Stabilizing a three-center single-electron metal–metal bond in a fullerene cage

Abstract

Trimetallic carbide clusterfullerenes (TCCFs) encapsulating a quinary M₃C₂ cluster represent a special family of endohedral fullerenes with an open-shell electronic configuration. Herein, a novel TCCF based on a medium-sized rare earth metal, dysprosium (Dy), is synthesized for the first time. The molecular structure of Dy₃C₂@I_h(7)-C₈₀ determined by single crystal X-ray diffraction shows that the encapsulated Dy₃C₂ cluster adopts a bat ray configuration, in which the acetylide unit C₂ is elevated above the Dy₃ plane by ∼1.66 Å, while Dy–Dy distances are ∼3.4 Å. DFT computational analysis of the electronic structure reveals that the endohedral cluster has an unusual formal charge distribution of (Dy₃)⁸⁺(C₂)²⁻@C₈₀⁶⁻ and features an unprecedented three-center single-electron Dy–Dy–Dy bond, which has never been reported for lanthanide compounds. Moreover, this electronic structure is different from that of the analogous Sc₃C₂@I_h(7)-C₈₀ with a (Sc₃)⁹⁺(C₂)³⁻@C₈₀⁶⁻ charge distribution and no metal–metal bonding.

---

Introduction

Endohedral clusterfullerenes featuring encapsulation of an unstable metallic cluster within a fullerene cage have been attracting considerable interest over the past two decades due to the versatility of the types of encapsulated cluster and its charge transfer to the outer fullerene cage.^1–8 Soon after the discovery of Sc₃N@C₈₀ as the first endohedral clusterfullerene encapsulating a trimetallic nitride cluster in 1999,⁹ the first metal carbide clusterfullerene (CCF), Sc₂C₂@C₈₄, was identified by Shinohara and coworkers in 2001.¹⁰ The identification that a C₂ moiety can be encapsulated in the form of metal carbide revealed a competition between CCFs M₂C₂@C_2n−2 and genuine dimetallofullerenes (di-EMFs) M₂@C_2n in the hot carbon vapor during the fullerene formation. Importantly, CCFs and di-EMFs have different electronic structures and metal valence states. For Sc, Y, and heavy lanthanides, di-EMFs feature a (M₂)⁴⁺@C_2n⁴⁻ distribution with M–M bonding and formally divalent metal M²⁺,^11–14 while the charge distribution in dimetallic CCFs is (M₂)⁶⁺(C₂)²⁻@C_2n⁴⁻ with the more traditional M³⁺ state of rare-earth metals.^8,15,16 Thus, by accumulating a certain negative charge, the C₂ group has a paramount influence on the valence state of endohedral metal atoms.

The C₂ group in CCFs has a variable formal charge state adapting to the metal cluster composition, from (C₂)²⁻ in M₂C₂@C_2n to (C₂)⁶⁻ in Sc₄C₂@C₈₀.^17,18 In Sc₃C₂@C₈₀, the first trimetallic CCF (TCCF) and the only TCCF characterized by single-crystal X-ray diffraction (SC-XRD),¹⁹ the acetylide unit has a formal (C₂)³⁻ charge state.²⁰ A similar charge distribution is also proposed for Lu₃C₂@C₈₈.²¹ In mixed-metal M₂TiC₂@C₈₀ TCCFs (M = Sc, Lu, Dy), the acetylide unit adopts a (C₂)⁴⁻ state to balance the tetravalent Ti⁴⁺.^22,23 In these five TCCFs, rare-earth metals are present in their 3+ oxidation state and thus do not form metal–metal bonding interactions. Interestingly, three-center M–M bonding was predicted in Y₃@C₈₀ and Er₃@C₇₄ trimetallofullerenes (tri-EMFs) based on computational analysis, but structural characterization of such tri-EMFs is still lacking.^24,25

In this work, we aim at filling the gap in the knowledge about rare-earth TCCFs and synthesized a novel TCCF based on a medium-sized rare-earth metal, dysprosium (Dy) – Dy₃C₂@C₈₀. Astonishingly, SC-XRD and computational analysis of Dy₃C₂@C₈₀ revealed that, dramatically different from the analogous Sc₃C₂@C₈₀ based on the small-sized metal Sc, Dy₃C₂@C₈₀ exhibits an unusual (Dy₃)⁸⁺(C₂)²⁻@C₈₀⁶⁻ formal charge distribution and features a unique three-center single-electron Dy–Dy–Dy bond, which has never been reported in molecular lanthanide chemistry.^26–28

Results and discussion

Synthesis, isolation and molecular structure of Dy₃C₂@I_h(7)-C₈₀

Dy₃C₂@C₈₀ was synthesized by a modified Krätschmer–Huffman DC arc discharge method using a mixture of Dy₂O₃ and graphite (molar ratio of Dy : Ti : C = 1 : 1 : 15) as the raw material under 200 mbar He and 10 mbar N₂ gas.^29–32 Isolation of Dy₃C₂@C₈₀ was fulfilled by three-step HPLC (see ESI S1† for details). The purity of the isolated Dy₃C₂@C₈₀ was confirmed by laser desorption time-of-flight (LD-TOF) mass spectroscopic (MS) and HPLC analyses (Fig. 1a and b). The LD-TOF MS spectrum shows a dominant mass peak at m/z = 1471.5 with the isotopic distribution resembling the calculated one for Dy₃C₈₂. The HPLC profile with a single peak at a retention time of 53.5 min confirms the high purity of the isolated Dy₃C₈₂. Based on the integrated area of the corresponding HPLC peaks, the yield of Dy₃C₈₂ relative to Dy₃N@I_h(7)-C₈₀ is estimated to be 1 : 320 (see ESI Table S1†).

Fig. 1 Confirmation of the purity and determination of the molecular structure of Dy₃C₂@I_h-C₈₀. (a) LD-TOF mass spectrum of purified Dy₃C₂@I_h-C₈₀; the insets show the measured and calculated isotopic distributions of Dy₃C₈₂. (b) HPLC profile of fraction A-3-3 (4.6 × 250 mm Buckyprep column; toluene as an eluent; flow rate 1.0 ml min⁻¹; injection volume 1 ml; 25 °C). (c) The thermal ellipsoid drawing of Dy₃C₂@I_h(7)-C₈₀·2DPC. Only the major Dy₃C₂ site is shown. Solvent molecules and H atoms are omitted for clarity.

The molecular structure of Dy₃C₈₂ was established by SC-XRD using co-crystallization with a recently developed decapyrrylcorannulene (DPC) host.³³ Mixing a CS₂ solution of fullerene with a toluene solution of DPC afforded cocrystal Dy₃C₂@I_h(7)-C₈₀·2(DPC)·3(C₇H₈), and its SC-XRD analysis confirmed unambiguously that the isolated Dy₃C₈₂ is Dy₃C₂@I_h(7)-C₈₀ TCCF. The cocrystal falls into the monoclinic P2₁/c space group, its asymmetric unit cell is composed of four complete Dy₃C₂@I_h-C₈₀ fullerene molecules and four pairs of 2DPC. Fig. 1c illustrates the relative orientations of the Dy₃C₂@I_h(7)-C₈₀ and the DPC molecules in the cocrystal, in which only one orientation of the fullerene cage together with the major site of the Dy₃C₂ cluster is shown for clarity. The nearest DPC-cage distances are 3.322(11) and 3.341(12) Å, respectively, which are comparable to those of the cocrystals of DPC with other endohedral fullerenes.³³ It is noteworthy that so far only two TCCFs with a homometallic M₃C₂ cluster have been reported, Sc₃C₂@I_h(7)-C₈₀ (ref. 19 and 34–38) and Lu₃C₂@D₂(35)-C₈₈,²¹ both with rare-earth metals of relatively small ionic radius (R³⁺ is 0.75 Å for Sc and 0.86 Å for Lu³⁹), and only Sc₃C₂@I_h(7)-C₈₀ has been structurally determined by SC-XRD.¹⁹ Therefore, Dy₃C₂@I_h(7)-C₈₀ represents the first TCCF based on a medium-sized rare-earth metal (R(Dy³⁺) = 0.91 Å) with the molecular structure determined unambiguously by SC-XRD.

While the I_h(7)-C₈₀ cage is fully ordered in the crystal, Dy atoms are disordered over 14 positions, whose site occupancies range from 0.051(2) to 0.501(2) (see ESI Fig. S2b†). The further structural analysis is based on the major Dy1–Dy3 sites with the largest occupancies of 0.482(2), 0.501(2) and 0.429(3), respectively. Dy atoms are located near the junctions of one pentagon and two hexagons as shown in Fig. 2a, with the shortest Dy–C(cage) distances of 2.202(16), 2.253(12), and 2.254(9) Å, suggesting a strong metal–cage interaction. The Dy⋯Dy distances are 3.381(1), 3.361(2), and 3.389(1) Å, respectively, which are all shorter than Sc⋯Sc distances in Sc₃C₂@I_h(7)-C₈₀ (3.407(3), 3.712(3), and 3.989(3) Å)¹⁹ or Dy⋯Dy distances in Dy₃N@I_h(7)-C₈₀ (3.476(3), 3.541(3), and 3.564(3) Å).⁴⁰

Fig. 2 Configuration of the Dy₃C₂ cluster within Dy₃C₂@I_h(7)-C₈₀. (a) Position of the Dy₃C₂ cluster within Dy₃C₂@I_h(7)-C₈₀ (major Dy sites) with the closest fullerene fragments. (b) The geometric structure of the Dy₃C₂ cluster with bond lengths determined by single-crystal X-ray diffraction. The numbers above the carbon atoms indicate the vertical displacements of carbon atoms above the Dy₃ planes. Distances are in Å. C and Dy atoms are shown in cyan and orange, respectively.

In view of a considerable disorder of heavy Dy atoms, positions of endohedral carbons are not very well defined. The best refinement results were obtained for a bat ray configuration of the Dy₃C₂ cluster shown in Fig. 2. The center of the C₂ unit is elevated above the Dy1–Dy2–Dy3 plane by ∼1.66 Å. The bat ray configuration was also found in the SC-XRD study of Sc₃C₂@I_h(7)-C₈₀, but the elevation of the C₂ unit above the Sc₃ plane is considerably smaller (0.44 Å).¹⁹ Besides, the C–C distance within the Dy₃C₂ cluster is determined to be 1.134(22) Å, which is smaller than that within Sc₃C₂ (1.29(3) Å (ref. 19)). Overall, these results reveal the strong influence of the metal size on the geometry of the encapsulated M₃C₂ cluster within the TCCF. The rare earth metals with larger size push the C₂ unit further above the M₃ plane and at the same time afford shorter M⋯M distances.

DFT computational studies of M₃C₂@I_h(7)-C₈₀ (M = Dy, Sc)

The different shapes of the Dy₃C₂ and Sc₃C₂ clusters motivated us to conduct a DFT computational study (Priroda and Orca codes^41,42) of the molecular and electronic structure of M₃C₂@C₈₀ (M = Dy, Sc) molecules. In Sc₃C₂@C₈₀, earlier computations and SC-XRD studies revealed two quasi-isoenergetic structures of the Sc₃C₂ unit.^19,20 In the lower-energy bat ray configuration, the C₂ unit is almost parallel to the Sc₃ plane and is elevated above it (Fig. 3a). The trifoliate structure has a trigonal bipyramidal shape with the Sc₃ triangle in the base and two C atoms in the apexes. Using these structures as starting configurations for optimization of Dy₃C₂@C₈₀ showed that Dy₃C₂ also has a minimum for a trifoliate configuration albeit with high relative energy, but the bat ray configuration changed considerably during optimization by elevating the C₂ unit by 1.620 Å above the Dy₃ plane and giving a structure quite similar to that found by SC-XRD. Analogous results were obtained in computations of Y₃C₂@C₈₀ (see the ESI†). Thus, the larger ionic radii of Dy³⁺ and Y³⁺ do not afford enough place for the C₂ unit when it is too close to the M₃ plane. Importantly, this bat ray structure of the Dy₃C₂ cluster is 89 kJ mol⁻¹ more stable than the trifoliate configuration. Different positions of the C₂ group also result in substantial differences in the DFT-optimized Dy⋯Dy distances. For the bat ray structure, the predicted distances of 3.440, 3.382, and 3.408 Å agree well with the SC-XRD results, while the trifoliate configuration has longer Dy⋯Dy distances of 3.637, 3.651, and 3.939 Å. The transition state connecting these two cluster configurations is found at a relative energy of 140 kJ mol⁻¹. The bat ray structure with an elevated C₂ unit similar to the Dy₃C₂ cluster is also found to be an energy minimum for Sc₃C₂@C₈₀, but this configuration is 65 kJ mol⁻¹ less stable than the trifoliate and flattened bat-ray configurations. Thus, in agreement with the experimental results, our calculations confirm that the lowest energy configuration of the Dy₃C₂ cluster has relatively short Dy⋯Dy distances and a significantly elevated position of the C₂ unit, which is different from the optimal geometry of the Sc₃C₂ cluster.

Fig. 3 DFT computational study on the electronic structure of M₃C₂@C₈₀ (M = Dy, Sc). (a) DFT-optimized configurations of the Dy₃C₂ (top) and Sc₃C₂ (bottom) clusters in M₃C₂@C₈₀ and their relative energies. Dy is depicted in green, Sc is shown in magenta, and “TS” means transition state. (b) DFT-computed valence spin density distribution in Dy₃C₂@C₈₀. (c) Singly occupied MO (SOMO) in Dy₃C₂@C₈₀. (d) Kohn–Sham MO energy levels in the lowest energy configurations of Dy₃C₂@C₈₀ and Sc₃C₂@C₈₀ (red and pale blue lines denote the occupied and unoccupied components of the singly occupied MO, and dashed arrows highlight the gap between the SOMO components). Computations are performed with the PBE functional, def2-TZVPP basis set for C and Sc, and 4f-in-core ECP55MWB-II basis set of Dolg and coworkers for Dy.^43–46

Confinement of the Dy₃C₂ cluster within the I_h-C₈₀ cage not only changes the molecular structure in comparison to the Sc congener, but also results in a completely new bonding situation. Previous studies showed that the electronic configuration of Sc₃C₂@C₈₀ is best described as (Sc₃)⁹⁺(C₂)³⁻@C₈₀⁶⁻ with the spin density mainly localized on the C₂³⁻ unit and Sc atoms featuring their typical Sc³⁺ oxidation state.²⁰ Surprisingly, in the Dy analog the charge distribution changes to (Dy₃)⁸⁺(C₂)²⁻@C₈₀⁶⁻ (see ESI S5† for the analysis of the canonical and Pipek–Mezey localized orbitals of the cluster). Acetylide unit C₂ in Dy₃C₂@C₈₀ has negligible spin population, while the main part of spin density is shared between three Dy ions (Fig. 3b). This spin distribution reflects the shape of the SOMO, which has Dy–Dy–Dy bonding character and corresponds to the three-center single-electron bonding of all three Dy atoms (Fig. 3c) with main contributions from the 6s, 6p, and 5d atomic orbitals of Dy. The presence of the M–M bonding in medium-size rare-earth M₃C₂@C₈₀ TCCFs is further corroborated by topological analysis of the electron density as described in the ESI.† Dimetallofullerenes were already shown to stabilize unique single-electron lanthanide–lanthanide bonds.^47–50 Dy₃C₂@C₈₀ described in this work is the first example of the new lanthanide bonding motif, i.e., three-center single-electron Dy–Dy–Dy bond. Three-center metal–metal bonding was earlier predicted in trimetallofullerene Y₃@C₈₀, but in this molecule the Y–Y–Y bonding MO is occupied by two electrons.^22,23 Interestingly, Dy₃C₂@C₈₀ with a trifoliate cluster configuration has a more traditional (Dy₃)⁹⁺(C₂)³⁻@C₈₀⁶⁻ electron distribution without metal–metal bonding (see the ESI† for further analysis). Thus, variation of the C₂ unit position within the M₃C₂ cluster has a strong influence on the valence state of metal atoms.

To clarify whether the shape of the Dy₃C₂ cluster in Dy₃C₂@C₈₀ results from the templating effect of the fullerene cage or has a universal character for rare-earth TCCFs, we performed optimization of the molecule with a larger fullerene cage, Y₃C₂@C₈₈ (Y³⁺ has a comparable ionic radius (0.90 Å) to Dy³⁺). Three different shapes of the cluster were used in the starting configurations: trifoliate, planar bat ray, and bat ray with an elevated C₂ unit like in Dy₃C₂@C₈₀. In all cases, optimization converged to the structure with a flattened bat ray configuration, and re-optimization of the latter with Dy instead of Y also left the shape intact (see ESI Fig. S4†). This result agrees with the earlier study of Lu₃C₂@C₈₈, which predicted a planar shape of the Lu₃C₂ cluster, with the C₂ unit lying in the triangle plane of three metals.²¹ Finally, an earlier computational study of Y₃C₂@C₉₆ also favoured the flattened shape of the cluster.²⁴ Based on these results, we conclude that the unique structure of the Dy₃C₂ cluster supporting the Dy–Dy–Dy bonding in Dy₃C₂@C₈₀ is a result of the favourable matching between the fullerene cage size and metal size. When the metal is smaller like Sc or the fullerene cage has a larger inner space (say, C₈₈ or C₉₆), the M₃C₂ cluster tends to adopt a flattened bat ray shape, which does not lead to direct metal–metal bonding.

Electronic properties of Dy₃C₂@I_h(7)-C₈₀

We next characterized the electronic properties of Dy₃C₂@I_h(7)-C₈₀ by UV-vis-NIR spectroscopy and electrochemistry (Fig. 4). The electronic absorption spectrum of yellow Dy₃C₂@I_h(7)-C₈₀ solution in toluene looks featureless without discernible absorption peaks and is similar to the spectrum of Sc₃C₂@I_h(7)-C₈₀. Both compounds have their absorption onset at around 850–900 nm (Fig. 4a). Electrochemical characterization by cyclic voltammetry (CV) and differential pulse voltammetry (DPV) gives more distinct characteristics of the compound (Fig. 4b). Dy₃C₂@I_h(7)-C₈₀ exhibits one reversible oxidation step at +0.19 V, one reversible reduction step at E_1/2 = −0.99 V, and one irreversible reduction with a peak potential of −1.65 V. The electrochemical gap ΔE_EC of Dy₃C₂@I_h(7)-C₈₀ is thus 1.18 V. It is remarkable that the ΔE_EC gap of Sc₃C₂@I_h(7)-C₈₀ is only 0.47 V, as its first oxidation and reduction potentials are found at −0.03 V and −0.50 V, respectively (see ESI Fig. S10 and Table S6†).³⁴ Both TCCFs have their SOMO localized on the M₃C₂ cluster, but their different electronic configurations result in a considerable difference of the frontier MO energies (Fig. 3d).

Fig. 4 Electronic properties of Dy₃C₂@I_h(7)-C₈₀. (a) UV-vis-NIR absorption spectra of Dy₃C₂@I_h(7)-C₈₀ and Sc₃C₂@I_h(7)-C₈₀ dissolved in toluene. Inset: the photograph of Dy₃C₂@I_h(7)-C₈₀ solution in toluene. (b) Cyclic voltammogram (top) and differential pulse voltammogram (bottom) of Dy₃C₂@I_h(7)-C₈₀ measured in o-DCB solution with TBAPF₆ as the supporting electrolyte (scan rate: 100 mV s⁻¹); ferrocene Fe(Cp)₂ was added as the internal standard. Each redox step of Dy₃C₂@C₈₀ is marked with a solid dot; the asterisk denotes the oxidation peak of ferrocene. Variation of the formal charge of the Dy₃ fragment in different potential ranges is indicated.

The Dy–Dy–Dy-bonding nature of the SOMO in (Dy₃)⁸⁺ indicates that reduction and oxidation should substantially change the Dy–Dy bonding situation in Dy₃C₂@I_h(7)-C₈₀. Indeed, DFT calculations show that Dy–Dy bond lengths in the anion shorten to 3.363/3.366/3.320 Å as a result of the stronger Dy–Dy bonding provided by the population of the Dy₃-bonding MO by the second electron in the (Dy₃)⁷⁺ fragment. In the Dy₃C₂@C₈₀⁺ cation, the Dy–Dy distances increase to 3.668/3.473/3.542 Å as the metal–metal bonding is diminished when the Dy₃-bonding MO is completely depopulated in (Dy₃)⁹⁺. The electrochemical gap of 1.10 V estimated by DFT with the polarized continuum model of the solvent agrees well with the experimental value, confirming the reliable interpretation of the redox process by theory.

Conclusions

In summary, for the first time we synthesized and isolated a novel TCCF based on a medium-sized rare-earth metal, dysprosium (Dy) – Dy₃C₂@I_h(7)-C₈₀. Its molecular structure elucidated by SC-XRD reveals Dy₃C₂ with a bat ray configuration similar to Sc₃C₂@C₈₀. But the larger size of Dy pushes the acetylide unit above the Dy₃ plane and at the same time results in shorter Dy–Dy distances than Sc–Sc distances in Sc₃C₂@C₈₀. The electronic configuration of Dy₃C₂@C₈₀ is described as (Dy₃)⁸⁺(C₂)²⁻@C₈₀⁶⁻ and features a unique three-center single-electron Dy–Dy–Dy bond, which represents the first example of such bonding motif in molecular lanthanide chemistry. Furthermore, the population of the Dy–Dy–Dy-bonding SOMO can be changed electrochemically, providing access to ionic species with either strengthened or weakened Dy–Dy bonds. Our finding on the unprecedented three-center single-electron metal–metal bond offers new insight into the bonding theory of lanthanide coordination compounds.

Author contributions

S. F. Y. conceived and designed this research. F. J. and J. P. X. synthesized, isolated the samples and conducted characterizations. R. N. G. helped with sample separation and characterizations. M. Q. C. helped with X-ray crystallographic measurements. A. A. P. did the DFT calculations. X. M. X., Q. Y. Z. and S. Y. X. afforded decapyrrylcorannulene. F. J., J. P. X., A. A. P. and S. F. Y. co-wrote the paper, and all the authors commented on it.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was partially supported by the National Key Research and Development Program of China (2017YFA0402800), National Natural Science Foundation of China (51925206, U1932214, 21721001 and 91961113), and Deutsche Forschungsgemeinschaft (PO 1602/5 and PO 1602/7). We thank the staff of the BL17B beamline of the National Facility for Protein Science Shanghai (NFPS) at the Shanghai Synchrotron Radiation Facility for assistance during data collection and Ulrike Nitzsche for assistance with computational resources at IFW Dresden.

Notes and references

1. S. Yang, F. Liu, C. Chen, M. Jiao and T. Wei, Chem. Commun., 2011, 47, 11822–11839.
2. A. A. Popov, S. Yang and L. Dunsch, Chem. Rev., 2013, 113, 5989–6113.
3. X. Lu, L. Feng, T. Akasaka and S. Nagase, Chem. Soc. Rev., 2012, 41, 7723–7760.
4. S. Yang, T. Wei and F. Jin, Chem. Soc. Rev., 2017, 46, 5005–5058.
5. R. Guan, M. Chen, F. Jin and S. Yang, Angew. Chem., Int. Ed., 2020, 59, 1048–1073.
6. P. Yu, W. Shen, L. Bao, C. Pan, Z. Slanina and X. Lu, Chem. Sci., 2019, 10, 10925–10930.
7. D. Hao, L. Yang, B. Li, Q. Hou, L. Li and P. Jin, J. Phys. Chem. A, 2020, 124, 2694–2699.
8. W. Shen, S. Hu and X. Lu, Chem.–Eur. J., 2020, 26, 5748–5757.
9. S. Stevenson, G. Rice, T. Glass, K. Harich, F. Cromer, M. R. Jordan, J. Craft, E. Hadju, R. Bible, M. M. Olmstead, K. Maitra, A. J. Fisher, A. L. Balch and H. C. Dorn, Nature, 1999, 401, 55–57.
10. C. R. Wang, T. Kai, T. Tomiyama, T. Yoshida, Y. Kobayashi, E. Nishibori, M. Takata, M. Sakata and H. Shinohara, Angew. Chem., Int. Ed., 2001, 40, 397–399.
11. A. A. Popov, S. M. Avdoshenko, A. M. Pendás and L. Dunsch, Chem. Commun., 2012, 48, 8031–8050.
12. W. Shen, L. Bao, Y. Wu, C. Pan, S. Zhao, H. Fang, Y. Xie, P. Jin, P. Peng, F.-F. Li and X. Lu, J. Am. Chem. Soc., 2017, 139, 9979–9984.
13. C. Pan, W. Shen, L. Yang, L. Bao, Z. Wei, P. Jin, H. Fang, Y.-P. Xie, T. Akasaka and X. Lu, Chem. Sci., 2019, 10, 4707–4713.
14. X. Lu, S. Hu, W. Shen, G. Duan, L. Yang, P. Jin, Y. Xie and T. Akasaka, Chem.–Eur. J., 2019, 25, 11538–11544.
15. P. Jin, C. Tang and Z. Chen, Coord. Chem. Rev., 2014, 89, 270–271.
16. X. Lu, T. Akasaka and S. Nagase, Acc. Chem. Res., 2013, 46, 1627–1635.
17. K. Tan, X. Lu and C. R. Wang, J. Phys. Chem. B, 2006, 110, 11098–11102.
18. T.-S. Wang, N. Chen, J.-F. Xiang, B. Li, J.-Y. Wu, W. Xu, L. Jiang, K. Tan, C.-Y. Shu, X. Lu and C.-R. Wang, J. Am. Chem. Soc., 2009, 131, 16646–16647.
19. H. Fang, H. Cong, M. Suzuki, L. Bao, B. Yu, Y. Xie, N. Mizorogi, M. M. Olmstead, A. L. Balch, S. Nagase, T. Akasaka and X. Lu, J. Am. Chem. Soc., 2014, 136, 10534–10540.
20. K. Tan and X. Lu, J. Phys. Chem. A, 2006, 110, 1171–1176.
21. W. Xu, T.-S. Wang, J.-Y. Wu, Y.-H. Ma, J.-P. Zheng, H. Li, B. Wang, L. Jiang, C.-Y. Shu and C.-R. Wang, J. Phys. Chem. C, 2011, 115, 402–405.
22. K. Junghans, C. Schlesier, A. Kostanyan, N. A. Samoylova, Q. Deng, M. Rosenkranz, S. Schiemenz, R. Westerström, T. Greber, B. Büchner and A. A. Popov, Angew. Chem., Int. Ed., 2015, 54, 13411–13415.
23. K. Junghans, K. B. Ghiassi, N. A. Samoylova, Q. Deng, M. Rosenkranz, M. M. Olmstead, A. L. Balch and A. A. Popov, Chem.–Eur. J., 2016, 22, 13098–13107.
24. A. A. Popov, L. Zhang and L. Dunsch, ACS Nano, 2010, 4, 795–802.
25. Y.-J. Guo, H. Zheng, T. Yang, S. Nagase and X. Zhao, Inorg. Chem., 2015, 54, 8066–8076.
26. S. T. Liddle, Molecular metal-metal bonds, Wiley-VCH Verlag GmbH & Co. KGaA, 2015.
27. C.-S. Cao, Y. Shi, H. Xu and B. Zhao, Coord. Chem. Rev., 2018, 365, 122–144.
28. M. V. Butovskii and R. Kempe, New J. Chem., 2015, 39, 7544–7558.
29. S. Hu, P. Zhao, W. Shen, P. Yu, W. Huang, M. Ehara, Y. Xie, T. Akasaka and X. Lu, Nanoscale, 2019, 11, 13415–13422.
30. S. Yang, C. Chen, A. Popov, W. Zhang, F. Liu and L. Dunsch, Chem. Commun., 2009, 6391–6393.
31. C. Chen, F. Liu, S. Li, N. Wang, A. A. Popov, M. Jiao, T. Wei, Q. Li, L. Dunsch and S. Yang, Inorg. Chem., 2012, 51, 3039–3045.
32. F. Liu, T. Wei, S. Wang, J. Guan, X. Lu and S. Yang, Fullerenes, Nanotubes, Carbon Nanostruct., 2014, 22, 215–226.
33. Y.-Y. Xu, H.-R. Tian, S.-H. Li, Z.-C. Chen, Y.-R. Yao, S.-S. Wang, X. Zhang, Z.-Z. Zhu, S.-L. Deng, Q. Y. Zhang, S. F. Yang, S.-Y. Xie, R.-B. Huang and L.-S. Zheng, Nat. Commun., 2019, 10, 485.
34. Y. Iiduka, T. Wakahara, T. Nakahodo, T. Tsuchiya, A. Sakuraba, Y. Maeda, T. Akasaka, K. Yoza, E. Horn, T. Kato, M. T. H. Liu, N. Mizorogi, K. Kobayashi and S. Nagase, J. Am. Chem. Soc., 2005, 127, 12500–12501.
35. T. Wang, J. Wu, W. Xu, J. Xiang, X. Lu, B. Li, L. Jiang, C. Shu and C. Wang, Angew. Chem., Int. Ed., 2010, 49, 1786–1789.
36. B. Wu, T. Wang, Y. Feng, Z. Zhang, L. Jiang and C. Wang, Nat. Commun., 2015, 6, 6468.
37. E. Nishibori, I. Terauchi, M. Sakata, M. Takata, Y. Ito, T. Sugai and H. Shinohara, J. Phys. Chem. B, 2006, 110, 19215–19219.
38. S. Taubert, M. Straka, T. O. Pennanen, D. Sundholm and J. Vaara, Phys. Chem. Chem. Phys., 2008, 10, 7158–7168.
39. R. Shannon, Acta Crystallogr., Sect. A: Cryst. Phys., Diffr., Theor. Gen. Crystallogr., 1976, 32, 751–767.
40. S. F. Yang, S. I. Troyanov, A. A. Popov, M. Krause and L. Dunsch, J. Am. Chem. Soc., 2006, 128, 16733–16739.
41. D. N. Laikov and Y. A. Ustynuk, Russ. Chem. Bull., 2005, 54, 820–826.
42. F. Neese, F. Wennmohs, U. Becker and C. Riplinger, J. Chem. Phys., 2020, 152, 224108.
43. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865–3868.
44. F. Weigend and R. Ahlrichs, Phys. Chem. Chem. Phys., 2005, 7, 3297.
45. M. Dolg, H. Stoll, A. Savin and H. Preuss, Theor. Chim. Acta, 1989, 75, 173–194.
46. J. Yang and M. Dolg, Theor. Chem. Acc., 2005, 113, 212.
47. T. Zuo, L. Xu, C. M. Beavers, M. M. Olmstead, W. Fu, T. D. Crawford, A. L. Balch and H. C. Dorn, J. Am. Chem. Soc., 2008, 130, 12992–12997.
48. L. Bao, M. Chen, C. Pan, T. Yamaguchi, T. Kato, M. M. Olmstead, A. L. Balch, T. Akasaka and X. Lu, Angew. Chem., Int. Ed., 2016, 55, 4242–4246.
49. F. Liu, G. Velkos, D. S. Krylov, L. Spree, M. Zalibera, R. Ray, N. A. Samoylova, C.-H. Chen, M. Rosenkranz, S. Schiemenz, F. Ziegs, K. Nenkov, A. Kostanyan, T. Greber, A. U. B. Wolter, M. Richter, B. Büchner, S. M. Avdoshenko and A. A. Popov, Nat. Commun., 2019, 10, 571.
50. F. Liu, L. Spree, D. S. Krylov, G. Velkos, S. M. Avdoshenko and A. A. Popov, Acc. Chem. Res., 2019, 52, 2981–2993.

---

Footnotes

† Electronic supplementary information (ESI) available. CCDC 2057452. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/d1sc00965f

‡ These authors contributed equally to this work.

---