Synergism between cyclopentadienyl and amidinate ligands affording anionic scandium terminal imido complexes

Abstract

Terminal rare-earth imido complexes containing metal–nitrogen double bonds have received more attention in recent years due to their importance in group transformation and catalytic reactions. However, due to the large difference in the orbital energy between rare-earth metals and nitrogen, their synthesis is difficult and the product is easy to polymerize. Here, we use the combination of Cp* and amidinate ligands to inhibit the tetramerization and provide exclusively the first anionic rare-earth(III) terminal imido complexes with both electron-donating and electron-withdrawing groups. Chemical bond analysis further confirms the double-bond character, and the strong polarity of the RE=N bond, which could be described as three orbital interactions, is primarily derived from the imido nitrogen, while the contribution from the rare-earth metal is limited. The mechanistic study using DFT calculations shows that the formation of the RE=N bond involves the activation of two N–H bonds. Furthermore, the anionic rare-earth(III) terminal imido complex shows some interesting and unique reactivity towards isocyanates, isonitriles, phenylsilanes, and W(CO)₆. The work extends the multiple-bond chemistry between rare-earth metals and main group elements, and is expected to inspire the development of rare-earth organometallic chemistry and related fields.

---

Introduction

Transition metal terminal imido complexes have been extensively studied in the past four decades, because of their importance in group transformation and catalytic reactions.¹ In contrast, studies on rare-earth metal terminal imido complexes are limited despite recent significant advances.² This is mainly due to (i) the extremely large energy gap between valence orbitals of the rare-earth metal and nitrogen, and (ii) the large sizes and high Lewis acidities of rare-earth metals usually causing bimolecular decomposition or oligomerization, which renders them difficult to isolate and characterize.³ Although the initial attempt to make rare-earth metal terminal imido complexes can be traced back to 1991,⁴ the bridging species was obtained when the imido ligand was utilized. Afterwards, Xie,⁵ Gordon,⁶ Hessen⁷ and Piers et al.⁸ independently made many attempts, but either the bridging species or species stabilized by alkali were obtained. The isolation and characterization of true scandium terminal imido complexes were not achieved until 2010 by Chen et al.⁹ So far, the metal centers of rare-earth terminal imido complexes have been extended from Sc^III to Y^III,¹⁰ Dy^III,^10b Ho^III,^10b Lu^III,^10a and Ce^IV.¹¹

The isolation and characterization of trivalent rare-earth terminal imido complexes are attributed to the stability of these sterically encumbered monoanionic ligands and the substituent on the imido nitrogen. The steric hindrance inhibits dimerization,^5–7,12 trimerization,¹³ tetramerization⁴ and other oligomerizations.^5,14 The use of either C₅Me₄SiMe₃ (Cp′) or amidinate (Am, R′′C(R′N)₂⁻) ligands inevitably leads to the formation of tetrameric rare-earth imido complexes (Scheme 1a).¹⁵ Once tetramers form, they are difficult to depolymerize to rare-earth terminal imido complexes. Interestingly, the combination of C₅Me₅ (Cp*) and MeC(ⁱPrN)₂⁻ ligands inhibits the tetramerization to provide exclusively the first anionic rare-earth(III) terminal imido complexes (Scheme 1b). By using this strategy, rare-earth terminal imido complexes with strong electron-withdrawing groups (e.g. C₆F₅) on the imido nitrogen can be synthesized. Furthermore, some calculations have been used to study the mechanism of this synthesis and the bonding between scandium and nitrogen. Some of the following reactions also reflect the characteristics of such imido compounds from the experimental aspect.

Scheme 1 Terminal rare-earth imido complexes.

Results and discussion

Synthesis of scandium terminal imido compounds

Our group has reported the applications of complex 1 with the combination of Cp* and Am ligands in dinitrogen conversion and synthesis of rare-earth metallacycles.¹⁶ In this research, salt metathesis and deprotonation of 1 with 4 equiv. of KBn (2 equiv. per Sc) resulted in the formation of the benzylated complex, Cp*(Am)ScBn, 2 (Scheme 2). The crystal structure of 2 confirmed this C=C double bond (C8–C9 1.361(2) Å) in the amidinate group (Am) (Fig. 1), which is much shorter than the length of the C–C single bond in 1 (1.505(7) Å).¹⁷ The benzyl compound 2 can be used as a base to react with aniline containing different aryl groups to afford the corresponding imido compounds. When 2,6-diisopropylaniline having electron-donating properties was used, the corresponding scandium imido complex 3a, featuring an Sc=N double bond, could be obtained successfully (Scheme 2a). In 3a, the distance between Sc and N^imido is 1.9086(18) Å, a little longer than that in the previous report of the Sc=N distance (1.881(5) Å) in its scandium imido complex,^9b while the angle of Sc–N^imido–C is nearly linear (162.24(15)°). Although the bond length of Sc=N clearly belongs to the Sc=N double bond (Fig. 1-3a), the charge buildup on nitrogen affords close-contact pairing with the potassium ion. To obtain the imido compound with a terminal Sc=N double bond, [2.2.2]cryptand (crypt) was added to compound 3a, which led to the isolation of the separated ion pair (SIP) 4a in 95% yield (Fig. 1-4a). In 4a, the angle of Sc–N^imido–C is 175.40(15)°, which is closer to linear compared to 3a. What's more, the bond length of Sc=N decreased slightly (1.8897(17) Å). This counteracts the trend of increased electron density in the Sc=N double bond after removing the K cation.

Scheme 2 The synthesis of terminal scandium imido complexes.

Fig. 1 Molecular structures of 2; 3a; 3b; 4a; 4b. The hydrogen atoms are omitted for clarity, and ellipsoids are shown at the 30% probability level. Selected bond lengths (Å) and angles (°) of 2: Sc1–N1 2.0667(12), Sc1–N2 2.0813(12), Sc1–C1 2.2999(14), K1–N2 2.9930(12), K1–C4 3.1500(16), K1–C7 3.2269(15), K1–C8 2.8683(14), N1–C8 1.4006(18), N2–C8 1.4092(19), C8–C9 1.361(2), C2–C1–Sc1 116.37(9); 3a: Sc1–N1 1.9086(18), Sc1–N2 2.2470(18), Sc1–N3 2.2414(19), K1–N1 2.8270(18), K1–N2 3.0693(18), K1–N3 3.1078(19), N1–C1 1.347(3), N2–C13 1.335(3), N3–C13 1.336(3), C13–C14 1.509(3), C1–N1–Sc1 162.24(15); 3b: Sc1–N1 1.9547(13), Sc1–N2 2.2222(13), Sc1–N3 2.2142(14), K1–N1 2.7914(13), K1–F5 2.7615(11), K1–C1 3.2945(15), N1–C1 1.3394(19), N2–C7 1.340(2), N3–C7 1.335(2), C7–C8 1.515(2), C1–N1–Sc1 141.34(11); 4a: Sc1–N1 1.8897(17), Sc1–N2 2.2186(18), Sc1–N3 2.2492(17), N1–C1 1.353(3), N2–C13 1.333(3), N3–C13 1.327(3), C13–C14 1.516(3), C1–N1–Sc1 175.40(15); 4b: Sc1–N1 1.9172(18), Sc1–N2 2.1902(16), Sc1–N3 2.1911(17), N1–C1 1.310(3), N2–C7 1.332(3), N3–C7 1.335(3), C7–C8 1.509(3), C1–N1–Sc1 170.82(16).

In addition, there are no other reports on rare-earth imido compounds with electron-withdrawing groups other than 3,5-(CF₃)₂C₆H₃ being used successfully.¹¹ As shown in Scheme 2b, compound 2 reacted with aniline bearing a pentafluorophenyl group, resulting in the rapid formation of compound 3b at −30 °C as yellow crystals. X-ray diffraction studies revealed the double bond character of Sc=N in compound 3b (Sc1–N1 1.9547(13) Å) and the deviation of the Sc–N^imido–C angle from linearity (141.34(11)°) (Fig. 1-3b). Similarly, we added 18-crown-6 to this reaction and obtained the SIP 4b at −30 °C. Its crystal structure shows that the length of the Sc=N double bond (1.9172(18) Å) is shorter than that in compound 3b, and the angle of Sc–N^imido–C (170.82(16)°) is more linear (Fig. 1-4b). The length of the corresponding Sc=N double bond in 3b is a little longer than that found in 4b, which is consistent with the change of substituents in imido groups.

Computational studies

Based on the results above, terminal rare-earth imido complexes featuring a significant Sc=N double bond can be easily synthesized from 2 under mild conditions, whether in the presence of electron-withdrawing or electron-donating groups. To deepen understanding of these systems, we performed density functional theory (DFT) calculations to explore the nature of the Sc=N bond in these compounds. We first performed the geometry optimization calculations (see the ESI† for computational details) on the four molecules (3a, 4a, 3b, and 4b) at the PBE0-D3 level with the SDD basis set for Sc and 6-31G* for other elements. In 4a and 4b, the K cations are encompassed by cryptand or 18-crown-6, thereby having little interaction with the Sc=N double bond. Therefore, K(L)-free 4a′ and 4b′ were used for optimization. The four optimized structures 3a, 4a′, 3b, and 4b′ are presented in Fig. S30,† and the selected geometrical parameters suggest good agreement between the experimentally and theoretically determined structures. The Sc=N bond length in 3a is noticeably short (1.857 Å), which is very close to the previously reported bond length of the Sc=N double bond^9a,18 along with a Mayer bond order (MBO) of 1.58, confirming the obvious double bond character of 3a. For the Sc=N bond of 4a′, the slightly changed bond length (1.871 Å) and the MBO (1.49) reflect the negligible impact from the presence of the K cation. Similar results were also found in 3b and 4b′. The bond lengths (3b: 1.883 Å; 4b′: 1.876 Å) and the MBO values (3b: 1.21; 4b′: 1.36) of the Sc=N bonds confirm their double bond character and minimal interactions between the K cation and the Sc=N bond in 3b and 4b′. It should be noted that in the DFT-optimized structures, the Sc–N bond length is shorter than that in the crystal structure. Besides calculation errors, this is mainly due to the fact that the DFT optimization was carried out under gas-phase conditions, without considering the packing effects in the solid-state crystal.

The molecular orbital (MO) analysis was performed based on the DFT-optimized structures at the PBE0-D3 level, using the SDD basis set for Sc and def2-TZVP for the other elements. As shown in Fig. 2a, taking 3a as an example, the Sc=N bond of 3a can be described as three localized molecular orbitals (LMOs), σ₁, π₁ and π₂, which derive from the interactions between 3d orbitals of the Sc center and 2s/2p orbitals of the imido nitrogen. The strong polarity of the Sc=N bond is supported by orbital composition analysis from the Hirshfeld method. The imido nitrogen contributes significantly to the three orbital interactions, with 25.8%, 53.5% and 29.4%, respectively. In contrast, the Sc center plays a relatively minor role, making smaller contributions (1.6%, 14.3% and 8.2%) to the respective bonding interactions. Similarly, the three LMOs in 4a′ (Fig. 2b) are mainly from the imido nitrogen (σ₁: 19.7%, π₁: 55.0%, π₂: 29.9%), while the Sc center provides the smaller contributions (σ₁: 1.5%, π₁: 15.5%, π₂: 8.7%). The imido nitrogen primarily acts as a basic site, highlighting the significant polarity of the RE=N bond. This further confirms the previous interpretation of the difference of Sc=N bond lengths in 3a and 4a. On the other hand, Fig. 2c and d show that the Sc=N double bonds of 3b and 4b′ exhibit similar polar properties, even in the presence of the electron-withdrawing groups. These electron-withdrawing groups further increase the polarity of the Sc=N bond, significantly reducing the contribution of the π₁ orbital on the nitrogen atom.

Fig. 2 Plots of the localized molecular orbitals (LMOs), including σ₁, π₁, and π₂, calculated at the PBE0-D3 level with the SDD basis set for Sc and def2-TZVP for other elements, for the DFT-optimized structures of (a) 3a, (b) 4a′, (c) 3b, and (d) 4b′. The percentages in red and blue indicate the contributions of the imido nitrogen and scandium. Color codes for atoms: silver – C; blue – N; brown – Sc; purple – K. The H atoms are omitted for clarity. The isovalue is 0.05.

According to the above DFT analysis, the bonding characteristics of these four compounds are essentially identical, with bond lengths ranging from 1.85 to 1.88 Å and bond orders between 1.2 and 1.6, reflecting clear double bond characteristics. Orbital analysis indicates the formation of one σ interaction and two π interactions between the d orbitals of the Sc center and the p orbitals of the N atom. These observations are consistent with other reported RE=N double bonds.^9a,11a,19 The calculated bond lengths for these compounds fall within the range of 1.85 to 2.0 Å, with bond orders between 1.3 and 1.7.^9a,11a Moreover, a similar bonding pattern (one σ and two π interactions) was also observed in these compounds. These are generally consistent with our results.

Furthermore, we were curious about the formation of the polar Sc=N double bond and therefore explored the formation mechanism of these compounds through DFT calculations in the gas phase (PBE0-D3/SDD for Sc, 6-31G* for other elements) and in a solvent (PBE0-D3/SDD for Sc, def2-TZVP for other elements). We chose the K-free benzyl 2′ as a starting model compound and 4a′ as a targeted model for simplicity, and intuitively proposed the first mechanism (the black line). As indicated in Fig. 3, we initially explored this pathway: ArNH₂ firstly coordinates to the scandium center, leading to the formation of a four-membered ring transition state, TS1′, which subsequently gives an intermediate, IM1′, featuring an Sc–N single bond, accompanied by the release of toluene. Then, the Sc=N double bond intermediate IM2′ will be obtained via a four-membered ring transition state TS2′ with a high-energy barrier (27.0 kcal mol⁻¹). In the final step, the product 4a′ is obtained with an exceptionally high-energy barrier (42.8 kcal mol⁻¹) in the proton-transfer process viaTS3′. The high-energy barriers of TS2′ and TS3′ render this mechanism ineffective. After checking the Mulliken charge of potential active sites, we found that the carbon in the methylene group of the amidinate ligand exhibits nucleophilicity, along with reduced steric hindrance. Based on the above information, we proposed the second mechanism, as depicted by the red line in Fig. 3. Here, 2′ and ArNH₂ first undergo the N–H activation process viaTS1, leading to the formation of an ionic pair IM1, with a relatively low energy barrier (10.1 kcal mol⁻¹). Subsequently, IM1 transforms into energetically favorable IM2. Afterward, IM2 undergoes another N–H activation process, crossing over a four-membered cyclic transition state TS2 with the highest energy barrier of 19.5 kcal mol⁻¹, and then gives 4a′ by eliminating toluene. The result suggests that this pathway of two N–H bond activations, leading to the formation of 4a′, is feasible at room temperature.

Fig. 3 Two DFT-calculated energy profiles for the formation of 4a′. The energies in the gas phase and solvent were calculated at the levels of PBE0-D3/SDD for Sc, 6-31G* for other elements (gas phase) and PBE0-D3/SDD for Sc, def2-TZVP for other elements (solvent).

Reactivity studies

The previously reported compounds with polar RE=N bonds have demonstrated a wide range of reactivity, including insertion reactions, acid–base reactions, addition reactions, and metathesis reactions.^{8b,9b,11b,18b,20} This leads us to speculate that these compounds may exhibit diverse reactivity due to the presence of the polar Sc=N double bond. In order to study the chemical properties of this kind of imido compound, the more stable compound 4a was chosen as the substrate to react with a series of small molecules, some of which can show novel reactivities. In the reaction of 4a with trimethylsilyl-substituted isocyanate (TMSNCO), the Sc=N double bond can react with the polar C=O double bond through [2 + 2] cycloaddition to provide complex 5 with an Sc–N–O–C four-membered ring (Scheme 3a). The crystal structure of 5 shows that the length of Sc–N^imido (2.1594(15) Å) can be considered as that of an Sc–N single bond, and C=N^TMS (1.304(3) Å) can be recognized as a double bond (Fig. 4-5). Such a reaction of the scandium terminal imido complex with phenyl isocyanate was reported by Chen et al. to afford the isomerization product after [2 + 2] cycloaddition.^18b In our reaction, the trimethylsilyl substituent in the substrate has larger steric hindrance than the 2,6-diisopropyl group in the reactant, so no rearrangement of the imido fragment occurs.

Scheme 3 Reactions of compound 4a with isocyanate, isonitrile, phenylsilane and hexacarbonyltungsten.

Fig. 4 Molecular structure of the compounds 5–8. The counter cation [K(crypt)]⁺ and hydrogen atoms are omitted for clarity, and ellipsoids are shown at the 30% probability level. Selected bond lengths (Å) and angles (°) of 5: Sc1–O1 2.0735(14), Sc1–N1 2.1596(16), Sc1–N2 2.2542(19), Sc1–N3 2.1844(19), O1–C21 1.332(2), N1–C1 1.410(2), N1–C21 1.378(2), N2–C13 1.320(4), N3–C13 1.330(4), C13–C14 1.521(4), O1–C21–N1 109.75(15); 6: Sc1–N1 2.1059(12), Sc1–N2 2.2152(13), Sc1–N3 2.2284(13), Sc1–C1 2.2076(14), N1–C1 1.3825(18), N1–C10 1.4215(19), N2–C22 1.334(2), N3–C22 1.329(2), N4–C1 1.3114(19), N4–C2 1.4011(19), C22–C23 1.513(2), N1–C1–Sc1 67.36(7), N4–C1–Sc1 170.76(11); 7: Sc1–N1 2.0790(15), Sc1–N2 2.2064(16), Sc1–N3 2.2471(18), Sc1–H1 1.950(19), Si1–N1 1.7637(18), Si1–H1A 1.406(19), Si1–H1B 1.491(18), N1–C1 1.413(2), N2–C19 1.334(3), N3–C19 1.328(3), C19–C20 1.520(3), Si1–N1–Sc1 100.10(7); 8: W1–C1 2.288(5), W1–C22 2.027(6), Sc1–O1 2.057(3), Sc1–N1 2.248(4), Sc1–N2 2.184(4), Sc1–N3 2.229(5), O1–C1 1.323(6), O2–C22 1.142(8), N1–C1 1.332(7), N 1–C2 1.433(7), N2–C14 1.320(8), N3–C14 1.328(8), C14–C15 1.523(9), O1–C1–N1 111.3(4).

Similarly, 4a can also react with isonitrile under mild reaction conditions, and the corresponding scandium aziridine 6 was obtained through [2 + 1] cycloaddition (Scheme 3b). The structure of 6 was confirmed by X-ray diffraction analysis. The length of C–N^imido (1.3826(18) Å) is closer to that of a C–N single bond while the length of the C–N^isonitrile bond (1.3113(19) Å) is closer to that of a C=N double bond (Fig. 4-6). This scandium-containing three-membered ring structure was obtained from a scandium imido compound and isonitrile for the first time. In this reaction, the carbon in isonitrile reacts as a carbene to undergo insertion into Sc=N instead of undergoing cycloaddition caused by electron distribution.

Besides, 4a can undergo an addition reaction with phenylsilane like that with an alkene (Scheme 3c). The ¹H NMR spectrum of compound 7 shows one broad singlet at 6.21 ppm which is close to that of the reported scandium terminal hydride [LScHNHDipp] (L = [MeC(N(Dipp))CHC(Me)(NCH₂CH₂NMe₂)]⁻, Dipp = 2,6-ⁱPr₂C₆H₃), which exhibits a broad signal at 6.39 ppm,^20d and one doublet at 5.25 ppm for two hydrogen atoms of SiH₂Ph. These NMR data are comparable to the data reported for scandium anilido hydride, whose X-ray structure was not obtained.^20a Besides, Anwander et al. also reported the reaction of a lanthanum diimide complex with phenylsilane, from which they obtained a lanthanum hydride complex with an La₂H₂ core.^20f Here, differently, we confirmed the structure of compound 7 from X-ray diffraction studies that showed the electropositive silicon group being added to nitrogen in the imido fragment, while electronegative hydrogen is attached to the metal center to obtain scandium hydride. As shown in Fig. 4, the length of the Sc–H bond (1.950(19) Å) in 7 is similar to that in scandium hydride synthesized from hydrogen (1.853(19) Å).^20d

In addition, 4a reacted with W(CO)₆ to produce an Sc–N–O–C four-membered ring for the first time (Scheme 3d). The crystallographic data show that the bond lengths of C–O (1.323(6) Å) and C–N (1.332(7) Å) are both between that of the corresponding single bond and double bond. Compared to the Sc–N–O–C four-membered ring in compound 5, the most notable difference in the data for 8 is the length of C–N (5: 1.377(2) Å; 8: 1.332(7) Å), indicating that there is a three-center four-electron bond (3c-4e) between C, N and O in compound 8.

The study of rare-earth metal imido complexes started in 1991,⁴ while the terminal imido complexes that can best reflect the metal–nitrogen double bond properties were not reported until 2010 by Chen et al.⁹ Since then, the studies of terminal rare-earth imido compounds are in the ascendant. Furthermore, the Lewis base assistance method they reported has become the most widely used synthesis method. Other researchers, such as Mindiola,^18a Cui,^18c Parvez,^8a Anwander,¹⁰ Mountford,^19b Gelfand^8b and others, have also tried this method and successfully captured terminal imido complexes. A Lewis acid assistance method was reported by Anwander,²¹ and a strong base assistance method was reported by Schelter, which produced the terminal imido Ce(IV) complex.¹¹ The synthetic methods of rare-earth imido compounds are very limited due to their particularity. In addition, the coordination of a Lewis base can not only stabilize such neutral rare-earth imido compounds, but also reduce their reactivity to some extent.^20c Therefore, we expected to find a balance between the stability and reactivity of imido compounds.

In our ligand system, the amidinate group mainly worked as a reservoir of protons. The terminal methyl group is somewhat acidic, which can react with the basic benzyl group. Also, the methylene part in compound 2 can react as a basic group to promote the dehydrogenation of aniline.²² Based on this, we used the coordination ability and basicity of the benzyl group to shape compound 2 with dianion properties. This method can yield salt-free anionic terminal imido complexes with the coordination of [2.2.2]cryptand and 18-crown-6.

Influence of the electronic effect on different anilines

In this work, we have successfully obtained four scandium terminal imido compounds with two kinds of anilines, which have an electron-donating effect and an electron-withdrawing effect, respectively. The length of the Sc–N double bond clearly shows little difference between these compounds. Compared with 3b and 4b containing electron-withdrawing substituents, compounds 3a and 4a have a relatively short Sc–N bond length (Fig. 1). In order to clarify such electronic influence, we conducted DFT calculations and found that in 3a and 4a, the electron-donating property of the isopropyl group allows the electrons of the scandium nitrogen double bond to be more evenly distributed among the metal center and nitrogen, which enhances the strength of the π bond between scandium and nitrogen. Therefore, the scandium imido compound containing an electron-donating substituent has better stability and milder reactivity.

In 4a′ and 4b′, the pentafluorophenyl group has a strong electron-withdrawing ability, which makes the polarity of Sc=N large and electrons more concentrated on nitrogen, resulting in weak interaction between Sc and N. This is confirmed by the smaller MBO (1.36) and more significant polarity of the Sc=N bond from the DFT calculations. At the same time, this strong electron-withdrawing ability enhances the nucleophilicity of nitrogen and improves the reactivity of compound 4b′, making it difficult to obtain stable products in subsequent activation reactions.

Conclusions

In conclusion, we have produced a novel method for the synthesis of rare-earth metal terminal imido compounds. These imido products not only reflect some classical characteristics of a polar metal nitrogen double bond but also have potential diverse reactivity. DFT calculations show that 2′ can react with ArNH₂ (Ar = 2,6-C₆H₃ⁱPr₂) to furnish the rare-earth metal terminal imido compound featuring the RE=N double bond. The chemical bond analysis further confirms the double bond character and the strong polarity of the RE=N bond, which could be described as three orbital interactions, which is primarily derived from the imido nitrogen, while the contribution from the rare-earth metal is limited. Such anionic terminal imido complexes exhibit interesting and unique reactivities, including insertion and cycloaddition reactions. Further research on the derivatization of such terminal imido compounds and the synthesis of imido compounds with other rare-earth metals is in progress.

Author contributions

All authors have given approval to the final version of the manuscript.

Data availability

The authors confirm that the data supporting the findings of this study are available within the article and its ESI.†

Conflicts of interest

The authors declare no competing financial interests.

Acknowledgements

This work was supported by the National Key R&D Program of China (no. 2021YFF0701600), the National Natural Science Foundation of China (no. 22371006 and 22131001), and the Beijing National Laboratory for Molecular Sciences (BNLMS-CXXM-202401). We thank the High-performance Computing Platform of Peking University for computational resources and the NMR facility of the Analytical Instrumentation Center. We acknowledge Prof. Dr Zhenfeng Xi of Peking University for useful discussions and comments on the manuscript.

References

1. (a) L. H. Gade and P. Mountford, New Transition Metal Imido Chemistry with Diamido-Donor Ligands, Coord. Chem. Rev., 2001, 216, 65–97; (b) A. P. Duncan and R. G. Bergman, Selective Transformations of Organic Compounds by Imidozirconocene Complexes, Chem. Rec., 2002, 2, 431–445; (c) N. Hazari and P. Mountford, Reactions and Applications of Titanium Imido Complexes, Acc. Chem. Res., 2005, 38, 839–849; (d) D. J. Mindiola, Oxidatively Induced Abstraction Reactions. A Synthetic Approach to Low-Coordinate and Reactive Early Transition Metal Complexes Containing Metal-Ligand Multiple Bonds, Acc. Chem. Res., 2006, 39, 813–821.
2. (a) O. T. Summerscales and J. C. Gordon, Complexes Containing Multiple Bonding Interactions between Lanthanoid Elements and Main-Group Fragments, RSC Adv., 2013, 3, 6682–6692; (b) E. Lu, J. Chu and Y. Chen, Scandium Terminal Imido Chemistry, Acc. Chem. Res., 2018, 51, 557–566; (c) D. Schädle and R. Anwander, Rare-Earth Metal and Actinide Organoimide Chemistry, Chem. Soc. Rev., 2019, 48, 5752–5805.
3. G. R. Giesbrecht and J. C. Gordon, Lanthanide Alkylidene and Imido Complexes, Dalton Trans., 2004, 33, 2387–2393.
4. A. A. Trifonov, M. N. Bochkarev, H. Schumann and J. Loebel, Reduction of Azobenzene by Naphthaleneytterbium: A Tetranuclear Ytterbium(III) Complex Combining 1,2-Diphenylhydrazido(2-) and Phenylimido Ligands, Angew. Chem., Int. Ed. Engl., 1991, 30, 1149–1151.
5. H.-S. Chan, H.-W. Li and Z. Xie, Synthesis and Structural Characterization of Imido-Lanthanide Complexes with a Metal-Nitrogen Multiple Bond, Chem. Commun., 2002, 38, 652–653.
6. J. C. Gordon, G. R. Giesbrecht, D. L. Clark, P. J. Hay, D. W. Keogh, R. Poli, B. L. Scott and J. G. Watkin, The First Example of a μ₂-Imido Functionality Bound to a Lanthanide Metal Center: X-Ray Crystal Structure and DFT Study of [(μ-ArN)Sm(μ-NHAr)(μ-Me)AlMe₂]₂ (Ar = 2,6-ⁱPr₂C₆H₃), Organometallics, 2002, 21, 4726–4734.
7. D. J. Beetstra, A. Meetsma, B. Hessen and J. H. Teuben, (Cyclopentadienylamine) Scandium (2,3-Dimethyl-1,3-Butadiene): A 1,3-Diene Complex of Scandium with Sc(I)-and Sc(III)-Like Reactivity, Organometallics, 2003, 22, 4372–4374.
8. (a) T. Chu, W. E. Piers, J. L. Dutton and M. Parvez, Synthesis and Reactivity of a Terminal Scandium Imido Complex, Organometallics, 2013, 32, 1159–1165; (b) E. A. Patrick, Y. Yang, W. E. Piers, L. Maron and B. S. Gelfand, A Monoanionic Pentadentate Ligand Platform for Scandium-Pnictogen Multiple Bonds, Chem. Commun., 2021, 57, 8640–8643.
9. (a) E. Lu, Y. Li and Y. Chen, A Scandium Terminal Imido Complex: Synthesis, Structure and DFT Studies, Chem. Commun., 2010, 46, 4469–4471; (b) E. Lu, J. Chu, Y. Chen, M. V. Borzov and G. Li, Scandium Terminal Imido Complex Induced C-H Bond Selenation and Formation of an Sc-Se Bond, Chem. Commun., 2011, 47, 743–745.
10. (a) D. Schädle, M. Meermann-Zimmermann, C. Schädle, C. Maichle-Mössmer and R. Anwander, Rare–Earth Metal Complexes with Terminal Imido Ligands, Eur. J. Inorg. Chem., 2015, 2015, 1334–1339; (b) T. E. Rieser, D. Schädle, C. Maichle-Mössmer and R. Anwander, Terminal Dysprosium and Holmium Organoimides, Chem. Sci., 2024, 15, 3562–3570.
11. (a) L. A. Solola, A. V. Zabula, W. L. Dorfner, B. C. Manor, P. J. Carroll and E. J. Schelter, An Alkali Metal-Capped Cerium(IV) Imido Complex, J. Am. Chem. Soc., 2016, 138, 6928–6931; (b) L. A. Solola, A. V. Zabula, W. L. Dorfner, B. C. Manor, P. J. Carroll and E. J. Schelter, Cerium(IV) Imido Complexes: Structural, Computational, and Reactivity Studies, J. Am. Chem. Soc., 2017, 139, 2435–2442; (c) Y. Wang, J. Liang, C. Deng, R. Sun, P.-X. Fu, B.-W. Wang, S. Gao and W. Huang, Two-Electron Oxidations at a Single Cerium Center, J. Am. Chem. Soc., 2023, 145, 22466–22474.
12. R. Thim, H. M. Dietrich, M. Bonath, C. Maichle-Mössmer and R. Anwander, Pentamethylcyclopentadienyl-Supported Rare-Earth-Metal Benzyl, Amide, and Imide Complexes, Organometallics, 2018, 37, 2769–2777.
13. K. D. Conroy, W. E. Piers and M. Parvez, Nucleophilic Degradation of a β-Diketiminato Ancillary by a Transient Scandium Hydride Intermediate, Organometallics, 2009, 28, 6228–6233.
14. J. C. Berthet, P. Thuéry and M. Ephritikhine, Polyimido Clusters of Neodymium and Uranium, Including a Cluster with an M₆(μ₃-N)₈ Core, Wiley Online Library, 2008.
15. (a) D. Cui, O. Tardif and Z. Hou, Tetranuclear Rare Earth Metal Polyhydrido Complexes Composed of “(C₅Me₄SiMe₃)LnH₂” Units. Unique Reactivities toward Unsaturated C-C, C-N, and C-O Bonds, J. Am. Chem. Soc., 2004, 126, 1312–1313; (b) C.-L. Pan, W. Chen, S. Song, H. Zhang and X. Li, Stabilization of Imidosamarium(III) Cubane by Amidinates, Inorg. Chem., 2009, 48, 6344–6346.
16. (a) Z.-J. Lv, Z. Huang, J. Shen, W.-X. Zhang and Z. Xi, Well-Defined Scandacyclopropenes: Synthesis, Structure, and Reactivity, J. Am. Chem. Soc., 2019, 141, 20547–20555; (b) Z.-J. Lv, Z. Huang, W.-X. Zhang and Z. Xi, Scandium-Promoted Direct Conversion of Dinitrogen into Hydrazine Derivatives Via N–C Bond Formation, J. Am. Chem. Soc., 2019, 141, 8773–8777; (c) M. Zhu, Z. Chai, Z.-J. Lv, T. Li, W. Liu, J. Wei and W.-X. Zhang, Selective Cleavage of the Strong or Weak C-C Bonds in Biphenylene Enabled by Rare-Earth Metals, J. Am. Chem. Soc., 2023, 145, 6633–6638.
17. X. Chen, H. Xu, Z. Xi and J. Wei, A (C₅Me₅)/Enolamido Lutetium Dinitrogen Complex Promoted by Deprotonation on the Amidinate Ligand, Chem. Commun., 2024, 60, 12750–12753.
18. (a) J. Scott, F. Basuli, A. R. Fout, J. C. Huffman and D. J. Mindiola, Evidence for the Existence of a Terminal Imidoscandium Compound: Intermolecular C-H Activation and Complexation Reactions with the Transient Sc=NAr Species, Angew. Chem., Int. Ed., 2008, 47, 8502–8505; (b) J. Chu, E. Lu, Z. Liu, Y. Chen, X. Leng and H. Song, Reactivity of a Scandium Terminal Imido Complex Towards Unsaturated Substrates, Angew. Chem., Int. Ed., 2011, 33, 7677–7680; (c) W. Rong, J. Cheng, Z. Mou, H. Xie and D. Cui, Facile Preparation of a Scandium Terminal Imido Complex Supported by a Phosphazene Ligand, Organometallics, 2013, 32, 5523–5529.
19. (a) B. F. Wicker, H. Fan, A. K. Hickey, M. G. Crestani, J. Scott, M. Pink and D. J. Mindiola, Evidence for the Existence of Terminal Scandium Imidos: Mechanistic Studies Involving Imido-Scandium Bond Formation and C-H Activation Reactions, J. Am. Chem. Soc., 2012, 134, 20081–20096; (b) B. A. Clough, S. Mellino, E. Clot and P. Mountford, New Scandium Borylimido Chemistry: Synthesis, Bonding, and Reactivity, J. Am. Chem. Soc., 2017, 139, 11165–11183; (c) T. Cheisson, K. D. Kersey, N. Mahieu, A. McSkimming, M. R. Gau, P. J. Carroll and E. J. Schelter, Multiple Bonding in Lanthanides and Actinides: Direct Comparison of Covalency in Thorium(IV)- and Cerium(IV)-Imido Complexes, J. Am. Chem. Soc., 2019, 141, 9185–9190; (d) H. H. Wilson, X. Yu, T. Cheisson, P. W. Smith, P. Pandey, P. J. Carroll, S. G. Minasian, J. Autschbach and E. J. Schelter, Synthesis and Characterization of a Bridging Cerium(IV) Nitride Complex, J. Am. Chem. Soc., 2023, 145, 781–786.
20. (a) J. Chu, E. Lu, Y. Chen and X. Leng, Reversible Addition of the Si-H Bond of Phenylsilane to the Sc=N Bond of a Scandium Terminal Imido Complex, Organometallics, 2013, 32, 1137–1140; (b) J. Chu, C. E. Kefalidis, L. Maron, X. Leng and Y. Chen, Chameleon Behavior of a Newly Synthesized Scandium Nitrilimine Derivative, J. Am. Chem. Soc., 2013, 135, 8165–8168; (c) J. Chu, X. Han, C. E. Kefalidis, J. Zhou, L. Maron, X. Leng and Y. Chen, Lewis Acid Triggered Reactivity of a Lewis Base Stabilized Scandium-Terminal Imido Complex: C-H Bond Activation, Cycloaddition, and Dehydrofluorination, J. Am. Chem. Soc., 2014, 136, 10894–10897; (d) X. Han, L. Xiang, C. A. Lamsfus, W. Mao, E. Lu, L. Maron, X. Leng and Y. Chen, Synthesis and Reactivity of a Scandium Terminal Hydride: H₂ Activation by a Scandium Terminal Imido Complex, Chem. – Eur. J., 2017, 23, 14728–14732; (e) J. Chu, C. Wang, L. Xiang, X. Leng and Y. Chen, Reactivity of Scandium Terminal Imido Complex toward Boranes: C(sp³)-H Bond Borylation and B-O Bond Cleavage, Organometallics, 2017, 36, 4620–4625; (f) R. Thim, D. Schädle, C. Maichle-Mössmer and R. Anwander, Rare-Earth Metal Diimide Complexes via Alkylaluminate Templating, Including a Ceric Derivative, Chem. – Eur. J., 2019, 25, 507–511.
21. T. E. Rieser, R. Thim-Spöring, D. Schädle, P. Sirsch, R. Litlabø, K. W. Törnroos, C. Maichle-Mössmer and R. Anwander, Open-Shell Early Lanthanide Terminal Imides, J. Am. Chem. Soc., 2022, 144, 4102–4113.
22. (a) D. A. Kissounko, A. Epshteyn, J. C. Fettinger and L. R. Sita, Chemoselective Deprotonations of a Cationic Zirconium Primary Amido Complex to Either a Neutral Zirconium Terminal Imido or a Noninterconverting Tautomer, Organometallics, 2006, 25, 1076–1078; (b) B. L. Yonke, A. J. Keane, P. Y. Zavalij and L. R. Sita, Mononuclear Tantalum(IV,d¹) Imido Complexes Supported by the Monocyclopentadienyl, Amidinate and Guanidinate Ligand Sets as Models to Explore Dinitrogen Fixation by “End-on-Bridged” Dinuclear {[Ta(IV,d¹)]}₂(μ,-η¹:η¹-N₂) Complexes, Organometallics, 2012, 31, 345–355.

---

Footnotes

† Electronic supplementary information (ESI) available. CCDC 2385556, 2410804 and 2385558–2385564. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d5qi00390c

‡ Tianyu Li and Dajiang Huang contributed equally to this paper.

---