The chemistry of divalent lanthanides has generated increasing interest in the past years due to their redox properties and unique reaction pathways. Notably, molecular complexes of low-valent lanthanides have been shown to be suitable one-electron reductants that are able to activate inert small molecules such as heteroallenes and dinitrogen in mild conditions. Lanthanide(II) complexes usually undergo mono-electron transfer, but multi-electron transfer can be achieved through multiple one-electron processes in polymetallic complexes. The first part of this thesis describes attempts to expand the redox chemistry of lanthanide complexes and lays the basis for a wider understanding of the chemical and reductive properties of low-valent polynuclear lanthanide complexes. Thus, Chapter 2, 3, and 4, reports the synthesis and characterization of different polynuclear lanthanides(II) complexes of various siloxide ligands. In Chapter 2, the outcome of carbon dioxide reduction (oxalate versus carbonate) by dinuclear Ln(II) tris(tert-butoxy)siloxide complexes, depending on the solvent polarity and on the nature of the Ln ions, was investigated. Furthermore, unprecedented reduction products could be isolated from the reaction with CS2, providing the first example of acetylenedithiolate ligand formation from metal reduction of carbon disulphide. In all cases, it has been shown that cooperative binding of the substrate plays an important role in the outcome of the reaction. Chapter 3 describes a rational route to the first metallasilsesquioxane of a dinuclear divalent lanthanide which can selectively effect the reductive disproportionation of CO2 and the two-electron reduction of azobenzene. In an attempt to confer increased reductive reactivity to Eu(II) ions, Chapter 4 describes the use of the tetraphenyldisiloxanediolate ligand to form polynuclear complexes of Eu(II) that can initiate for the first time carbon dioxide activation by Eu(II) ions in an isolated complex. Differently, until 2019, the +IV oxidation state of lanthanide in molecular complexes remained limited to a single lanthanide ion, cerium. It is only recently that the terbium ion was shown to be able to access the +IV oxidation state in molecular complexes. However, given the broad applications of cerium(IV), the discovery of other molecular complexes of Ln(IV) is of great interest for chemists. Chapter 5 describes the synthesis of the third example of a Tb(IV) complex displaying an open coordination sphere and the very first report of a molecular complex of praseodymium in the +IV oxidation state. The novel Ln(IV) complexes were characterized by UV-Visible, NMR, EPR and IR spectroscopy, magnetometry, single-crystal X-ray diffraction analysis, cyclic voltammetry, and computational studies. In particular, this provided an opportunity to measure and study the luminescent spectra of Pr(IV) for the first time, which upon binding to the siloxide ligand results in a large shift of the ligand triplet state. Furthermore, these complexes have been shown to be appropriate starting materials for the synthesis of Ln(IV) phosphine oxide adducts. Preliminary studies directed towards the isolation of Nd(IV) molecular complexes are discussed.