The 2016 crystallographic isolation of two Sc2O@C78 isomers related by a single-step Stone-Wales (SW) transformation supplied the 'missing links' in an interconversion network - SW rotations combined with C2 additions and extrusions - that connects the majority of solvent-extractable Sc2O@C2n (n = 38-41) cages, and DFT calculations indicated a notably low barrier for the SW step, attributed to catalysis by an exohedrally attached scandium atom. This turned the long-standing top-down versus bottom-up debate about fullerene formation into a concrete, atomically mapped question, yet the mechanism itself remains unassigned: no experiment has determined whether this network is actually traversed during arc-discharge synthesis, in which direction, or whether the computed low-barrier channels operate under real formation conditions. The problem is to establish the experimentally grounded formation mechanism of this clusterfullerene family: determine with direct experimental evidence whether the network is traversed predominantly top-down (shrinkage from larger cages), predominantly bottom-up (growth from smaller ones), or as an experimentally characterized competitive partition between both routes, and whether the computed low-barrier SW channel catalyzed by an exohedrally attached Sc atom operates under synthesis conditions. Reviews through 2024 continue to treat the fullerene formation mechanism as unresolved, and simulations to date offer only model-level, not experimental, answers.

Background

Fullerenes are hollow carbon cages made of pentagons and hexagons; endohedral clusterfullerenes additionally trap a metal-containing cluster, such as the oxide Sc2_2O, inside the cage (written Sc2_2O@C2n_{2n}). They are made almost exclusively by arc discharge between graphite electrodes doped with the metal, a plasma process at several thousand kelvin in helium, after which the products are recovered by solvent extraction of the soot and chromatographic separation. Only cages satisfying the isolated pentagon rule (IPR) - no two pentagons sharing an edge - are normally stable enough to survive this workflow, so the observable family is a biased, thermodynamically preselected subset of all cages formed.

Two classes of mechanism have been proposed for how fullerenes form. Bottom-up scenarios build the cage from small carbon fragments; top-down scenarios start from larger cages that shrink. Canonical quantum-chemical molecular dynamics simulations of the 'shrinking hot giant' road (Irle, Zheng, Wang and Morokuma, 2006) showed hot giant cages autocatalytically self-assembling and then shrinking by C2_2 loss toward C60_{60}. Both scenarios agree, however, that specific fullerenes are reached through a stepwise process of Stone-Wales transformations (SWTs) - the 90-degree rotation of a shared C-C bond that interconverts four hexagons with two pentagon-hexagon pairs, the elementary cage-isomerization step - combined with C2_2 additions (cage growth) and C2_2 extrusions (cage shrinkage). In empty C60_{60} the SWT carries a prohibitive barrier of about 7 eV, but prior density functional theory (DFT) established that charge transfer from encapsulated metal atoms lowers it appreciably (Choi, Kim, Han and Ihm, 2006), and that heteroatom substitution can lower it dramatically (Kabir, Mukherjee and Saha-Dasgupta, 2011), making SW-mediated rearrangement plausible inside metallofullerenes.

The bottleneck for testing any such pathway experimentally has been the scarcity of 'missing link' cages: isolated crystallographic structures of isomers that are actually adjacent on the hypothetical interconversion network. Hao et al. (2016) closed this gap for one family. They isolated and crystallographically characterized Sc2_2O@C2v_{2v}(3)-C78_{78} and Sc2_2O@D3h_{3h}(5)-C78_{78}, which differ by exactly one SW rotation, and showed that these two molecules are the key links in a well-defined pathway connecting most solvent-extractable Sc2_2O@C2n_{2n} (n=38n = 38-4141) cages through SWT and C2_2 addition/extrusion steps - the first formation pathway defined at the atomic level for a clusterfullerene family. A companion DFT study (Zhao, Li, Guo, Zhao and Zhao, 2016) had independently predicted these two isomers to be the thermodynamically stable Sc2_2O@C78_{78} species and to be related by a single SW transformation. Hao et al. further computed a notably low activation barrier for the SWT (a channel catalyzed by an exohedrally attached scandium atom that lowers the SW kinetic barrier), and their follow-up Account (Feng, Hao, Liu and Slanina, 2019) summarizes the resulting picture: two competitive transformation pathways - C2_2 loss and C2_2 addition - are established for the majority of oxide clusterfullerenes, but neither direction is selected.

What the structural map cannot tell is the direction of travel. The same network of SW rotations and C2_2 steps describes both a top-down cascade from larger cages and a bottom-up assembly from smaller ones; establishing the map does not determine which direction the chemistry actually follows in the arc reactor - or whether both routes coexist - nor whether the computed low-barrier SW channels are kinetically operative at the temperatures, metal vapor densities, and cooling trajectories of real synthesis. Resolving this matters practically: endohedral metallofullerene applications are constrained by low and unselective yields, and a verified formation mechanism is the precondition for rational, selective synthesis.

Problem Statement

For the Sc2_2O@C2n_{2n} (n=38n = 38-4141) clusterfullerene family synthesized by arc discharge, determine the actual formation mechanism of the carbon cages - i.e., whether the Stone-Wales/C2_2 interconversion network established crystallographically by Hao et al. is genuinely traversed during synthesis, and how. Concretely, two linked questions define the problem, both serving one objective, an experimentally grounded mechanism: (1) Direction: is the trajectory through the network predominantly top-down (progressive C2_2 extrusion and SW annealing from larger cages), predominantly bottom-up (C2_2 addition to smaller cages), or an experimentally characterized competitive partition between both routes, under the family's standard arc-discharge/helium synthesis conditions as reported by Hao et al. (2016)? The assignment must rest on direct experimental evidence sufficient to determine experimentally which of these outcomes obtains, as the authors themselves state that further experimental evidence is required and the source framing establishes two competitive pathways without selecting one. (2) Operative channels: does the computed low-barrier Stone-Wales channel - catalyzed by an exohedrally attached scandium atom that lowers the SW kinetic barrier - actually operate under synthesis conditions? A channel counts as operative here only on experimental evidence that SW isomerization actually proceeds through it - not merely on a computed barrier that is consistent with the observed products - and the discriminating signature must be identified. Kinetic quantification of the SW/C2_2 events relative to competing channels would further constrain such an assignment. How the operative channels depend on the encapsulated cluster (its size, geometry, and electron count) and on the reaction environment (metal vapor density, transient exohedral metal contacts, temperature and pressure trajectories) is a natural generalization strand - different transformation routes can be expected depending on the encapsulated species - but it lies beyond the completion criterion for this problem. The question requires linking the static structural map and DFT barriers to the real-time kinetics of cage growth and isomerization: an experimentally grounded assignment of how the SW/C2_2 network is traversed (direction, including possible competitive coexistence) and whether the computed exohedral-Sc-assisted low-barrier SW channel operates under synthesis conditions, rather than a picture merely consistent with computation.

References

  1. Hao, Yajuan; Tang, Qiangqiang; Li, Xiaohong; Zhang, Meirong; Wan, Yingbo; Feng, Lai; Chen, Ning; Slanina, Zdenek; Adamowicz, Ludwik; Uhlik, Filip (2016). Isomeric Sc2O@C78 Related by a Single-Step Stone-Wales Transformation: Key Links in an Unprecedented Fullerene Formation Pathway. Inorganic Chemistry 55, 11354-11361. DOI 10.1021/acs.inorgchem.6b01894. https://doi.org/10.1021/acs.inorgchem.6b01894
  2. Zhao, Pei; Li, Meng-Yang; Guo, Yi-Jun; Zhao, Rui-Sheng; Zhao, Xiang (2016). Single Step Stone-Wales Transformation Linking Two Thermodynamically Stable Sc2O@C78 Isomers. Inorganic Chemistry 55, 2220-2226. DOI 10.1021/acs.inorgchem.5b02591. https://doi.org/10.1021/acs.inorgchem.5b02591
  3. Irle, Stephan; Zheng, Guishan; Wang, Zhi; Morokuma, Keiji (2006). The C60 Formation Puzzle 'Solved': QM/MD Simulations Reveal the Shrinking Hot Giant Road of the Dynamic Fullerene Self-Assembly Mechanism. The Journal of Physical Chemistry B 110, 14531-14545. DOI 10.1021/jp061173z. https://doi.org/10.1021/jp061173z
  4. Choi, Woon Ih; Kim, Gunn; Han, Seungwu; Ihm, Jisoon (2006). Reduction of activation energy barrier of Stone-Wales transformation in endohedral metallofullerenes. Physical Review B 73, 113406. DOI 10.1103/PhysRevB.73.113406. https://doi.org/10.1103/PhysRevB.73.113406
  5. Kabir, M.; Mukherjee, S.; Saha-Dasgupta, T. (2011). Substantial reduction of Stone-Wales activation barrier in fullerene. Physical Review B 84, 205404. DOI 10.1103/PhysRevB.84.205404. https://doi.org/10.1103/PhysRevB.84.205404
  6. Slanina, Zdenek; Uhlik, Filip; Feng, Lai; Adamowicz, Ludwik (2017). Sc2O@C78: Calculations of the yield ratio for two observed isomers. Fullerenes, Nanotubes and Carbon Nanostructures 25, 124-127. DOI 10.1080/1536383X.2016.1250078. https://doi.org/10.1080/1536383X.2016.1250078
  7. Feng, Lai; Hao, Yajuan; Liu, Along; Slanina, Zdenek (2019). Trapping Metallic Oxide Clusters inside Fullerene Cages. Accounts of Chemical Research 52, 1802-1811. DOI 10.1021/acs.accounts.9b00206. https://doi.org/10.1021/acs.accounts.9b00206
  8. Bao, Lipiao; Yu, Pengyuan; Pan, Changwang; Shen, Wangqiang; Lu, Xing (2019). Crystallographic identification of Eu@C2n (2n = 88, 86 and 84): completing a transformation map for existing metallofullerenes. Chemical Science 10, 2153-2158. DOI 10.1039/c8sc04906h. https://doi.org/10.1039/c8sc04906h
  9. Shen, Wangqiang; Bao, Lipiao; Hu, Shuaifeng; Yang, Le; Jin, Peng; Xie, Yunpeng; Akasaka, Takeshi; Lu, Xing (2019). Crystallographic characterization of Lu2C2n (2n = 76-90): cluster selection by cage size. Chemical Science 10, 829-836. DOI 10.1039/c8sc03886d. https://doi.org/10.1039/c8sc03886d
  10. Guan, Runnan; Chen, Muqing; Xin, Jinpeng; Xie, Xiao-Ming; Jin, Fei; Zhang, Qianyan; Xie, Su-Yuan; Yang, Shangfeng (2021). Capturing the Missing Carbon Cage Isomer of C84 via Mutual Stabilization of a Triangular Monometallic Cyanide Cluster. Journal of the American Chemical Society 143, 8078-8085. DOI 10.1021/jacs.1c02428. https://doi.org/10.1021/jacs.1c02428
  11. Gu, Yong-Xin; Zheng, Hong; Li, De-Huai; Zhao, Xiang (2023). M@C78 (M = U, Th): Inherent Topological Connectivity Existed in Thermodynamically Stable Isomers and the Possibility of an Endohedral Fullerene Containing One Heptagon Ring. The Journal of Physical Chemistry A 127, 6012-6019. DOI 10.1021/acs.jpca.3c02266. https://doi.org/10.1021/acs.jpca.3c02266
  12. Yu, Yanling; Slanina, Zdenek; Wang, Feng; Yang, Ying; Lian, Yongfu; Uhlik, Filip; Xin, Baifu; Feng, Lai (2020). Ho2O@D3(85)-C92: Highly Stretched Cluster Dictated by a Giant Cage and Unexplored Isomerization. Inorganic Chemistry 59, 11020-11027. DOI 10.1021/acs.inorgchem.0c01512. https://doi.org/10.1021/acs.inorgchem.0c01512
  13. Li, Meng-Yang; Zhao, Yao-Xiao; Han, Yan-Bo; Yuan, Kun; Nagase, Shigeru; Ehara, Masahiro; Zhao, Xiang (2020). Theoretical Investigation of the Key Roles in Fullerene-Formation Mechanisms: Enantiomer and Enthalpy. ACS Applied Nano Materials 3, 547-554. DOI 10.1021/acsanm.9b02110. https://doi.org/10.1021/acsanm.9b02110
  14. Li, Mengyang; Zhao, Ruisheng; Dang, Jingshuang; Zhao, Xiang (2022). Theoretical study on the stabilities, electronic structures, and reaction and formation mechanisms of fullerenes and endohedral metallofullerenes. Coordination Chemistry Reviews 471, 214762. DOI 10.1016/j.ccr.2022.214762. https://doi.org/10.1016/j.ccr.2022.214762
  15. Crichton, Ryan A.; Zhang, Jianyuan (2022). Formation Mechanism of Fullerenes/Metallofullerenes. In: Handbook of Fullerene Science and Technology, Springer, 991-1020. DOI 10.1007/978-981-16-8994-9_44. https://doi.org/10.1007/978-981-16-8994-9_44
  16. Popov, Andrey M.; Lebedeva, Irina V.; Vyrko, Sergey A.; Poklonski, Nikolai A. (2021). Multiscale modeling strategy to solve fullerene formation mystery. Fullerenes, Nanotubes and Carbon Nanostructures 29, 755-766. DOI 10.1080/1536383X.2021.1900124. https://doi.org/10.1080/1536383X.2021.1900124 (preprint arXiv:2108.07598, https://arxiv.org/abs/2108.07598)
  17. Fan, Huichen; Liu, Zhenyu; Gan, Li-Hua; Wang, Chun-Ru (2024). The formation mechanism of Sc-based metallofullerenes: a molecular dynamics simulation study. Physical Chemistry Chemical Physics 26, 5499-5507. DOI 10.1039/d3cp05587f. https://doi.org/10.1039/d3cp05587f
  18. Liu, Fupin; Popov, Alexey A. (2024). Capturing Unstable Metallofullerenes. Inorganics 12, 48. DOI 10.3390/inorganics12020048. https://doi.org/10.3390/inorganics12020048
  19. Chuvilin, Andrey; Kaiser, Ute; Bichoutskaia, Elena; Besley, Nicholas A.; Khlobystov, Andrei N. (2010). Direct transformation of graphene to fullerene. Nature Chemistry 2, 450-453. DOI 10.1038/nchem.644. https://doi.org/10.1038/nchem.644
  20. Hou, Yameng; Kong, Xianglei (2024). Endometallofullerenes in the Gas Phase: Progress and Prospect. Inorganics 12, 68. DOI 10.3390/inorganics12030068. https://doi.org/10.3390/inorganics12030068