Nitrogen is the main component of the air and the precious resource that nature gives us. As a traditional scientific problem, how to realize the activation and efficient conversion of nitrogen under mild conditions and turn it into a resource that can be used by human beings is an important research topic for chemists and even the whole scientific community. Industrial ammonia synthesis technology (Haber-Bosch Process) is one of the greatest innovations of mankind in the 20th century. It has realized the industrial production of ammonia directly from nitrogen, and has made important contributions to the development of the world economy and the progress of human civilization^[1]. However, the reaction conditions are very harsh, requiring high temperature and high pressure (about 350-500 degrees Celsius, 50-200 standard atmospheric pressure), and huge energy consumption (it is estimated that the annual energy consumption accounts for about 1% -2% of the total global energy consumption).This technology does not meet the requirements of human sustainable development, and also forces scientists to find new methods (mild reaction conditions, low energy consumption) to achieve the activation and conversion of nitrogen^[2]. In recent years, with the continuous exploration and research of scientists, the methods of converting nitrogen into nitrogen-containing compounds have gradually enriched, which can be roughly divided into the following categories: (1) Artificial simulation of biological nitrogen fixation: This method is to simulate plants to fix nitrogen in the form of nitrogen-containing compounds through nitrogenase^[3]. There are two main advantages of artificially simulating biological nitrogen fixation: first, the conditions are mild, and second, the yield is very high (about 75% of the total amount of biological nitrogen fixation in the world every year). Artificial simulation of biological nitrogen fixation also has its disadvantages: the scope of application is small, because the microorganisms with biological nitrogen fixation function can only be symbiotic with a small number of leguminous plants, but not with most of the food crops we need, so this method can not replace the industrial synthesis of ammonia. (2) Photocatalytic nitrogen fixation: Photocatalytic nitrogen fixation is a method of synthesizing ammonia from nitrogen and water by using solar energy^{[4⇓⇓⇓~8]}. Solar energy is one of the cleanest energy sources, and it is undoubtedly a major breakthrough to use semiconductor photocatalysts to convert nitrogen into ammonia. For example, :TiO₂ as a classical photocatalyst and its composite materials have been widely studied and used for photocatalytic nitrogen fixation. Binary or multi-metal oxides and sulfides, as well as various carbonaceous materials are also the focus of recent research^{[9,10][11⇓⇓⇓⇓⇓⇓~18]}. However, photocatalytic nitrogen fixation also has some significant disadvantages, such as low conversion efficiency, electron-hole separation, low quantum efficiency, poor visible light absorption and poor selectivity, mainly due to the lack of active sites for adsorption and catalysis on the surface of semiconductor photocatalysts. And (3) catalytic nitrogen fixation by a metal complex: firstly, complexe that metal complex with nitrogen to form a complex, and further hydrogenating or protonate the complex through different electron or proton transfer paths to obtain amine or other nitrogen-containing organic molecules, as shown in figure 1^[19,20][21]. Metal complexes have great potential to realize nitrogen activation under mild conditions, and will be a hot research topic in the future. The design and synthesis of novel metal complexes is the focus of nitrogen activation and transformation, and the efficient synergistic effect of metal and ligand will be a powerful breakthrough to achieve catalytic nitrogen fixation under mild conditions.

Uranium is a typical example of the actinides, which have a range of oxidation States from divalent to hexavalent^[22]. In recent years, with the revival of organometallic chemistry of actinides in non-aqueous systems, the coordination chemistry of low-valent uranium has attracted wide interest of scientists^[23]. The results show that trivalent uranium has good catalytic activity for the activation of small molecules^[24⇓~26]. Therefore, it is of great scientific significance and potential industrial application value to summarize and summarize the activation of nitrogen by trivalent uranium.

In 1998, Scott et al. First reported a complex based on a Tren ligand, trivalent uranium, chelating a nitrogen molecule^[27]. They first reduced the tetravalent uranium precursor compound 1[U(NN'₃)Cl][NN'₃=N(CH₂CH₂NSi^tBuMe₂)₃] with potassium to obtain the mixed-valence chlorine-bridged compound 2[{U(NN'₃)}₂-μ-Cl], and then sublimated it at high temperature and high vacuum to obtain the important trivalent uranium compound 3[U(NN'₃)]^[28]. They found that 3 could react in a nitrogen atmosphere at one standard atmosphere to form the compound 4[{U(NN'₃)}₂(μ²-η²:η²-N₂)], and the color gradually changed from purple to red. The reaction can be carried out reversibly by the method of freezing and degassing. The reaction process is shown in fig. 2. The whole process can be traced by nuclear magnetic resonance hydrogen spectroscopy to determine the degree of reaction. The crystal structure of compound 4 shows that the nitrogen molecule connects the two uranium centers in a side-coordinated manner, and the uranium atoms are located outside the plane composed of three nitrogen atoms, about 0.8484 Å and 0.885 Å, respectively. The bond lengths of the top N — U are 2.555 (5) Å and 2.601 (5) Å, respectively, which are shorter than the N — U bond in compound 1 (about 2.77 Å) and the N — U bond in compound 2 (2.78 (2) Å). The bond lengths of uranium with the two nitrogen atoms in the nitrogen molecule are between 2.39 and 2.4444 Å. The N-N bond length in 4 is about 1.109 (7) Å, which is basically the same as the N-N bond length in nitrogen (1.0975 0975 Å), so the activation of nitrogen molecules has not been realized in a real sense^[29].

There are two ways of metal chelating nitrogen: end group coordination and side group coordination. In the end group coordination mode, two filled metal π orbitals form a M → N feedback bond with nitrogen. In the lateral coordination mode, only one metal π orbital and one δ-orbital are available^[30]. At the same time, the reason why the end group coordination is more favorable than the side group coordination is that the π orbital is more favorable than the δ-orbital.

In 1998, Cummins et al. Used a heteronuclear bimetallic strategy to achieve nitrogen activation^[31]. They first used Li(N[R]Ar)(OEt₂) as a substrate to react with UI₃(THF)₄ to obtain a yellow tetravalent uranium compound 5[UI(NRAr)₃],5, which was further reduced by Na/Hg to obtain a trivalent uranium compound 6[U(THF)(NRAr)₃]. 6 reacted with Mo(N[t-Bu]Ph)₃ at a ratio of 1:1 in toluene under a N₂ atmosphere to almost quantitatively form the orange crystalline compound 7a(Ph[t-Bu]N)₃Mo(μ-N₂)U(N[R]Ar)₃ (fig. 3). A reasonable guess is that the possible intermediate (N₂)Mo(N[t-Bu]-Ph)₃ is more likely to be captured by compound 6 than by the Mo(N[t-Bu]Ph)₃ itself. Using a similar approach, a more sterically hindered 7b(Ar[Ad]N)₃Mo(μ-N₂)U(N[R]Ar)₃ was successfully synthesized (fig. 3). At the same time, the more sterically hindered structural units of molybdenum make nitrogen more likely to chelate with bimetals through end-group coordination. Both 7 a and 7 B were further structurally determined by single crystal diffraction. The structures indicate that both U and Mo are in the + 4 oxidation state, indicating a two-electron reduction. Mo is a more efficient pi donor for diazonium complexes.

In 2002, Cloke et al. Effectively achieved nitrogen activation using a trivalent uranium complex derived from a pentacyclopentadiene ligand^[32]. Silylated pentacyclopentadiene dianions are an important class of ligands, and their chelated trivalent uranium complexes have been successfully bonded to carbon monoxide. Inspired by this work, they look forward to further using the complex for the activation of nitrogen. [UI₃(THF)₄] is an important raw material for the synthesis of trivalent uranium precursor compounds.However, in order to synthesize the pentacyclopentadiene trivalent uranium complex without strong ligand coordination for the subsequent activation of small molecules, they used uranium chips and HgI₂ to react at 320 ℃ in a sealed tube to realize the synthesis of UI₃. The reaction of UI₃ with equivalent amounts of KCp^* and K₂[C₈H₄{SiⁱPr₃-1,4}₂], respectively, gave the violet-black compound 8[U(η-Cp^*)(η-C₈H₄{SiⁱPr₃-1,4}₂) in 40% yield]. When compound 8 was placed in a nitrogen atmosphere, both nuclear magnetic resonance hydrogen spectrum and silicon spectrum produced new signals. After freezing, it was found that the signal peak of compound 8 completely disappeared, and a new dark green compound 9 was produced (fig. 4). The single crystal structure shows that compound 9 has a binuclear structure, and the nitrogen molecule forms a complex with two molecules of compound 8 through side coordination. The N-N bond length is 1.232 (10) Å, which is very close to the N-N double bond length (1.259 (4) Å) in [ (TmCp^*)₂(μ-η²:η²-N₂)], but significantly larger than the N-N triple bond length (1.109 (7) Å) in compound 4[(U{NN'₃})₂(μ-η²:η²-N₂)]^[33]. The above results also show that the reason why compound 3[U(NN'₃)] can not reduce nitrogen is probably due to the steric hindrance of the ligand in compound 4[(U{NN'})₂(μ-η²:η²-N₂)] and the small orbital overlap between the U center and the μ-η²:η²-N₂. Although the bond orders of N — N bonds in compounds 3 and 9 are significantly different, the bond lengths of U — N bonds are similar, which may be due to the different frontier orbital geometries in the two ligand environments.

The formation of compound 9 involves a transition from trivalent uranium to tetravalent uranium, accompanied by a two-electron reduction of the nitrogen molecule. However, even under a nitrogen atmosphere of 3.4 standard atmospheres, the reaction only proceeds to about 75%, and 9 is extremely susceptible to losing nitrogen molecules both in solution and in the solid state. The instability of compound 9 losing its N₂ to regenerate 8 may be due to the fact that the space is more crowded, and the parallel sandwich structure can be regenerated after releasing nitrogen.

In 2003, Evans et al reported the first new method of chelating nitrogen molecules with uranocene^[34]. Compound 10 needs to be efficiently converted to compound 11 at 5.4 standard atmospheres. Moreover, the reaction is reversible, with 11 being quantitatively converted to 10 when the pressure is reduced to 1 ATM (fig. 5). The single crystal structure shows that the nitrogen molecules are bonded to the low-valent uranium precursors through the end groups. The N-N bond length is 1.120 (14) Å, which is basically consistent with the bond length of free nitrogen (1.0975 0975 Å), so there is no real activation of nitrogen, which is due to the weak σ-donor and π-acceptor properties of N₂, resulting in the weak interaction between uranium and N₂. However, this was the first case of the formation of a monometallic -N₂ complex by terminal coordination of an f-block element at that time.

In 2011, Arnold et al. Found that the coordination of trivalent uranium precursor compound 12 with phenoxy complex to nitrogen could also be achieved to form compound 13 under the condition of 1standard atmosphere^[35]. The single crystal structure shows that the nitrogen molecules are complexed with two low-valent uranium precursors by side coordination. The N-N bond length is 1.189 (18) Å, which is slightly longer than that of compound 11 (1.120 (14) Å), but shorter than that of compound 9 (1.232 (10) Å), which also indicates that the uranium nucleus has a certain reduction effect on nitrogen. In compound 13, the π-symmetric orbital electrons of uranium feed back to the π^*- antibonding orbital of N₂, which makes the uranium have a stronger activation to the side-coordinated N₂ and form a more stable compound.

The above shows that although trivalent uranium has shown some potential as an important synthon in the chelation and activation of nitrogen, the conversion of nitrogen into important compounds such as amines or organic amines has not yet been reported.The main reason is that the above complexes can only play the role of coordination or two-electron reduction, and can not achieve further reduction and transformation of nitrogen.

In 2017, Mazzanti and others used the trivalent uranium complex derived from the large steric silanol ligand developed by their research group to realize the activation and efficient conversion of nitrogen^[36]. They first used KC₈ to reduce the previously reported tetravalent uranium precursor compound 14[K₃{U(OSi(O^tBu)₃)₃}₂(μ-N) to obtain the important trivalent uranium compound 14[K₃{U(OSi(O^tBu)₃)₃}₂(μ-N). Compound 14 has higher chemical stability and better crystallinity compared with the analogues of Cs, which also provides a prerequisite for the exploration of subsequent reactions. The single crystal structure of 14 is shown to be a nitrogen-bridged two trivalent uranium-chelated building block, counterbalanced by three potassium ions. Moreover, the existence of the silicon-oxygen-potassium framework also enables the flexible silanol ligand to better connect the two uranium nuclei, thus stabilizing the trivalent uranium compound more effectively. Compound 14 was efficiently converted to compound 15 in 68% yield at room temperature in solid form or dissolved in toluene under a nitrogen atmosphere of 1 ATM. The single crystal structure of 15 shows that the N-N bond length is 1.521 (18) Å, which is very close to the N-N bond length of NH₂-NH₂ (1.4747 Å). This indicates that nitrogen is reduced by four electrons and uranium is oxidized from trivalent to pentavalent. The valence state of uranium is further confirmed by electron paramagnetic resonance and variable temperature magnetism. They believe that the flexible silicon-oxygen-potassium skeleton plays a key role in the bending of the U-N-U core to chelate nitrogen. This is also the first time that trivalent uranium complexes have been found to be activated by nitrogen through four-electron reduction.

Ammonia can be obtained in 25% ~ 42% yield by treating 15 with hydrochloric acid. Ammonia can be obtain in 77% yield if it is first reduced with hydrogen and then further treat with acid. If only hydrogen is reduced, it takes two to three weeks for the substrate to be completely converted. When compound 15 was treated with two equivalents of PyHCl, the N-N bond was cleaved to give compound 16. When PyHCl was further increased to twenty equivalents, NMR showed that ammonium chloride was the only product. Compound 15 is reacted with carbon monoxide to give the oxocyanate compound 17. The single crystal structure of 17 shows that the dinuclear uranium is bridged by oxygen and two OCN^- fragments. At the same time, pentavalent uranium is reduced to tetravalent uranium. At the same time, compound 15 was treated with 2,4,6-tri-tert-butylphenol to give NH^2- and {{\mathrm{N}}_2}^{4-} bridged dinuclear uranium compound 18, which reacted with carbon monoxide to give NH^2- and NCO^- bridged mixed-valence U (III) and U (IV) dinuclear uranium compound 19, and nitrogen was removed (Fig. 7).

In 2019, Mazzanti et al. Reported the activation of nitrogen by an oxygen-bridged dinuclear uranium precursor supported by a sterically hindered silicon-oxygen ligand^[37]. They used trivalent uranium compound 20 as the starting material and treated it with nitric oxide complex of nitrogen heterocyclic carbene to obtain oxygen-bridged tetravalent uranium compound 21. 21 is reduced by KC₈ to give the potassium ion-counterbalanced oxygen-bridged trivalent uranium compound 22. 22 can undergo a four-electron redox reaction with nitrogen to give compound 23. The single crystal structure shows that the N-N bond length is 1.40 (1) Å, which is very close to the N-N bond length of hydrazine (1.4747 Å), and is also within the range of the normal N-N single bond length of hydrazine compounds (1.377 ~ 1.548548 Å). The above results favorably indicate that nitrogen undergoes a four-electron reduction to an N-N single bond and that uranium is oxidized from trivalent to pentavalent. Compound 23 reacted with three equivalents of carbon monoxide to give compound 24. The single crystal structure shows that the dinuclear uranium is bridged by two oxygen and one NCN^2- unit. Compound 24 is further reduced to dinuclear uranium (VI) compounds by further reaction with 17 equivalents of carbon monoxide, while the NCN^2- unit is converted to nitrogen-containing compounds NCNCO for removal. However, unlike compound 15, which can be efficiently converted to a nitrogen-containing compound, compound 23, whether treated with acid or reduced with hydrogen, does not yield the corresponding nitrogen-containing compound, but is removed as nitrogen. This large difference is due to the significant difference in reaction performance caused by the different binding modes and electronic structures of the alkali metal ions (fig. 8).

In 2020, Arnold et al. Realized the reduction and functionalization of nitrogen by using dinuclear uranium complexes with rigid frameworks, and could effectively convert them into secondary silylamine compounds^[38]. They used meta-substituted tetraarylphenols as ligands to obtain dinuclear tetravalent uranium compounds 26. From the single crystal structure of 26, it can be seen that due to the coordination unsaturation of uranium, there is an electrostatic interaction between the bridged aryl and uranium, which increases its stability. After the reduction of 26 in a nitrogen atmosphere KC₈, nitrogen is activated with the activation of carbon-hydrogen bonds to give the polymetal chelated hydrazine adduct 27. The single crystal structure of 27 shows that the carbon-hydrogen bond of triarylmethane is activated, the hydrogen migrates to the nitrogen atom, and the C-U bond is formed. At the same time, the four potassium-aryl interactions play a role in charge balancing and enhance the overall structural stability. Both compounds 26 and 27 can reduce nitrogen to an amine compound. 26 was reduced with metallic potassium under nitrogen and further acidified with osmium and silanized with Me₃SiCl to give the disilylamine derivative. 27 were treated with pyridine hydrochloride to give ammonium chloride, the original ligand, uranium tetrachloride and potassium chloride, respectively; Ammonia can be obtained by acidification with tritium (fig. 9). The above results show that the cleavage of N — N bonds and the formation of nitrogen-containing compounds can be achieved by acidification. They believe that the formation of aryl-bridged cryptins from the bimetallic biligand metal macrocyclic precursors provides a key site for the activation of nitrogen.

Low-valent uranium complexes have shown great potential for nitrogen activation. The realization of nitrogen activation and conversion with mild conditions and low energy consumption is of great significance to the sustainable development of human beings, so the application prospect is also very broad. Compare with that nitrogen activation catalyzed by the transition metal complex, the nitrogen activation promoted by the low-valent uranium has more remarkable advantages, because the uranium is highly polarized, the energy level of the 5 f orbital of the uranium is more easily affect by the crystal field splitting effect due to the relativistic effect, and the covalent bond component of the formed complex is higher; Moreover, trivalent uranium has a strong reduction ability, which is an important reason for the efficient catalytic activation of nitrogen by low-valent uranium complexes. Of course, uranium is radioactive to a certain extent, so there are potential dangers and hazards. As a result, the conditions for the practical operation of uranium-containing compounds are more demanding and stringent than those for transition metals. However, there are still some problems to be solved: (1) there are very few low-valent uranium complexes available for nitrogen activation; (2) It has been reported that the activation and conversion of nitrogen require the destruction of the complex itself, so the catalytic efficiency is low. Therefore, how to realize the catalytic conversion of nitrogen activation and the recycling of low-valent uranium precursor catalyst is an important research content of low-valent uranium catalytic nitrogen activation in the future. Of course, there is still a long way to go because it requires reasonable design of relevant ligands and precise construction of their complexes. However, we believe that with the deepening of research, the catalytic cycle of low-valent uranium complexes in nitrogen activation will be realized in a real sense in the near future.