which is common practice at the paper Theoretica Chimica Acta, since it is also possible to publish in German and even in Latin! This fact undoubtedly limited the direct impact of the 1973 publication. This would be a second article published in 1976 (this time in English in chemical physics) and describe a more detailed version of the model, which will give it the critical impetus and encourage other groups to develop similar dielectric models.

• How is the solvent introduced into the dielectric model and what is its effect on the result of the calculations?

Conceptually, the model is very simple. The solvent is represented by a continuous dielectric medium surrounding the solute. The latter is supposed to be in a “hollow” created in the centre. In the presence of the electric charges carried by the solute, the dielectric medium becomes polarized. This creates an electrostatic potential that interacts with the electrons and nuclei. The mathematical expression for voltage is obtained by solving the classical electrostatic equations. This potential is then added to the Hamiltonian of the solute for which the wave function can be computed by the usual methods of quantum chemistry. Of course, the introduction of an external potential due to the solvent causes in turn the polarization of the charges of the solute, leading to a system of nonlinear equations that must be iteratively solved (Self-consistent reaction field equations). The complexity of the mathematical mechanism of the problem depends largely on the chosen shape of the cavity. In the first model of 1973, a spherical or elliptical form was considered, leading to an analytical solution to the potential consisting of the sum of the multipolar contributions. The case for a cavity with a general shape that better fits the structure of the molecule will be developed later, using numerical calculations. The great advantage of this model is that it makes it possible to rationalize experimentally observed solvent effects from relatively simple theoretical calculations of quantum chemistry. These effects sometimes modify the structure, spectral properties and stability of solutes in a very radical way. It therefore exerts a great influence on chemical equilibria and molecular interaction.

• Has this model evolved since its birth? Is it still in use and what is its future?

It is clear that the model has evolved a lot since its inception. There are many approaches today, depending on how you solve the dielectric equations and estimate the non-electrostatic terms. It is available in most quantum chemistry programs and allows for routine studies in solution. So the model continues to be used daily by a large number of researchers. At Nancy, we have gradually moved towards more detailed solvent models, which combine the techniques of quantum chemistry and statistical mechanics. This development is necessary to describe the dissolution phenomenon from a dynamic point of view, taking into account the microscopic structure of the solvent. However, dielectric methods, despite their obvious limitations, remain very useful because the additional computational cost compared to isolated molecules is very low. They can therefore be used to easily troubleshoot issues of interest before implementing more complex techniques, if necessary. With regard to their future, a major challenge will likely be their extension to the study of unconventional media, such as solvents compatible with green chemistry (eutectic solvents, ionic liquids, supercritical liquids, etc.) or interfaces. Some efforts have already been made in this direction, but there is still a long way to go.