Nitrogen reduction and functionalization by a multimetallic uranium nitride complex

Molecular nitrogen (N-2) is cheap and widely available, but its unreactive nature is a challenge when attempting to functionalize it under mild conditions with other widely available substrates (such as carbon monoxide, CO) to produce value-added compounds. Biological N-2 fixation can do this, but the industrial Haber–Bosch process for ammonia production operates under harsh conditions (450 degrees Celsius and 300 bar), even though both processes are thought to involve multimetallic catalytic sites (1, 2). And although molecular complexes capable of binding and even reducing N-2 under mild conditions are known, with co-operativity between metal centres considered crucial for the N-2 reduction step (1-14), the multimetallic species involved are usually not well defined, and further transformation of N-2-binding complexes to achieve N–H or N–C bond formation is rare (2, 6, 8, 10, 15, 16). Haber noted (17), before an iron-based catalyst was adopted for the industrial Haber–Bosch process, that uranium and uranium nitride materials are very effective heterogeneous catalysts for ammonia production from N-2. However, few examples of uranium complexes binding N-2 are known (18-22), and soluble uranium complexes capable of transforming N-2 into ammonia or organonitrogen compounds have not yet been identified. Here we report the four-electron reduction of N-2 under ambient conditions by a fully characterized complex with two U-iii ions and three K+ centres held together by a nitride group and a flexible metalloligand framework. The addition of H-2 and/or protons, or CO to the resulting N-2(4-)complex results in the complete cleavage of N-2 with concomitant N-2 functionalization through N–H or N–C bond-forming reactions. These observations establish that a molecular uranium complex can promote the stoichiometric transformation of N-2 into NH3 or cyanate, and that a flexible, electron-rich, multimetallic, nitride-bridged core unit is a promising starting point for the design of molecular complexes capable of cleaving and functionalizing N-2 under mild conditions.

Small Molecule Activation by Multimetallic Uranium Complexes

Marta Falcone

The development of new processes for the selective and sustainable transformation of abundant small molecules constitutes one of the major research areas in inorganic and organometallic chemistry. These molecules as CO2, CO, N2 and H2 are abundant reservoirs of chemical energy, readily available sources of carbon, nitrogen and hydrogen. In a patent from 1909, Haber reported that uranium and uranium nitride were very efficient catalysts for ammonia synthesis. In order to perform multi-electron transfer, multimetallic complexes need to be designed. Reactivity of uranium bridging nitride remained unexplored since we investigated the nitride based reactivity of a dinuclear uranium(IV) bridging nitride complex towards small molecules such as CO2, yielding the first example of a uranium complex containing a bridging dicarbamate ligand, or CO, resulting in the complete cleavage in mild conditions of one of the strongest bonds in nature, or H2, yielding one of the few examples in f-elements of metal hydride, which is readily transferred to substrates such as CO2. Beyond the ligand based reactivity, we were interested in reducing the two metal centers to access a diuranium(III) bridging nitride, to pursue nitrogen activation. This complex was shown to have high flexibility, provided by the siloxide ligands, high reducing power and a bridging atom, the nitride, that holds together the uranium centers, with the possibility to bend upon nitrogen activation. This approach led us to the isolation of a hydrazido complex, resulting from the four-electron reduction of dinitrogen by the two metal centers. Moreover, complete cleavage of the NâN single bond was achieved by addition of CO to give two cyanate ligands or by addition of H2 to give ammonia. The nitride atom linker is nonetheless, too reactive to act as a spectator and it is involved in reactivity. Hence, we target the isolation of the analogue dinuclear U(III) complex with an oxide ligand as linker. This complex showed the same reactivity towards nitrogen but DFT revealed a more ionic character of the UâO bond nature with respect to the UâN bond, leading to different reactivity of the hydrazido fragment upon addition of CO and upon addition of H2, showing a smaller extent of activation in the oxo complex with respect to the nitride. The diuranium(III) bridging oxide complex showed interesting reactivity with H2, forming a bis-hydride complex capable of reducing CO2 beyond the formate level, to methanol.

EPFL2019

Four-electron Reduction and Functionalization of N₂ by a Uranium(III) Bridging Nitride

Marta Falcone, Marinella Mazzanti

N2 is a cheap and widely available but very unreactive molecule. Notably, the only industrial process for the conversion of dinitrogen is the Haber-Bosch process to form ammonia, but under very harsh conditions (high temperature and pressure). Here, we review our recent research leading to the reduction of dinitrogen by four electrons, under ambient conditions by a uranium(III) bridging nitride, K3 U-N-U, where two uranium(III) cations are linked by a nitride group and a flexible metal-ligand scaffold. We also show that the bound dinitrogen can be further functionalized under mild conditions: the addition of acid, hydrogen and protons or carbon monoxide to the uranium hydrazido complex yields to the cleavage of the N-N single bond and to the formation of new N-H and N-C bonds.

2018

Transition metal clusters on h-BN/Rh(111)

Hamed Achour

The main aim of this dissertation is to study the catalytic aspects of very small transition metal clusters supported on h-BN/Rh(111). The catalytic reactions of interest were ammonia synthesis and CO oxidation on mass selected and soft landed iron and platinum clusters, respectively. A special focus was devoted to the investigation of the stability of these very small clusters on h-BN before and after the reaction. It was found that Pt7 clusters soft landed at room temperature with an energy of 1.2 eV per atom and annealed to 700 K exhibit Smoluchowski ripening. Above this temperature, these clusters undergo partial intercalation between the h-BN monolayer and the Rh(111). The intercalation initiates at 900 K and becomes more pronounced when the clusters are annealed under gas reaction conditions. A high catalytic activity was observed when the Pt catalyst remains supported on h-BN. In this case, the reaction starts at 480 K and follows Langmuir-Hinshelwood mechanism. However, the catalytic activity strongly reduces when the Pt clusters undergo partial intercalation. Nevertheless, in this case the reaction starts at only 380 K, revealing a reduction by 100 K in the Pt poisoning as a result of substrate effect and charge redistribution. In a second trail, the cluster-h-BN interaction was studied through h-BN irradiation with Pt clusters, at room temperature, within an energy window of 30-416 eV/atom. The results show that even irradiation at 30 eV/atom can lead to entire non thermal intercalation of the clusters. The energetic Pt clusters were found to be site selective as they settle only under the h-BN wires, in contrast to soft landed Pt7 clusters that were found to settle at the side edge of the h-BN depressions. After being exposed to irradiated Pt clusters, at an energy above 100 eV/atom, the h-BN layer starts to display visual cavity-type defects. These defects, induced by collision, disappeared after annealing to 600 K under gas reaction leaving behind an h-BN with no visual defects. CO temperature desorption spectroscopy (TDS) indicates that the intercalated Pt is inactive towards CO oxidation due to h-BN screening. Subsequently, we were able to produce a catalyst system made of soft-landed Pt/h-BN/intercalated Pt/Rh(111) using a combination between soft and energetic Pt deposition with which a reduction by 100 K in CO poisoning was obtained. Finally, ammonia synthesis was conducted on soft landed Fe clusters, supported on h-BN/Rh(111) and deposited at 100 K under Ultra High Vacuum (UHV). Scanning tunneling microscopy (STM) displays two imaging states of the as deposited clusters; a ring state surrounding h-BN depression which disappears above 300 K and a dot like structure. The Fe clusters are found to grow by Ostwald ripening after annealing to 600 K and above this temperature, they endure partial intercalation. It was found that, in the presence of iron, the nitrogen reaction gas experienced an exchange with the nitrogen species forming the h-BN layer. TPR (Temperature Programmed Reduction) of N

_2

with hydrogen revealed that nitrogen reduction occurs at 620 K under high vacuum following the Haber-Bosch method. The detection of reaction intermediates NH and NH

_2

together with desorbed atomic nitrogen, during ammonia synthesis, confirms that NH

_3

formation involves the stepwise hydrogenation of adsorbed nitrogen.

EPFL2017