Using theoretical chemistry to understand the properties of Polyoxometalates and their potential as energy storage materials

Abstract

Polyoxometalates (POMs) are very appealing compounds as these transition metal oxide nanoclusters exhibit the ability to store multiple electrons in a reversible manner. Recently, POM-based energy storage devices, like redox flow batteries (RFBs) and alkali-ion batteries, have been extensively employed, since they meet equally the need of higher energy demand and low impact on the environment. However, POM-based technologies are still at early stage of development, mostly because of the difficulties to understand POM electronic behaviour. In this PhD thesis, it has been presented a theoretical study of POMs in different environments in order to understand the basic mechanisms behind their behaviour. In the first part, a general overview of electronic structures of POMs is given. Furthermore, the advantages and limits of POM-based batteries are discussed in detail. In Chapter 2 the state-of-the-art theoretical approaches used to study POMs are discussed, in particular, strong emphasis is given to the density functional theory (DFT), and both classical and quantum molecular dynamics (MD). In the third Chapter, the redox properties of POMs are investigated by means of simulations in an implicit and explicit environment. The implicit solvation is semiquantitative, and uncertainties arise due to the limit of the model. Quantum MD simulations reveal that an explicit environment can improve the calculated redox potentials of POMs, providing useful insights into their molecular nature. A spectroscopic study of x-ray absorption near-edge spectra (XANES) and extended x-ray absorption fine structure (EXAFS) is presented in Chapter 4. These techniques alongside with first-principles calculations have shown to be powerful tools to unveil the structure-property relationships of super reduced POMs. Chapter 5 is devoted to the study of self-assembly process of POMs. Classical MD simulations show that a rich network of hydrogen bonds mediates the POM-POM interaction, and their agglomeration strongly depends on total charge. Furthermore, first-principles calculations illustrate the effect of POM agglomeration on their redox potential, and catalytic efficiency towards the hydrogen evolution reaction. The results from this study show that it is now possible to adopt a range of computational approaches to understand the properties of POMs in different physical contexts. Specifically, the advantages and limits of DFT have been highlighted when computing the redox potentials of POMs, showing that further accuracy and insights into their electronic structure can be achieved by explicitly including the solvent molecules. For instance, it has been shown that the POM ability to undergo multiple redox reductions is due to possibility of delocalizing further electron density over all metallic atoms, regardless of the POM type. This behaviour is linked to their molecular structure, which undergoes an elongation of metal-oxygens bond lengths and formation of metal-metal bonds when further electrons are added to POMs. Furthermore, the inclusion of an explicit environment was shown to be an important factor to understand other properties of POMs, like the profile of their x-ray spectra or the self-assembly process. In the first case, QM/MM calculations shows that the polariazation of POM electron density returns more realistic molecular structures with respect to static DFT calculations, thus influencing the sensitivity of their simulated x-ray spectra. On the other hand, MD simulations revealed that the dynamical behaviour of POMs and the formation of long-lived agglomerates depends on several factors, like the total charge of POM, its counter ions, and solvent.