Open AccessCCS ChemistryRESEARCH ARTICLE1 Nov 2021Rare-Earth Metal Boroxide with Formal Triple Metal–Oxygen Orbital Interaction: Synthesis from B(C6F5)3·H2O and Radical-Anion Ligated Rare-Earth Metal Amides Haihan Yan†, Botao Wu†, Xiao-Kun Zhao, Chao Yu, Junnian Wei, Han-Shi Hu, Wen-Xiong Zhang and Zhenfeng Xi Haihan Yan† Beijing National Laboratory for Molecular Sciences, MOE Key Laboratory of Bioorganic Chemistry and Molecular Engineering, College of Chemistry, Peking University, Beijing 100871 †H. Yan and B. Wu contributed equally to this work.Google Scholar More articles by this author , Botao Wu† Beijing National Laboratory for Molecular Sciences, MOE Key Laboratory of Bioorganic Chemistry and Molecular Engineering, College of Chemistry, Peking University, Beijing 100871 †H. Yan and B. Wu contributed equally to this work.Google Scholar More articles by this author , Xiao-Kun Zhao Department of Chemistry, Key Laboratory of Organic Optoelectronics and Molecular Engineering of Ministry of Education, Tsinghua University, Beijing 100084 Google Scholar More articles by this author , Chao Yu Beijing National Laboratory for Molecular Sciences, MOE Key Laboratory of Bioorganic Chemistry and Molecular Engineering, College of Chemistry, Peking University, Beijing 100871 Google Scholar More articles by this author , Junnian Wei Beijing National Laboratory for Molecular Sciences, MOE Key Laboratory of Bioorganic Chemistry and Molecular Engineering, College of Chemistry, Peking University, Beijing 100871 Google Scholar More articles by this author , Han-Shi Hu Department of Chemistry, Key Laboratory of Organic Optoelectronics and Molecular Engineering of Ministry of Education, Tsinghua University, Beijing 100084 Google Scholar More articles by this author , Wen-Xiong Zhang *Corresponding author: E-mail Address: [email protected] Beijing National Laboratory for Molecular Sciences, MOE Key Laboratory of Bioorganic Chemistry and Molecular Engineering, College of Chemistry, Peking University, Beijing 100871 Google Scholar More articles by this author and Zhenfeng Xi Beijing National Laboratory for Molecular Sciences, MOE Key Laboratory of Bioorganic Chemistry and Molecular Engineering, College of Chemistry, Peking University, Beijing 100871 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.020.202000587 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail Rare-earth metal boroxides are not only rare and of great synthetic challenge but also important theoretically due to their unrevealed bonding characteristics. In this paper, we report the synthesis, characterization, and bonding characteristics of the 4f-block metal boroxide complexes MesDAD2 LuOB(C6F5)X ( 4-Lu, X = C6F5; 5-Lu, X = NC13H10) with two chelating radical-anionic 1,4-diazabutadiene (DAD) ligands. The synthesis of 4-Lu or 5-Lu involved three steps, during which the radical-anionic DAD ligands could be kept stable: (1) the reaction of LuCl3, MesDAD, and potassium graphite (KC8) afforded the diradical rare-earth metal chloride MesDAD2 LuCl· tetrahydrofuran (THF) ( 1-Lu); (2) the salt metathesis of 1-Lu with two alkali metal amides gave lutetium bis(diazabutadiene) amides 2-Lu or 3-Lu, respectively; and (3) 2-Lu or 3-Lu reacted with B(C6F5)3·H2O, resulting in the formation of rare-earth metal boroxides 4-Lu or 5-Lu. The radical-anionic nature of DAD ligands in these complexes were proven by their bond lengths and coordination modes in single-crystal structures. DFT calculations clearly show the bonding of 4-Lu can be divided into three major types of Lu–O orbital interactions, which can be categorized as σ-donation, π-donation, and d-π-conjugation. The triple Lu–O orbital interactions cooperatively make a Lu–O single bond and contribute to the linear Lu–O–B structure. Download figure Download PowerPoint Introduction The elucidation of the electronic structure of organometallic species is of great importance in providing insights into the nature of the chemical bond, and helping us discover structurally new molecules.1–5 Main group or transition-metal boroxides are well regarded as electron deficient analogues of metal alkoxides and aryloxides;6 however, the bonding characters in these boroxide complexes, especially linear ones, have not been studied in detail. Generally, one of the lone pair electrons on oxygen in metal boroxides can donate to suitable orbitals on the metal center, while the other pair is favored to donate to the empty 2p-orbital on boron, leading to a formal B–O double bond (Scheme 1a, A). The donation to boron has an important contribution to the B–O bond, while the contribution of the donation of oxygen electron density to the metal center in metal boroxides is weak and often omitted. As far as we are aware, the formal triple M–O orbital interaction with the significant contribution to M–O bond in metal boroxides remains elusive (Scheme 1a, B). In contrast to the main group or transition-metal boroxides, there is only one report of a rare-earth metal boroxide, Sc boroxide, which is prepared by reduction of B(C6F5)3·H2O with a formal Sc(I) complex.7 Thus, the construction of rare-earth metal boroxides with the formal triple M–O orbital interaction is an attractive and challenging area from both an experimental and theoretical perspective. Scheme 1 | (a–c) Metal boroxides and rare-earth metal–amide complexes with chelating radical ligands. Download figure Download PowerPoint Notably, rare-earth metal–amide complexes are one of the most common and important species in rare-earth organometallic chemistry and coordination chemistry.8–18 Benefiting from the strong Coulombic interaction as well as the prominent covalent interaction between rare-earth metal centers and amide ligands, rare-earth metal–amide complexes are traditionally stable, and can be combined with many types of auxiliary ligands to tune their structures and properties. However, the compatibility between amide ligands and chelating radical-anionic ligands in rare-earth complexes is poor. In fact, making rare-earth metal–amide complexes with chelating radical-anionic ligands generally suffers from (Scheme 1b): (1) the mismatch between soft chelating radicals and hard rare-earth metal centers or hard amide ligands, which makes the ligand redistribution between radical-anionic ligands and amide ligands occur easily,19 and (2) the redox reactivity of radical anions weakening their stability.20–22 Although some rare-earth metal complexes contain radical-anionic ligands, such as 1,4-diazabutadienes (DADs),19–29 iminopyridines (IPys),30–33 bipyridines (Bipys),34–36 and α-diones,37–39 there are only three examples of rare-earth metal complexes bearing both radical ligands and chelating amide ligands, such as bis(2-isopropylaminophenyl)amide, monoanionic adduct of Ipy, or dianionic DAD-type ligands.40–42 Simple amides have yet to be introduced into this kind of rare-earth metal complex. Herein, we report the synthesis and structures of rare-earth metal amides 2 and 3, which are supported by two moderately bulky radical-anionic MesDAD ligands. Further protonolysis reaction of 2-Lu and 3-Lu with B(C6F5)3·H2O afforded the first 4f-block metal boroxides 4-Lu and 5-Lu, in which the Lu–O–B motif is linear (Scheme 1c). The bonding properties of 4-Lu were studied in detail by theoretical calculations, and three types of orbital interactions which mainly contribute to this linear Lu–O–B structure were found. Experimental Methods Synthesis of complex 1 Anhydrous rare-earth metal chlorides [RECl3 (0.50 mmol)] and MesDAD (292 mg, 1.00 mmol) were added to a 50 mL round-bottom flask. THF (20 mL) was added and the mixture was stirred at room temperature. KC8 (135 mg, 1.00 mmol) was gradually added, and the solution immediately turned dark red. After stirring at room temperature for 2 h, solvent was removed under vacuum, and the residue was extracted with toluene (15 mL) and filtered. The filtrate was evaporated under vacuum to give 1-Sc or 1-Lu as yellow or orange powder. Synthesis of complexes 2 and 3 Anhydrous RECl3 (0.50 mmol) and MesDAD (292 mg, 1.00 mmol) were added to a 50 mL round-bottom flask. THF (20 mL) was added and the mixture was stirred at room temperature. KC8 (135 mg, 1.00 mmol) was gradually added, and the solution immediately turned dark red. After stirring at room temperature for 2 h, lithium amide (0.50 mmol) was added and the mixture was stirred for 1 h. After removing solvent under vacuum, the residue was extracted with n-hexane (45 mL) and filtered. The filtrate was evaporated under vacuum to give 2 or 3 as dark red to black powder. Synthesis of complexes 4-Lu and 5-Lu Freshly prepared 2-Lu or 3-Lu (0.10 mmol) was dissolved in toluene (3 mL) in a 25 mL round-bottom flask. B(C6F5)3·H2O (53 mg, 0.10 mmol) dissolved in toluene (5 mL) was gradually added. After stirring at room temperature for 1 h, solvent was removed under vacumm, and the residue was extracted with n-hexane (15 mL) and filtered. The filtrate was evaporated under vacuum to give the product 4-Lu and 5-Lu as dark red powder. Further experiment details and characterization data are presented in the Supporting Information. Results and Discussion The reaction of anhydrous rare-earth metal chloride (RE = Sc and Lu), MesDAD, and KC8 in THF yielded rare-earth bis(diazabutadiene) chloride 1 (RE = Sc and Lu) (Scheme 2). The scandium complex 1-Sc easily decomposed during the crystallizing process, but the lutetium complex 1-Lu could be recrystallized in Et2O/THF to give single crystals suitable for X-ray diffraction analysis. The structure of 1-Lu (Figure 1 and Supporting Information Figure S1) consists of a distorted octahedral geometry, in which the lutetium center is coordinated with four nitrogen atoms of two MesDAD, one chloride, and one THF molecule. The lutetium atom is nearly coplanar with the N–C–C–N skeleton of DAD ligand, with an average Lu–N bond length of 2.374 Å. The four N–C bond lengths and two C–C bond lengths (Table 1) in DAD ligands are all comparable with the reported DAD radical anions.24,43,44 Scheme 2 | Synthesis of 2 and 3. Download figure Download PowerPoint Table 1 | Lengths of N–C, C–C, and Lu–NDAD Bonds in DAD-Lu Skeletons for Selected Complexes Complex N–C (Å) C–C (Å) Lu–NDAD (Å) 1-Lu 1.334(3), 1.331(3) 1.395(3), 1.396(3) 2.4028(16), 2.3369(16) 1.342(3), 1.339(3) 2.3894(17), 2.3703 (18) 2-Lu 1.328(3), 1.335(3) 1.398(3), 1.398(3) 2.3316(18), 2.3540(18) 1.337(3), 1.342(3) 2.3402(17), 2.3255(17) 3-Lu 1.327(3), 1.342(3) 1.403(3), 1.396(3) 2.3301(18), 2.3284(18) 1.345(3), 1.334(3) 2.3032(18), 2.3356(18) 4-Lu 1.332(5), 1.330(5) 1.406(6), 1.408(6) 2.318(3), 2.290(3) 1.342(5), 1.328(5) 2.313(3), 2.316(3) 5-Lu 1.340(3), 1.335(3) 1.390(4), 1.396(4) 2.2998(19), 2.320(2) 1.326(3), 1.336(3) 2.2866(19), 2.342(2) The isolated 1-Lu reacted with lithium amide LiNHDipp (Dipp = 2,6-diisopropylphenyl), resulting in the formation of 2-Lu. After many trials, we found in situ generation of 1-Lu from LuCl3, MesDAD, and KC8 was more convenient than isolating 1-Lu. Thus, an efficient in situ route was established to afford 2-Lu (Scheme 2). Replacing LuCl3 with ScCl3 provides 2-Sc. When lithium dihydroacridinide (LiNC13H10) was utilized, the corresponding 3-Sc and 3-Lu could be obtained. Figure 1 | Molecular structures of 1-Lu (left), 2-Lu (middle), and 3-Lu (right) with thermal ellipsoids at 35% probability. H atoms and non-coordinating solvents are omitted for clarity. Download figure Download PowerPoint All the molecular structures of 2-Sc,Lu and 3-Sc,Lu were confirmed by single-crystal X-ray diffraction analysis (see Figures S3–S6 and Supporting Information for details). 2-Sc and 2-Lu (Figure 1) were both crystallized in the monoclinic space group P21/c, with a distorted square pyramidal coordination sphere. The amide ligand is located on the axial direction of the square pyramidal. For 2-Lu, the average Lu–NDAD bond length is 2.338 Å (Table 1 and see Tables S7–S9 and Supporting Information for more details), which is shorter than the Lu–NDAD bonds in hexa-coordinated 1-Lu (avg. 2.375 Å). The crystal structure of 3 is similar to 2. The RE–NDAD distances are slightly shorter in 3-Sc or 3-Lu compared with the corresponding 2-Sc or 2-Lu (see Tables S4–S15). These bond lengths on the DAD skeletons still correspond to the DAD ligands as radical anions. It should be noted that despite reports of coordination of two or more sterically favored chelating radical ligands, such as IPy or Bipy, with a rare-earth metal center,36,40 similar reports for DAD ligands remain rare,42 especially with Sc and Lu. The ligand redistribution of DAD ligands in this salt metathesis reaction inhibits the synthesis of rare-earth metal–amide complexes with two chelating radical-anionic DAD ligands. The generation of 2-Sc,Lu and 3-Sc,Lu can be attributed to two aspects: (1) the moderately bulky mesityl groups on DAD ligands suppress the ligand redistribution process, and (2) the nucleophilic attack of a bulky amide on 1-Sc,Lu displaces THF, which releases the tension in the crowded rare-earth metal center, and strengthens the RE–NDAD bonds by lowering the coordination number. As far as we are aware, 2-Sc,Lu and 3-Sc,Lu are the first neutral rare-earth metal–amide complexes with multiple chelating radical-anionic ligands. With these rare-earth bis(diazabutadiene) amides in hand, we aim to explore the reactivity of amide ligands while maintaining stability of the DAD ligands. The protonolysis reaction of amide ligands in 2-Lu or 3-Lu with different proton sources was tested. Reaction with water, alcohols, or phenols did not produce isolable products, which might be because the side reaction of a proton with a DAD ligand would result in the degeneration of the products. The reaction of 2-Lu or 3-Lu with boronic acids, such as Ar2BOH, Mes2BOH, and (C6F5)2BOH easily generated a messy mixture. However, a sterically bulky proton source, B(C6F5)3·H2O, with moderate acidity reacted well. The reaction of 2-Lu with B(C6F5)3·H2O generated bis(diazabutadiene) lutetium boroxide 4-Lu by the protonolysis of −NHDipp and the elimination of C6F5H, whereas the reaction of 3-Lu with B(C6F5)3·H2O gave 5-Lu, in which the amido group was transferred to the boron center (Scheme 3). To the best of our knowledge, there is no report of such metal amido-boroxide complexes. The reactivity of proton in B(C6F5)3·H2O toward Lu–N amides is suitable, while it is low toward the sterically bulkier and less basic DAD radical anions. Thus, the selective protonolysis reaction with B(C6F5)3·H2O could be achieved. Scheme 3 | Reaction of 2-Lu and 3-Lu with B(C6F5)3·H2O. Download figure Download PowerPoint The single-crystal structures of 4-Lu and 5-Lu are shown in Figure 2 and Supporting Information Figures S7 and S8. The bond lengths of DAD-Lu skeleton of 4-Lu and 5-Lu changed little compared with 2-Lu or 3-Lu (Table 1 and Supporting Information Tables S16 and S21). The B–O bond length in 4-Lu [1.297(5) Å] is close to those reported data for Ti, Zr, and U complexes45–51 as well as some B=O double bonds in oxoboranes,52 which indicates that the O→B π donation is strong. The Lu–O bond length [2.066(3) Å] is slightly longer than those found in lutetium alkoxides or aryloxides (2.00–2.06 Å). The B–O–Lu bond angle (175.5(3)°) is significantly larger than the B–O–Sc bond angle (∼150°) in the scandium complex LSc[OB(C6F5)2]2 [L = (Et2NCH2CH2NCMe)2CH].7 Thus, the B–O–Lu motif in 4-Lu can be assigned a linear structure. For 5-Lu, the B–O bond [1.321(3) Å] becomes longer, and the Lu–O bond [2.0561(17) Å] is slightly shorter than those found in 4-Lu. The B–O–Lu bond angle [168.93(17)°] is also smaller than that found in 4-Lu and significantly larger than the B–O–Sc bond angle (∼150°) in the scandium boroxide.7 This shows that the B–N bond can compensate for the low charge density on the boron atom, which weakens the O→B π donation and further strengthens the O→Lu π donation. As far as we are aware, 4-Lu and 5-Lu are the first 4f-block metal boroxides. Figure 2 | Molecular structures of 4-Lu (left) and 5-Lu (right) with thermal ellipsoids at 35% probability. H atoms are omitted for clarity. Selected bond lengths (Å) and bond angles (°) for 4-Lu: B1–O1 1.297(5); O1–Lu1 2.066(3); B1–O1–Lu1 175.5(3), 5-Lu: B1–O1 1.321(3); O1–Lu1 2.0561(17); B1–O1–Lu1 168.93(17). Download figure Download PowerPoint To understand the nature of the linear-like Lu–O–B structure more clearly, theoretical calculations were performed using 4-Lu as an example. The optimized bond distances and bond angles fit well with the single-crystal structure of 4-Lu. Natural-localized molecular orbital (NLMO) from natural bond orbital (NBO) analysis was used for the Lu–O–B structure. The NLMO diagram for Lu–O–B bonding of 4-Lu is exhibited in Figure 3a. The bonding mode among three atoms can be indicated by four NLMOs: σB–O, πB–O, nO(σ), and nO(2p). The B=O bond contains 56.4% σ-bonding (σB–O) and 43.6% π-bonding [37.8% from πB–O and 5.8% from nO(2p)] orbitals. The atomic hybrid-orbital composition based on NLMO shows that the πB–O mainly demonstrates a 2p–2p π-bonding between boron and oxygen, while the σB–O illustrates the orbital hybridization of boron and oxygen. As shown in Figure 3b, the bonding orbital of O is sp0.48 hybridized and B is sp2.54 hybridized. In addition, the B=O bond is strongly polarized, which is evinced by the large difference in orbital components of oxygen and boron (80–84.2% vs 12.5–18.7%), and the dipole moment direction (B→O). For the bonding between oxygen and lutetium, we found the contributions of three NLMO components, which are sorted in descending order: nO(2p) (39.4%), nO(σ) (34.3%), and πB–O (26.3%), where nO(2p) and nO(σ) mainly represent lone pairs of oxygen, and πB–O is mainly the π-orbital of B=O bond. A small amount of lutetium (2.2–3.3%) is involved in πB–O, nO(2p), and nO(σ), reflecting their contribution to Lu–O bonding. To be specific, the orbital interaction between oxygen and lutetium contains the 2p(O)→5d(Lu) π-coordination in nO(2p), the σ-coordination from oxygen to lutetium via hybridized orbitals in nO(σ) (Figure 3b and sp2.1 for O; sp0.3d2.41f0.3 for Lu) and the weak p-π conjugation between 5d(Lu) and π(B=O), respectively. Unfortunately, there is no report about bonding analysis on a similar level of theory for traditional lutetium alkoxides. To make a rational comparison between lutetium boroxides and alkoxides, NLMO analysis was conducted for a reported lutetium alkoxide complex [(Me3Si)2NC(NiPr)2]2LuOtBu,53 in which the Lu–O interactions were only found as pure O→Lu σ and π donations (see Figure S10 and Supporting Information for details). These results clearly show the electron delocalization effect of boron in 4-Lu. Interestingly, the diradical characteristics of these complexes are only reflected on the two (DAD)•− ligands. Figure 3 | (a) Selected Lu–O–B fragment containing NLMO diagram of 4-Lu. (b) Selected atomic hybrid-orbit composition for NLMO in 4-Lu (isovalue = 0.05) Download figure Download PowerPoint Energy decomposition analysis combined with natural orbitals for chemical valence (EDA-NOCV) was also conducted for better interpretation of the triple orbital interaction between lutetium of [(DAD)2Lu]+ and oxygen of boroxide ligand. As shown in Table 2, the contribution of the orbital interaction term ΔEorb (29.1%) in 4-Lu is weaker than the electrostatic attraction ΔEelstat (63.5%). The dispersion forces provide the remaining 7.4% to the total attraction, which is non-negligible. So the interaction between [(DAD)2Lu]+ and boroxide is primarily ionic, but the orbital interactions are still important especially for the formation of the linear Lu–O–B structure. Table 2 | EDA-NOCV Results of 4-Lua Term Description Energy (kcal/mol) ΔEint −152.97 ΔEdisp −19.93 (7.4%) ΔEPauli 115.11 ΔEelstatb −170.23 (63.5%) ΔEorbb −77.92 (29.1%) ΔEorb(1)c (DAD)2Lu+←O−(lp)σ-donation −22.4 (28.8%) ΔEorb(2)c (DAD)2Lu+←OB−(lp+π)π-donation −18.49 (23.7%) ΔEorb(3)c (DAD)2Lu+←OB−(π)d-π-conjugation −12.2(15.7%) ΔErestc −20.33 (26.1%) aThe interacting fragments are (DAD)2Lu+ and boroxide anion. bThe values in parentheses give the percentage contribution to the total attractive interactions ΔEelstat + ΔEorb + ΔEdisp. cThe values in parentheses give the percentage contribution to the total orbital interactions ΔEEorb. The most important information about the orbital interactions between the [(DAD)2Lu]+ fragment and the boroxide ligand comes from the breakdown of the ΔEorb term into the pairwise orbital contributions. Four major orbital interactions can be identified by inspecting the deformation densities Δρ(1)∼Δρ(3) (Figure 4), which show the sum of the α and β contributions to the orbital interactions and simultaneously associate with ΔEorb(1)∼ΔEorb(3) contributing 68.2% of the total ΔEorb (Table 2). The strongest orbital interaction Δρ(1), which is associated with ΔEorb(1) (i.e., −22.4 kcal/mol) contributing 28.8% of the total ΔEorb, comes mainly from the σ-donation from the lone pair of oxygen in boroxide ligand to the 5d orbital of Lu in [(DAD)2Lu]+. The second contribution (ΔEorb(2) = −18.5 kcal/mol) mainly corresponds to the in-plane π-donation from the 2p(O) in the boroxide fragment to the 5d orbital of Lu in [(DAD)2Lu]+, as well as the conjugation between B=O π-bond and the 5d orbital of Lu, which could not be completely separated. The third contribution [ΔEorb(2) = −12.2 kcal/mol] mainly arises from the out-of-plane d-π conjugation between the B=O π-bond and the 5d orbital of Lu-center. From the sum of these three energy values (−53.09 kcal/mol), the covalent (or dative) component of Lu–O interaction can be recognized as a single bond, in which all three orbital interactions have moderate contributions. This result is consistent with the NLMO calculation result. In addition, the ΔErest is a bit large (26.1%), which indicates that there are more secondary orbital interactions in this system. Figure 4 | Shape of the charge deformation densities (1)–(3) associated with the orbital interactions ΔEorb(1)–(3) for 4-Lu (isovalue = 0.001). The eigenvalues (υ) of the fragment orbitals give the size of the charge migration. The color code for the charge transfer is red→blue. Gray, carbon; white, hydrogen; blue, nitrogen; light green, fluorine; red, oxygen; orange (big), lutetium; orange (small), boron. Download figure Download PowerPoint Furthermore, the Mayer bond order of Lu–O bond is 0.78, which also indicates the strong coordination effect between O and Lu. Thus, the linear Lu–O–B structure in 4-Lu is formed mainly because of two aspects: (1) the restricted geometry with the –C6F5 and –Mes as the hindered group; (2) the triple orbital interaction between O and Lu, which can only be strengthened under the linear conformation. In contrast, the Mayer bond order of B–O is 1.59, which is slightly low for a B=O double bond. This is due to the πB–O orbital involved by 5d(Lu) according to the above discussion. In other words, the [(DAD)2Lu]+ fragment has a certain degree of electron-withdrawing conjugated effect on the B=O bond. In addition, we also propose the mechanism for this reaction. Generally, as shown in Scheme 4, for both 2-Lu and 3-Lu, the first C6F5H could be released through a concerted or stepwise proton transfer and elimination process to generate Int-1 with O–Lu bond and B–N bond. Then, in the second elimination step for 2-Lu, the liberation of sterically bulky –NHDipp is more favored than –C6F5, thus 4-Lu could be obtained as the final product. However, for 3-Lu, the steric difference between –NC13H10 and –C6F5 is not large, so the cleavage of the weak B–C bond rather than the stronger B–N bond is favored resulting in the elimination of another C6F5H with the formation of 5-Lu. Scheme 4 | Possible pathways for the reaction of 2-Lu and 3-Lu with B(C6F5)3·H2O. Download figure Download PowerPoint Conclusion The synthesis, characterization, and bonding characteristics of the 4f-block metal boroxide complexes MesDAD2 LuOB(C6F5)X ( 4-Lu, X = C6F5; 5-Lu, X = NC13H10) with two chelating radical-anionic DAD ligands were achieved. Complexes 4-Lu or 5-Lu were prepared by the reaction of B(C6F5)3·H2O with lutetium bis(diazabutadiene) amides 2-Lu or 3-Lu, which were generated from the salt metathesis of the diradical rare-earth metal chloride 1-Lu with LiNHDipp or LiNC13H10, respectively. The radical-anionic MesDAD ligands in those complexes remained stable during such transformations, proving that ligand redistribution and redox process were suppressed by the moderately bulky mesityl substituents. The structure of 4-Lu was studied by DFT calculations. Three major orbital interactions were found in the linear Lu–O–B structure, which clearly depicted the bonding type and their contributions in metal boroxides. Supporting Information Supporting Information is available and includes additional experimental details and characterization data for complexes 1– 5, X-ray crystallographic data of 1-Lu, 2-Sc, 3-Lu, 3-Sc, 4-Lu, and 5-Lu, and DFT calculation details. Conflict of Interest There is no conflict of interest to report. Funding Information This work was supported by the Natural Science Foundation of China (nos. 21725201 and 21890721). Acknowledgments The DFT calculation was supported by High-performance Computing Platform of Peking University. The authors thank Prof. Yanfeng Dang from Tianjin University, Prof. Jun Zhu from Xiamen University, Mr. Chang-Su Cao from Tsinghua University, and Miss. Xianlu Cui from Nanjing Tech University for the discussions on theoretical calculations.