Fe N-Heterocyclic Carbene Complexes as Promising Photosensitizers

The photophysics and photochemistry of transition metal complexes (TMCs) has long been a hot field of interdisciplinary research. Rich metal-based redox processes, together with a high variety in electronic configurations and excited-state dynamics, have rendered TMCs excellent candidates for interconversion between light, chemical, and electrical energies in intramolecular, supramolecular, and interfacial arrangements. In specific applications such as photocatalytic organic synthesis, photoelectrochemical cells, and light-driven supramolecular motors, light absorption by a TMC-based photosensitizer and subsequent excited-state energy or electron transfer constitute essential steps. In this context, TMCs based on rare and expensive metals, such as ruthenium and iridium, are frequently employed as photosensitizers, which is obviously not ideal for large-scale implementation. In the search for abundant and environmentally benign solutions, six-coordinate Fe-II complexes ((FeL6)-L-II) have been widely considered as highly desirable alternatives. However, not much success has been achieved due to the extremely short-lived triplet metal-to-ligand charge transfer((MLCT)-M-3) excited state that is deactivated by low-lying metal-centered (MC) states on a 100 fs time scale. A fundamental strategy to design useful Fe-based photosensitizers is thus to destabilize the MC states relative to the 3MLCT state by increasing the ligand field strength, with special focus on making e(g) sigma* orbitals on the Fe center energetically less accessible. Previous efforts to directly transplant successful strategies from (RuL6)-L-II complexes unfortunately met with limited success in this regard, despite their close chemical kinship. In this Account, we summarize recent promising results from our and other groups in utilizing strongly sigma-donating N-heterocyclic carbene (NHC) ligands to make strong-field (FeL6)-L-II complexes with significantly extended 3MLCT lifetimes. Already some of the first homoleptic bis(tridentate) complexes incorporating (CNHCNpyridineCNHC)-N-boolean AND-C-boolean AND)-type ligands gratifyingly resulted in extension of the (MLCT)-M-3 lifetime by more than 2 orders of magnitude compared to the parental Fe(tpy)(2) (tpy = 2,2':6',2 ''-terpyridine) complex. Quantum chemical (QC) studies also revealed that the (MC)-M-3 instead of the (MC)-M-5 state likely dictates the deactivation of the (MLCT)-M-3 state, a behavior distinct from traditional (FeL6)-L-II complexes but rather resembling Ru analogues. A heteroleptic Fe-II NHC complex featuring mesoionic bis(1,2,3-triazol-S-ylidene) (btz) ligands also delivered a 100-fold elongation of the (MLCT)-M-3 lifetime relative to its parental Fe(bpy)(3) (bpy = 2,2'-bipyridine) complex. Again, a Ru-like deactivation mechanism of the (MLCT)-M-3 state was indicated by QC studies. With a COOH functionalized homoleptic complex, a record (MLCT)-M-3 lifetime of 37 ps was recently observed on an Al2O3 nanofilm. As a proof of concept, it was further demonstrated that the significant improvement in the (MLCT)-M-3 lifetime indeed benefits efficient light harvesting with Fe-II NHC complexes. For the first time, close-to-unity electron injection from the lowest-energy (MLCT)-M-3 state to a TiO2 nanofilm was achieved by a stable Fell complex. This is in complete contrast to conventional (FeL6)-L-II-derived photosensitizers that could only make use, of high-energy photons. These exciting results significantly broaden the understanding of the fundamental photophysics and photochemistry of d(6) Fe-II complexes. They also open up new possibilities to develop solar energy-converting materials based on this abundant, inexpensive, and intrinsically nontoxic element.

Activation of small molecules promoted by multimetallic complexes supported by redox active ligands

Davide Toniolo

The activation of small molecule is a topic of high current interest. A variety of homogeneous end heterogeneous catalysts have been studied to be able to promote chemical transformation of industrial relevance under mild condition and therefore decrease the costs. Often the most performant catalysts are composed by rare and precious metals such as Pd, Pt, Rh, Ir, making the catalyst expensive and not suitable for large scale reactions. Lots of effort is put in researching ways to substitute precious and rare metal while keeping catalytic efficiency and selectivity. F-elements have shown to be able to perform many chemical transformations that are common to d-metals, promoting also in some cases unusual chemical reactivity. Differently from transition metals, f-elements present a unique coordina-tion chemistry, governed by electrostatic interaction with the ligand environment and steric constrain. During my PhD I focused my attention on the synthesis of highly reactive, low valent metal complex-es based of f-elements for the activation of small molecules. In particular, two main pathways have been followed to promote the multi-electrons transfer necessary for the transformation of the substrate: The first approach consists in the use of redox active ligands to support the metal centre and act as electron reservoir in order to increase the number of electrons exchanged during the reaction. This method has been used for d-, f- and d- mixed f-elements complexes, resulting to be an elegant way for the preparation of multimetallic compounds capable of promoting multi-electron transfer reaction. In the second approach, divalent classic lanthanides have been stabilized thanks to the bulky tris(tert-butoxide)siloxide ligand that resulted a to be a useful tool for the formation of highly reactive bimetal-lic complexes. The reactivity of the synthetized complexes towards small molecules such as CO2, CS2 and arenes has been tested, proving the ability of those compounds to act as multielectron reducing reagents. Moreo-ver, we proved that multidentate Schiff base ligands are capable of stabilizing low valent d-metal com-plexes allowing the formation bonding interaction between different metals with the enhancement of the magnetic properties.

EPFL2020

Enantiopure helical (supra)molecular arrays containing lanthanides

Marco Antonio Lama

The diastereoselective synthesis of chiral coordination compounds with d metals proved the effectiveness of the pinene-bipyridine approach in predetermining the chirality at the metal centre. The strong Lewis acidic character of lanthanide cations required the modification of this class of ligands in order to adapt their coordination properties to the binding affinity of Ln(III) toward hard donor atoms like oxygen. The pinene bipyridine carboxylic derivative (+)-L- (figure 1) has been designed and synthesized to form configurationally stable lanthanide complexes. This ligand proved its effectiveness as chiral building block for the synthesis of oligometallic supramolecular structures. Indeed a self-assembly process takes place with complete diastereoselectivity between the enantiomerically pure (-) or (+) L- and Ln(III) ions (Ln = La, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er) in methanol, giving rise to trinuclear enantiopure structures with general formula Ln3(L)6(µ3-OH)(H2O)32 , possessing predetermined helical chirality. The ligands in this case display a new version of supramolecular helical chirality since they are wrapping around a trimetallic core instead of one single metal centre. The same components afford a second polynuclear structure if the reaction is carried out in CH3CN. In this medium a three-dimensional tetranuclear array self-assembles (with general formula Ln4(L)9(µ3-OH)2). The nine ligands are forming three distinct helical domains, two supramolecular as in the case of the trinuclear array and one classical around a single metal centre, with predetermined configuration. The two classes of compounds have been structurally characterized in solid state (X-ray and IR) and in solution (ES-MS, NMR). Their chiroptical properties were investigated by means of Circular Dichroism and Circularly Polarized Luminescence (for the emitting compounds). Trinuclear and tetranuclear arrays are shown to interconvert in CH3CN, in a process mediated by water. This process was rationalized taking into account the nature of the Ln(III) ions, the initial concentration of the components and the addition or removal of water in solution. The availability of both the enantiomers of the ligand prompted us to study the self-recognition capabilities of the two systems. 1H-NMR and CD studies reveal an almost complete chiral recognition in the self-assembly leading to the trinuclear complex. The process leading to tetranuclear species appears instead perturbed, probably due to the higher number of fundamental components involved (4 metals and 9 ligands instead of 3 metals and 6 ligands). Figure 1: Reaction conditions leading to the synthesis of trinuclear and tetranuclear Ln(III) complexes with ligand (+)-L-. The introduction of a carboxylic moiety onto the 6' posititon of the (-)-5,6- and (-)-4,5- pinene bipyridine lead to the tridentate ligands (-)-HL1 and (-)-HL2, respectively. Mononuclear neutral complexes are obtained upon reaction with Ln(ClO4)3. Different extents of diastereoselectivity are pointed out in the two systems by X-ray analysis and CD spectroscopy. In particular the ligand (-)-HL1, in which the pinene moiety is next to the binding site, seems to be more effective in predetermining the configuration at the metal centre. A low diastereoselectivity is instead expressed in the reactions with (-)-HL2 as revealed by the X-ray structure of [Pr{(-)-L2}3] (both Δ and Λ isomers detected) and by the CD spectrum void of signal.

EPFL2006

Low-dimensional supramolecular architectures at metal surfaces

Sebastian Stepanow

This thesis is concerned with supramolecular architectures assembled at metal surfaces. The investigations pursued two objectives. On the one hand, the focus was placed on the fabrication of low-dimensional surface-supported network structures by applying the concepts of conventional three-dimensional supramolecular chemistry at surfaces. Three main driving forces were explored for the construction of low-dimensional assemblies: hydrogen bonding, metal coordination and ionic interactions. The realized structures were characterized by scanning tunneling microscopy (STM) under ultra-high vacuum conditions. In particular, the interplay between the adsorbate-substrate coupling and lateral adsorbate interactions was addressed in the experiments. Further, the electronic and magnetic properties of the fabricated coordination structures were investigated. This characterization was done by x-ray absorption experiments at the synchrotron facility in Grenoble. Moreover, the response of the metal units and open cavity structures to the adsorption of large organic and small gas molecules was studied by temperature controlled STM experiments. The first part of the thesis deals with hydrogen bonded supramolecular assemblies. Simple aromatic carboxylic acids were employed to study the assembly principles and in particular effects arising from the substrate and molecular structure. The investigation of TPA (1,4-benzenedicarboxylic acid) adlayers on Cu(100) at different substrate temperatures revealed how the protonation status of the carboxyl moieties affects the topology of the assemblies. Further, the influence of the molecular backbone functionality and symmetry is addressed in comparative investigations on TPA and PDA (2,5-pyrazinedicarboxylic acid) as well as BDA (4,4'-biphenyldicarboxylic acid) and SDA (4',4" trans-ethene-l,2-diyl-bisbenzoic acid) adlayers on Cu(100). In the second part of the thesis, metal-ligand interactions were employed for the construction of supramolecular architectures at a Cu(100) surface. The concepts of conventional coordination chemistry were successfully applied at the surface. Iron atoms in conjunction with rod-like aromatic molecular linkers comprising carboxyl, pyridyl and hydroxyl functional groups assemble into a rich variety of open network structures with distinctly arranged mono- and diiron coordination centers and well-defined cavities. The results show that the adsorbate-substrate coupling considerably affects the local coordination environment of the iron units when the molecular backbone length is varied. The occurrence of structural isomerism in open network structures is addressed in a comparative study of heterofunctional and homofunctional linker molecules. It is shown that the former ligand allows only one topologically pure structure while the other linker forms two phases with different topology. The hydroxyl ligand has been used to construct distorted hexagonal coordination networks containing threefold coordinated single iron atoms. The comparison between the networks on Cu(100) and Ag(111) surfaces shows marked differences, which are attributed to adsorbate-substrate interactions. Finally, a novel design strategy for the fabrication of two-dimensional coordination structures at surfaces is introduced by using alkali metal ions in conjunction with carboxylate ligands. The interaction between the components is dominated by electrostatic forces and is of intermediate bond strength as in the case of hydrogen bonding and transition metal coordination. It is expected, that substrate symmetry and registry plays a larger role in the network formation, because the ionic interaction is not directional. The chemical as well as magnetic properties of the fabricated coordination networks are examined in the last part of the thesis. The adsorption of C60 molecules on open network structures reveals how cavity size and chemical functionality affect the bond strength to the guest molecules as well as the number of accommodated C60 molecules. Since transition metal clusters, and in particular diiron units, are the catalytic active centers in various proteins, the reactivity of various mono- and diiron coordination structures to molecular oxygen was investigated in temperature controlled STM experiments. In particular, the focus was placed on the structural relaxation upon gas adsorption and chemical reaction. At last, the magnetic properties of the metal coordination centers are addressed in XMCD experiments. The iron units are found to be paramagnetic at temperatures down to 5 K and show distinct magnetic anisotropy and magnetization loops. These experiments demonstrate that surface-supported metal coordination structures are a promising class of materials for applications in the field of heterogenous catalysis and surface magnetism.

EPFL2005