Search

Controlled crystallization of fullerene molecules into ordered molecular assemblies is important for their applications. However, the morphology engineering of fullerene[C70 ] assemblies is challenging, and complicated architectures have rarely been reported due to the low molecular symmetry of C70 molecules, which makes their crystallization difficult to control and the low production yield as well. Herein, with the assistance of solvent intercalation, a general reprecipitation approach is reported to prepare morphologically controllable C70 microcrystals with mesitylene as a good solvent and n-propanol as a poor solvent in one solvent system without replacing specific solvents. A series of C70 microcrystals with high uniformity from perfect cubes and defective hoppers to novel cruciform-pillars are obtained by intentionally tuning C70 concentration and the volume ratio of mesitylene to n-propanol. Among them, novel cruciform-pillar-shaped microcrystals are obtained for the first time by further decreasing the amount of mesitylene in the solvent-intercalated microcrystals. Notably, the C70 concentration is a key parameter for the selective growth of C70 hopper, rather than the volume ratio of mesitylene to n-propanol. Interestingly, the hopper-shaped microcrystals exhibit excellent photoluminescence properties relative to those of cubes and cruciform-pillars owing to the enhanced light absorption, proving their potential applications in optoelectronic devices. This study offers new insights into the morphology-controlled synthesis of other micro/nanostructured organic microcrystals and the fine tuning of photoluminescence properties of organic crystals.

The chemical properties of a prototypical labile mono-EMF, Y@Cs(6)-C82, have been systematically disclosed for the first time via a Bingel–Hirsch reaction. Three mono-adduct isomers, namely, 2a, 2b and 2c out of 44 possibilities for the Y@Cs(6)-C82 cage have been readily isolated, demonstrating surprisingly high regioselectivity. Crystallographic results of 2b unambiguously confirm its molecular structure with a singly bonded bromomalonate group attached onto the Cs(6)-C82 cage. Further computational results rationalize that the high regioselectivity is a consequence of the localization of high spin density and large frontier molecular orbital distribution on the corresponding carbon atoms stemming from the encapsulation of an yttrium atom into the low-symmetry Cs(6)-C82 cage with three-electron transfer from the metal to the cage.

Successful isolation and characterization of a series of Er-based dimetallofullerenes present valuable insights into the realm of metal-metal bonding. These species are crystallographically identified as Er2 @Cs (6)-C82 , Er2 @C3v (8)-C82 , Er2 @C1 (12)-C84 , and Er2 @C2v (9)-C86 , in which the structure of the C1 (12)-C84 cage is unambiguously characterized for the first time by single-crystal X-ray diffraction. Interestingly, natural bond orbital analysis demonstrates that the two Er atoms in Er2 @Cs (6)-C82 , Er2 @C3v (8)-C82 , and Er2 @C2v (9)-C86 form a two-electron-two-center Er-Er bond. However, for Er2 @C1 (12)-C84 , with the longest Er⋅⋅⋅Er distance, a one-electron-two-center Er-Er bond may exist. Thus, the difference in the Er⋅⋅⋅Er separation indicates distinct metal bonding natures, suggesting a distance-dependent bonding behavior for the internal dimetallic cluster. Additionally, electrochemical studies suggest that Er2 @C82-86 are good electron donors instead of electron acceptors. Hence, this finding initiates a connection between metal-metal bonding chemistry and fullerene chemistry.

To investigate the intrinsic reactivity of atomic nitrogen, which had previously been accomplished only by examining its decay in the gas phase using special equipment, a nitrogen atom was inserted into a series of molecule-encapsulating C60 and C70 fullerenes. Among the studied endofullerenes, H2 @C70 was able to encapsulate an additional nitrogen atom within the fullerene cage under radiofrequency plasma conditions. The product was analyzed by ESR spectroscopy and mass spectrometry in solution, which revealed that the nitrogen atom with a quartet ground state does not react but weakly interact with the H2 molecule, thus demonstrating the utility of such fullerenes as "nanoflasks".

Additions of adamantylidene (Ad) to M3 N@Ih -C80 (M=Sc, Lu) and Sc3 N@D5h -C80 have been accomplished by photochemical reactions with 2-adamantyl-2,3'-[3H]-diazirine (1). In M3 N@Ih -C80 , the addition led to rupture of the [6,6]- or [5,6]-bonds of the Ih -C80 cage, forming the [6,6]-open fulleroid as the major isomer and the [5,6]-open fulleroid as the minor isomer. In Sc3 N@D5h -C80 , the addition also proceeded regioselectively to yield three major isomeric Ad mono-adducts, despite the fact that there are nine types of C-C bonds in the D5h -C80 cage. The molecular structures of the seven Ad mono-adducts, including the positions of the encaged trimetallic nitride clusters, have been unambiguously determined through single-crystal XRD analyses. Furthermore, results have shown that stepwise addition of Ad to Lu3 N@Ih -C80 affords several Ad bis-adducts, two of which have been isolated and characterized. The X-ray structure of one bis-adduct clearly revealed that the second Ad addition took place at a [6,6]-bond close to an endohedral metal atom. Theoretical calculations have also been performed to rationalize the regioselectivity.

We compared the chemical reactivity of D2d(23)-C84 and that of Sc2C2@D2d(23)-C84, both having the same carbon cage geometry, in the photolysis of 2-adamantane-2,3′-[3H]-diazirine, to clarify metal-atom doping effects on the chemical reactivity of the carbon cage. Experimental and computational studies have revealed that the chemical reactivity of the D2d(23)-C84 carbon cage is altered drastically by endohedral Sc2C2 doping. The reaction of empty D2d(23)-C84 with the diazirine under photoirradiation yields two adamantylidene (Ad) adducts. NMR spectroscopic studies revealed that the major Ad monoadduct (C84(Ad)-A) has a fulleroid structure and that the minor Ad monoadduct (C84(Ad)-B) has a methanofullerene structure. The latter was also characterized using X-ray crystallography. C84(Ad)-A is stable under photoirradiation, but it interconverted to C84(Ad)-B by heating at 80 °C. In contrast, the reaction of endohedral Sc2C2@D2d(23)-C84 with diazirine under photoirradiation affords four Ad monoadducts (Sc2C2@C...

Charge transfer in metallofullerene/porphyrin cocrystals is revealed for the first time. Originated from the different solvents for crystallization, distinct stacking manners are presented in the two types of cocrystals. It is demonstrated that intermolecular packing, next to the nature of the corresponding electron donors and acceptors, dominates the charge transfer processes.

We report that Ce@C2v(9)-C82 forms a centrosymmetric dimer when co-crystallized with Ni(OEP) (OEP = octaethylporphyrin dianion). The crystal structure of {Ce@C2v(9)-C82}2⋅2[Ni(OEP)]⋅4 C6H6 shows that a new C−C bond with a bond length of 1.605(5) A connects the two cages. The high spin density of the singly occupied molecular orbital (SOMO) on the cage and the pyramidalization of the cage are factors that favor dimerization. In contrast, the treatment of Ni(OEP) with M@C2v(9)-C82 (M = La, Sc, and Y) results in crystallization of monomeric endohedral fullerenes. A systematic comparison of the X-ray structures of M@C2v(9)-C82 (M = Sc, Y, La, Ce, Gd, Yb, and Sm) reveals that the major metal site in each case is located at an off-center position adjacent to a hexagonal ring along the C2 axis of the C2v(9)-C82 cage. DFT calculations at the M06-2X level revealed that the positions of the metal centers in these metallofullerenes M@C2v(9)-C82 (M = Sc, Y, and Ce), as determined by single-crystal X-ray structure studies, correspond to an energy minimum for each compound.

Although the major isomers of M@C82 (namely M@C2v (9)-C82 , where M is a trivalent rare-earth metal) have been intensively investigated, the lability of the minor isomers has meant that they have been little studied. Herein, the first isolation and crystallographic characterization of the minor Y@C82 isomer, unambiguously assigned as Y@Cs (6)-C82 by cocrystallization with Ni(octaethylporphyrin), is reported. Unexpectedly, a regioselective dimerization is observed in the crystalline state of Y@Cs (6)-C82 . In sharp contrast, no dimerization occurs for the major isomer Y@C2v (9)-C82 under the same conditions, indicating a cage-symmetry-induced dimerization process. Further experimental and theoretical results disclose that the regioselective dimer formation is a consequence of the localization of high spin density on a special cage-carbon atom of Y@Cs (6)-C82 which is caused by the steady displacement of the Y atom inside the Cs (6)-C82 cage.

The awesome allotropy of carbon yields innumerable topologically possible cage structures of molecular carbon. This field is also related to endohedral metallofullerenes constructed by metal-atom encapsulation. Stable and soluble empty fullerenes and endohedral metallofullerenes are available in pure form in macroscopic amounts from carbon arc production or other physical processes followed by extraction and subsequent chromatographic separation. However, many other unidentified fullerene species, which must be reactive and insoluble in their pristine forms, remain in soot. These "missing" species must have extremely small HOMO-LUMO gaps and may have unconventional cage structures. Recent progress in this field has demonstrated that reactive fullerenes can be salvaged by exohedral derivatization, which can stabilize the reactive carbon cages. This concept provides a means of preparing macroscopic amounts of unconventional fullerenes as their derivatives.

Bis-silylated and bis-germylated derivatives of Lu3 N@Ih -C80 (3, 4, 5) were successfully synthesized by the photochemical addition of disiliranes 1 a, 1 b or digermirane 2, and fully characterized by spectroscopic, electrochemical, and theoretical studies. Interestingly, digermirane 2 reacts more efficiently than disiliranes 1 a and 1 b because of its good electron-donor properties and lower steric hindrance around the Ge-Ge bond. The 1,4-adduct structures of 3, 4, 5 were unequivocally established by single-crystal X-ray crystallographic analyses. The electrochemical and theoretical studies reveal that the energy gaps between the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) of the 1,4-adducts are remarkably smaller than those of Lu3 N@Ih -C80 , because the electron-donating groups effectively raise the HOMO levels. It is also observed that germyl groups are slightly more electron-donating than the silyl groups on the basis of the redox properties and the HOMO-LUMO energies of 4 and 5. Bis-silylation and bis-germylation are effective and versatile methods for tuning the electronic characteristics of endohedral metallofullerenes.

To activate oxidative and/or reductive electron transfer reactions, N-pyridyl-substituted Sc3N@I(h)-C80 (4) and C60 (3) fulleropyrrolidines have been prepared and axially coordinated to electron-rich (1) or electron-deficient (2) Zn(II)phthalocyanines (Zn(II)Pcs) through zinc-pyridyl, metal-ligand coordination affording a full-fledged family of electron donor-acceptor ensembles. An arsenal of photophysical assays as they were carried out with, for example, 1/4 and 2/4 show unambiguously that a Zn(II)Pc-to-Sc3N@I(h)-C80 photoinduced electron transfer takes place in the former ensemble, whereas a Sc3N@I(h)-C80-to-Zn(II)Pc electron transfer occurs in the latter ensemble. To the best of our knowledge, this is the first time that a fullerene-based molecular building block shows an electron transfer dichotomy, namely acting both as electron-acceptor or electron-donor, and its outcome is simply governed by the electronic nature of its counterpart. In light of the latter, the present work, which involves the use of Sc3N@I(h)-C80, one of the most abundant and easy-to-purify endohedral metallofullerenes, is, on one hand, a paradigmatic change and, on the other hand, an important milestone en-route toward the construction of easy-to-prepare molecular materials featuring switchable electron transfer reactivity.

We demonstrate that a finite-length (10,0) carbon nanotube (CNT) with two fullerene caps, namely D5(450)-C100, is an ideal prototype to study the mechanical responses of small CNTs upon endohedral metal doping. Encapsulation of a large La2C2 cluster inside D5(450)-C100 induces a 5% axial compression of the cage, as compared with the structure of La2@D5(450)-C100. Detailed crystallographic analyses reveal quantitively the flexibility of the [10]cyclacene-sidewall segment and the rigidity of the pentagon-dominating caps for the first time. The internal C2-unit acts as a molecular spring that attracts the surrounding cage carbon atoms through strong interactions with the two moving lanthanum ions. This is the first crystallographic observation of the axial compression of CNTs caused by the internal stress, which enhances our knowledge about the structural deformation of novel carbon allotropes at the atomic level.

The chemical properties of carbide-cluster metallofullerenes (CCMFs) remain largely unexplored, although several new members of CCMFs have been discovered recently. Herein, we report the reaction between Sc2 C2 @C3v (8)-C82 , which is viewed as a prototypical CCMF because of its high abundance, and 3-triphenylmethyl-5-oxazolidinone (1) to afford the corresponding pyrrolidino derivative Sc2 C2 @C3v (8)-C82 (CH2 )2 NTrt (2; Trt=triphenylmethyl). Single-crystal X-ray crystallography studies of 2 revealed that the reaction takes place at a [6,6]-bond junction, which is directly over the encapsulated C2 unit and is far from either of the two scandium atoms. On the basis of theoretical calculations and by considering previously reports, we have found that a hexagonal carbon ring on the cage of Sc2 C2 @C3v (8)-C82 is highly reactive toward different reagents due to the overlap of high p-orbital axis vector (POAV) angles and large LUMO coefficients. We propose that this highly concentrated area of reactivity is generated by the encapsulation of the Sc2 C2 cluster because this region is absent from the empty fullerene C3v (8)-C82 . Moreover, the absorption and electrochemical results confirm that derivative 2 is more stable than pristine Sc2 C2 @C3v (8)-C82 , thus illuminating its potential applications.

The thermal reaction of the endohedral metallofullerene La2 @D2 (10611)-C72 , which contains two pentalene units at opposite ends of the cage, with 5,6-diphenyl-3-(2-pyridyl)-1,2,4-triazine proceeded selectively to afford only two bisfulleroid isomers. The molecular structure of one isomer was determined using single-crystal X-ray crystallography. The results suggest that the [4+2] cycloaddition was initiated in a highly regioselective manner at the C-C bond connecting two pentagon rings of C72 . Subsequent intramolecular electrocyclization followed by cycloreversion resulted in the formation of an open-cage derivative having three seven-membered ring orifices on the cage and a significantly elongated cage geometry. The reduction potentials of the open-cage derivatives were similar to those of La2 @D2 -C72 whereas the oxidation potentials were shifted more negative than those of La2 @D2 -C72 . These results point out that further oxidation could occur easily in the derivatives.

Fullerenyl radicals can be generated by addition of a free radical to a fullerene surface, by nucleophilic addition followed by one-electron oxidation, or by thermal dissociation of singly bonded fullerene dimers. However, fullerenyl radicals are usually very reactive and generally cannot be isolated. On the contrary, we have found that the reactions of the dimetallic endofullerenes, La2@Ih-C80 and La2@D5h-C80, with 3-chloro-5,6-diphenyltriazine resulted in mono-addition of the triazinyl radical to the fullerene cages to yield isolable fullerenyl radicals. The unusual stability of these fullerenyl radicals arises from the confinement of the unpaired electron to an internal, metal-metal bonding orbital. Accordingly, the fullerene cage protects the radical center from other reactive species. Furthermore, we demonstrate that the fullerenyl radical adduct of La2@Ih-C80 reacts with toluene to afford additional benzylation. Interestingly, the benzylated derivative is diamagnetic in solution, while it forms a paramagnetic dimer when crystallized.

The endohedral fullerene once erroneously identified as Sc3@C82 was recently shown to be Sc3C2@Ih-C80, the first example of an open-shell cluster metallofullerene. We herein report that benzyl bromide (1) reacts with Sc3C2@ Ih-C80 via a regioselective radical addition that affords only one isomer of the adduct Sc3C2@Ih-C80(CH2C6H5) (2) in high yield. An X-ray crystallographic study of 2 demonstrated that the benzyl moiety is singly bonded to the fullerene cage, which eliminates the paramagnetism of the endohedral in agreement with the ESR results. Interestingly, X-ray results further reveal that the 3-fold disordered Sc3C2 cluster adopts two different configurations inside the cage. These configurations represent the so-called "planar" form and the computationally predicted, but not crystallographically characterized, "trifoliate" form. It is noteworthy that this is the first crystallographic observation of the "trifoliate" form for the Sc3C2 cluster. In contrast, crystallographic investigation of a Sc3C2@Ih-C80/Ni(OEP) cocrystal, in which the endohedral persists in an open-shell structure with paramagnetism, indicates that only the former form occurs in pristine Sc3C2@ Ih-C80. These results demonstrate that the cluster configuration in EMFs is highly sensitive to the electronic structure, which is tunable by exohedral modification. In addition, the electrochemical behavior of Sc3C2@Ih-C80 has been markedly changed by the radical addition, but the absorption spectra of the pristine and the derivative are both featureless. These results suggest that the unpaired electron of Sc3C2@Ih-C80 is buried in the Sc3C2 cluster and does not affect the electronic configuration of the cage.

Reactions of Sc3N@Ih-C80 with aziridine derivatives were conducted to afford the corresponding mono-adducts. A pair of diastereomers of the mono-adduct [5,6]-pyrrolidino-Sc3N@Ih-C80 was isolated and characterized by means of mass spectrometry, vis-NIR absorption spectroscopy, and electrochemical measurements. Structural analysis of the mono-adducts was conducted by NMR and single-crystal X-ray structure determinations.

Although all the pure-carbon fullerene isomers above C60 reported to date comply with the isolated pentagon rule (IPR), non-IPR structures, which are expected to have different properties from those of IPR species, are obtainable either by exohedral modification or by endohedral atom doping. This report describes the isolation and characterization of a new endohedral metallofullerene (EMF), La2@C76, which has a non-IPR fullerene cage. The X-ray crystallographic result for the La2@C76/[Ni(II)(OEP)] (OEP=octaethylporphyrin) cocrystal unambiguously elucidated the C(s)(17,490)-C76 cage structure, which contains two adjacent pentagon pairs. Surprisingly, multiple metal sites were distinguished from the X-ray data, which implies dynamic behavior for the two La(3+) cations inside the cage. This dynamic behavior was also corroborated by variable-temperature (139)La NMR spectroscopy. This phenomenon conflicts with the widely accepted idea that the metal cations in non-IPR EMFs invariably coordinate strongly with the negatively charged fused-pentagon carbons, thereby providing new insights into modern coordination chemistry. Furthermore, our electrochemical and computational studies reveal that La2@C(s)(17,490)-C76 has a larger HOMO-LUMO gap than other dilanthanum-EMFs with IPR cage structures, such as La2@D(3h)(5)-C78 and La2@I(h)(7)-C80, which implies that IPR is no longer a strict rule for EMFs.

The dream target of artificial photosynthesis is the realization of long-lived radical ion pair states that power catalytic centers and, consequently, the production of solar fuels. Notably, magnetic field effects, especially internal magnetic field effects, are rarely employed in this context. Here, we report on a linear Lu3N@Ih-C80-PDI electron donor-acceptor conjugate, in which the presence of the Lu3N cluster exerts an appreciable electron nuclear hyperfine coupling on the charge transfer dynamics. As such, a fairly efficient radical ion pair intersystem crossing converts the initially formed singlet radical ion pair state, (1)[(Lu3N@Ih-C80)(•+)-PDI(•-)], to the corresponding triplet radical ion pair state, (3)[(Lu3N@Ih-C80)(•+)-PDI(•-)]. Most notably, the radical ion pair state lifetime of the latter is nearly 1000 times longer than that of the former.

This report describes a helicity-selective photoreaction of single-walled carbon nanotubes (SWNTs) with disulfide in the presence of oxygen. The SWNTs were characterized using absorption, photoluminescence (PL), Raman, and X-ray photoelectron spectroscopy, scanning electron microscopy, and current–voltage (I–V) measurements. Results showed remarkable helicity-selective (metallic SWNTs/semiconducting SWNTs and diameter) functionalization of SWNTs. The reaction rate decreases in the order of metallic SWNTs > semiconducting SWNTs and small-diameter SWNTs > large-diameter SWNTs. Control experiments conducted under various experimental conditions and ESR and femtosecond laser flash photolysis measurements revealed that the helicity-selective reaction proceeds via a photoinduced electron transfer reaction. The PL and I–V measurements showed that the photoreaction is effective not only to control SWNT conductivity but also for the band gap modulation of semiconducting SWNTs.

Location recognition at the molecular scale provides valuable information about the nature of functional molecular materials. This study presents a novel location sensing approach based on an endohedral metallofullerene, Ce@C82, using its anisotropic magnetic properties, which lead to temperature-dependent paramagnetic shifts in (1)H NMR spectra. Five site-isomers of Ce@C82CH2-3,5-C6H3Me2 were synthesized to demonstrate the spatial sensing ability of Ce@C82. Single-crystal structures, absorption spectra, and density functional theory calculations were used to select the plausible addition positions in the radical coupling reaction, which preferentially happens on the carbon atoms with high electron density of the singly occupied molecular orbital (SOMO) and positive charge. Temperature-dependent NMR measurements demonstrated unique paramagnetic shifts of the (1)H peaks, which were derived from the anisotropic magnetism of the f-electron in the Ce atom of the isomers. It was found that the magnetic anisotropy axes can be easily predicted by theoretical calculations using the Gaussian 09 package. Further analysis revealed that the temperature-dependent trend in the shifts is clearly predictable from the distance and relative position of the proton from the Ce atom. Hence, the Ce-encapsulated metallofullerene Ce@C82 can provide spatial location information about nearby atoms through the temperature-dependent paramagnetic shifts of its NMR signals. It can act as a molecular probe for location sensing by utilizing the anisotropic magnetism of the encapsulated Ce atom. The potentially low toxicity and stability of the endohedral fullerene would make Ce@C82 suitable for applications in biology and material science.

Single crystals of three soluble Yb@C(82) isomers, namely, Yb@C(2)(5)-C(82), Yb@C(s)(6)-C(82), and Yb@C(2v)(9)-C(82), cocrystallized with Ni(II)(octaethylporphyrin), allowed accurate crystallographic elucidation of their molecular structures in terms of both cage symmetry and metal location. Multiple metal positions were found in all these isomers, but the major metal sites were found in some specific regions within these cages. Specifically, the Yb(2+) ion prefers to reside close to a hexagonal ring in Yb@C(2)(5)-C(82) and Yb@C(2v)(9)-C(82) but a [5,6,6]-junction carbon atom in Yb@C(s)(6)-C(82). Theoretical calculations at the B3LYP level revealed that these metal positions all correspond to energy minima from the electrostatic potential maps and give rise to the most stable configurations of these Yb@C(82) isomers. Furthermore, it is noteworthy that this is the first report on X-ray crystallographic studies of such metallofullerenes with the popular C(2v)(9)-C(82) encapsulating a divalent metal ion, described as M(2+)@[C(2v)(9)-C(82)](2-).

Functionalization of endohedral metallofullerenes has been shown to differ depending on photochemical or thermal pathways. We report that Lu(3)N@I(h)-C(80) reacts with thermally generated bis(2,6-diethylphenyl)silylene with high selectivity and forms monosilylated derivative 1b. Unexpectedly, 1b undergoes photochemical conversion to afford isomer 1a under ambient light. These adducts were characterized using NMR, visible-near-IR spectroscopy, and matrix-assisted laser desorption ionization time-of-flight mass spectrometry. Single-crystal X-ray structure determination of 1a reveals a rare example of an open 1,2-adduct at the [5,6]-ring junction of the I(h)-C(80) cage. The electrochemical study reveals that the redox potentials of 1a and 1b are shifted cathodically compared to those of pristine Lu(3)N@I(h)-C(80) and that monosilylation is effective to fine-tune the electronic properties of endohedral metallofullerenes as well as empty fullerenes. Density functional theory calculations were also performed, which provide a theoretical basis for the structures and the behavior of the encapsulated Lu(3)N cluster.

It has been an interesting finding that reactions of M@C(2v)(9)-C(82) (M = Y, La, Ce, Gd) with diazirine adamantylidene (AdN(2), 1) gave rise to only two monoadduct isomers, indicating that the cage reactivity of monometallofullerenes is not dependent on the type of the internal metal. However, we found here that Sc@C(2v)(9)-C(82) shows an exceptional chemical reactivity toward the electrophile 1, affording four monoadduct isomers (2a-d). Single-crystal X-ray diffraction crystallographic results of the most abundant isomer (2a) confirm that the addition takes place at a [6,6]-bond junction which is very close to the internal metal ion. Theoretical calculations reveal that 2 out of the 24 nonequivalent cage carbons of Sc@C(2v)(9)-C(82) are highly reactive toward 1, but only one cage carbon of the other M@C(2v)-C(82) (M = Y, La, Ce, Gd) is sufficiently reactive. The exceptional chemical property of Sc@C(2v)(9)-C(82) is associated with the small ionic radius of Sc(3+), which allows stronger metal-cage interactions and makes back-donation of charge from the cage to the metal more pronounced. Our results have provided new insights into the art of altering the chemical properties of fullerene molecules at the atomic level, which would be useful in the future in utilizing EMFs in quantum computing systems.

The reaction of tetrazine 1 with Sc(3)C(2)@C(80) exclusively affords the open-cage derivative 2 instead of the expected C(2)-inserted derivative 3 bearing a four-membered ring, as previously obtained for C(60). The structure of 2 has been firmly established by NMR spectroscopy and theoretical calculations. EPR spectroscopy shows that a single Sc atom of the Sc(3)C(2) cluster gets located within the bulge created by the bridging addend, which is a first step toward release of the internal metal atoms.

Endohedral metallofullerenes (EMFs) have novel structures and properties that are closely associated with the internal metallic species. Benzyl radical additions have been previously shown to form closed-shell adducts by attaching an odd number of addends to open-shell EMFs (such as Sc3 C2 @Ih -C80 ) whereas an even number of groups are added to closed-shell EMFs (for example Sc3 N@Ih -C80 ). Herein we report that benzyl radical addition to the closed-shell La2 @Ih -C80 forms a stable, open-shell monoadduct instead of the anticipated closed-shell bisadduct. Single-crystal X-ray diffraction results show the formation of a stable radical species. In this species, the La-La distance is comparable to the theoretical value of a La-La covalent bond and is shorter than reported values for other La2 @Ih -C80 derivatives, providing unambiguous evidence for the formation of direct La-La bond.

A Sc(2)C(84) isomer, previously assumed to be Sc(2)@C(84), is unambiguously identified as a new carbide cluster metallofullerene Sc(2)C(2)@C(s)(6)-C(82) using both NMR spectroscopy and X-ray crystallography. The (13)C-nuclei signal of the internal C(2)-unit was observed at 244.4 ppm with a 15% (13)C-enriched sample. Temperature-dependent dynamic motion of the internal Sc(2)C(2) cluster is also revealed with NMR spectrometry. Moreover, the chemical property of Sc(2)C(2)@C(s)(6)-C(82) is investigated for the first time using 3-triphenylmethyl-5-oxazolidinone (1) which provides a 1,3-dipolar reagent under heating. Regarding the low cage symmetry of this endohedral which contains 44 types of nonequivalent cage carbons, it is surprising to find that only one monoadduct isomer is formed in the reaction. Single-crystal X-ray results of the isolated pyrrolidino derivative Sc(2)C(2)@C(s)(6)-C(82)N(CH(2))(2)Trt (2) reveal that the addition takes place at a [6,6]-bond junction, which is far from either of the two Sc atoms. Such a highly regioselective addition pattern can be reasonably interpreted by analyzing the frontier molecular orbitals of the endohedral. Electronic and electrochemical investigations reveal that adduct 2 has a larger highest occupied molecular orbital-lowest unoccupied molecular orbital (HOMO-LUMO) gap than pristine Sc(2)C(2)@C(s)(6)-C(82); accordingly, it is more stable.

Endohedral metallofullerenes are promising materials in biomedical and material sciences. In particular, they are of interest as agents for magnetic resonance imaging (MRI), photovoltaic devices, and semimetallic components. The synthesis of chiral endofullerenes represents one step further in the potential use of these carbon allotropes; however, this step has not been addressed so far. In this regard, enantiopure endofullerenes are expected to open new avenues in fields in which chirality is a key issue. Here, the synthesis and characterization of the first chiral endohedral metallofullerenes, namely, chiral bis-adducts of La@C(72), are reported. Eight optically active isomers were obtained by enantioselective 1,3-dipolar cycloaddition of a N-metalated azomethine ylide onto a non-isolated-pentagon rule metallofullerene derivative, La@C(72)(C(6)H(3)Cl(2)), catalyzed by a copper chiral complex. The chiral bis-adducts of La@C(72), isolated by nonchiral HPLC, showed optical purities as high as 98% as revealed by the remarkable positive or negative Cotton effects observed in the circular dichroic spectra.

We describe the regioselective cycloaddition of La(2)@I(h)-C(80) with tetracyanoethylene oxide (TCNEO), which enabled the formation of the corresponding adduct having a tetracyanotetrahydrofuran moiety. X-ray crystallographic analysis revealed that the cycloaddition took place as a [5,6] addition. Along with dynamic swing motion of the metal atoms, the results of this electrochemical study demonstrate that TCNEO addition enhanced the electron-accepting character of La(2)@I(h)-C(80) and that the first reduction potential of the adduct reached -0.21 V versus the ferrocene/ferrocenium couple, which is the lowest value reported for endohedral metallofullerenes and their derivatives to date.

Unambiguous X-ray crystallographic results of the carbene adduct of Sc(2)C(82) reveal a new carbide cluster metallofullerene with the unexpected C(2v)(5)-C(80) cage, that is, Sc(2)C(2)@C(2v)(5)-C(80). More interestingly, DFT calculations and NMR results disclose that the dynamic motion of the internal Sc(2)C(2) cluster depends strongly on temperature. At 293 K, the cluster is fixed inside the cage with two nonequivalent Sc atoms on the mirror plane, thereby leading to C(s) symmetry of the whole molecule. However, when the temperature increases to 413 K, the (13)C and (45)Sc NMR spectra show that the cluster rotates rapidly inside the C(2v)(5)-C(80) cage, featuring two equivalent Sc atoms and weaker metal-cage interactions.

Regioselective cycloaddition of adamantylidene carbene to Pr@C(2v)(9)-C82 affords the first derivative of praseodymium-containing metallofullerenes. Single-crystal X-ray crystallographic data in combination with theoretical studies demonstrate that the addition is dictated by the single metal ion encapsulated inside the fullerene cage.

Photochemical reactions of C(60) with siliranes (1a-d) afford adducts of four types (2a-5b) as carbosilylated and hydrosilylated C(60) derivatives. Characterization of these adducts was conducted using MS, UV, NMR spectroscopy, and single-crystal X-ray analyses. In particular, the first example of the crystal structure of a closed 1,2-adduct at the 5,6-ring junction of the C(60) cage is provided by single-crystal X-ray analysis of 3b. Electrochemical analyses also revealed unique redox properties of the products 2b-5b, which depend on the regiochemistry of the functionality, in addition to the substituents on the C(60) cage. Theoretical calculations offer bases for the experimentally observed redox properties and relative stabilities of the silylated products.

The decakis(trifluoromethyl)fullerene C(1)-C(70)(CF(3))(10), in which the CF(3) groups are arranged on a para(7)-meta-para ribbon of C(6)(CF(3))(2) edge-sharing hexagons, and which has now been prepared in quantities of hundreds of milligrams, was reacted under standard Bingel-Hirsch conditions with a bis-pi-extended tetrathiafulvalene (exTTF) malonate derivative to afford a single exTTF(2)-C(70)(CF(3))(10) regioisomer in 80 % yield based on consumed starting material. The highly soluble hybrid was thoroughly characterized by using 1D (1)H, (13)C, and (19)F NMR, 2D NMR, and UV/Vis spectroscopy; matrix-assisted laser desorption ionization (MALDI) mass spectrometry; and electrochemistry. The cyclic voltammogram of the exTTF(2)-C(70)(CF(3))(10) dyad revealed an irreversible second reduction process, which is indicative of a typical retro-Bingel reaction; whereas the usual phenomenon of exTTF inverted potentials (E1ox>E2ox), resulting in a single, two-electron oxidation process, was also observed. Steady-state and time-resolved photolytic techniques demonstrated that the C(1)-C(70)(CF(3))(10) singlet excited state is subject to a rapid electron-transfer quenching. The resulting charge-separated states were identified by transient absorption spectroscopy, and radical pair lifetimes of the order of 300 ps in toluene were determined. The exTTF(2)-C(70)(CF(3))(10) dyad represents the first example of exploitation of the highly soluble trifluoromethylated fullerenes for the construction of systems able to mimic the photosynthetic process, and is therefore of interest in the search for new materials for photovoltaic applications.

Purely semiconducting single-walled carbon nanotubes (s-SWNTs) with a narrow diameter distribution have been produced for HiPco SWNTs. A facile technique combining microwave irradiation with mixed-acid-assisted dispersion has proven efficient for enrichment of s-SWNTs. Using this process, both electronic type-dependent (metallic versus semiconducting) and diameter-dependent separation of SWNTs were simultaneously realized. By mildly controlling experimental conditions, metallic single-walled carbon nanotubes (m-SWNTs) and smaller s-SWNTs were preferentially removed, yielding purely diameter distribution-narrowed semiconducting nanotubes. Furthermore, the chemical structure of carbon nanotubes was restored after thermal annealing, as indicated by Raman and UV absorption investigations. The dual effects of semiconducting enrichment and diameter distribution narrowing, along with facile procedures, make this method promising for large-scale separation of SWNTs and therefore for industrial fabrication of nanotube-based electronics.

The two regioisomers of endohedral pyrrolidinodimetallofullerenes M(2)@I(h)-C(80)(CH(2))(2)NTrt (M = La, Ce; Trt = trityl) were synthesized, isolated, and characterized. X-ray crystallographic analyses of [6,6]-La(2)@I(h)-C(80)(CH(2))(2)NTrt and [6,6]-Ce(2)@I(h)-C(80)(CH(2))(2)NTrt revealed that the encapsulated metal atoms are located at the slantwise positions on the mirror plane that parallels the pyrrolidine ring. Paramagnetic NMR analyses of [6,6]- and [5,6]-Ce(2)@I(h)-C(80)(CH(2))(2)NTrt were also carried out to clarify the metal positions. As for the [6,6]-adduct, the metal positions obtained by paramagnetic NMR analysis agree well with the X-ray structure. In contrast, paramagnetic NMR analysis of the [5,6]-adduct showed that the two Ce atoms are collinear with the pyrrolidine ring. We also compared the observed paramagnetic effects of the pyrrolidinodimetallofullerenes with those of other cerium-encapsulating fullerene derivatives such as bis-silylated Ce(2)@I(h)-C(80) and a carbene adduct of Ce(2)@I(h)-C(80). We found that the metal positions can be explained by the electrostatic potential maps of the corresponding [6,6]- and [5,6]-adducts of [I(h)-C(80)(CH(2))(2)NTrt](6-). These findings clearly show that metal positions inside fullerene cages can be controlled by means of the addition positions of the addends. In addition, the radical anions of the pyrrolidinodimetallofullerenes were prepared by bulk controlled-potential electrolysis and characterized by X-band EPR spectral study.

Carbon nanotubes (CNTs) are anticipated as an important new material for use in nanotechnology applications because of their excellent mechanical and electrical properties. For their development, a highly stable dispersion of debundled CNTs is indispensable. Herein we present a new method to enhance dispersibility of single-walled carbon nanotubes (SWNTs) with proteins using alcohols as co-solvents. Addition of fluoroalcohols in solution increased the SWNT dispersion by more than one order of magnitude without protein denaturation. Enhancement of SWNT dispersion through addition of alcohols was attributed to the decreased hydrophobic interaction among SWNTs. This novel approach enables us to produce biofunctional CNTs such as one-dimensional nanobiosensors and drug carriers that can penetrate cells.

Herein we show the synthesis and characterization of the second known Ce(2)@C(80) isomer. A (13)C NMR spectroscopic study revealed that the structure of the second isomer has D(5h) symmetry. Paramagnetic NMR spectral analysis and theoretical calculation display that the encapsulated Ce atoms circulate two-dimensionally along a band of ten contiguous hexagons inside a D(5h)-C(80) cage, which is in sharp contrast to the three-dimensional circulation of two Ce atoms in an I(h)-C(80) cage. The electronic properties were revealed by means of electrochemical measurements. The D(5h) isomer of Ce(2)@C(80) has a much smaller HOMO-LUMO gap than cluster fullerenes (M(3)N@C(80), M = Sc, Tm, and Lu) with the same D(5h)-C(80) cages. The chemical reactivity was investigated by using disilirane as a chemical probe. The high thermal reactivity toward 1,1,2,2-tetramesityl-1,2-disilirane is consistent with the trends of the redox potentials and the lower LUMO level of the D(5h) isomer of Ce(2)@C(80) compared with that of C(60).

Many insoluble metallofullerenes are called missing metallofullerenes. These metallofullerenes have not yet been isolated despite their detection in raw soot using mass spectrometry. They have been anticipated to show unique structures and properties that differ considerably from those of soluble metallofullerenes. Herein, a missing metallofullerene, La@C(80), was extracted and isolated as a derivative, La@C(80)(C(6)H(3)Cl(2)). The structure of La@C(80)(C(6)H(3)Cl(2)) was determined unambiguously using single-crystal X-ray crystallographic analysis. In fact, La@C(80) has a novel C(80) cage, C(2v)-C(80) (80:3), which has not been reported. The calculation of La@C(80) with C(2v)-C(80) (80:3) cage shows that La@C(2v)-C(80) (80:3) has a radical character as well as small ionization potential (Ip) and electron affinity (Ea), which are indicative of high reactivity. On the basis of those results, La@C(2v)-C(80) (80:3) is expected to interact strongly and bond with amorphous carbon or other fullerenes in soot that is made insoluble in common organic solvents.

A new chemical method has been applied to isolate isomers of C78. The photochemical cycloaddition of a mixture of C78(D3) and C78(C2v) with a disilirane affords only the mono-adduct of the C2v isomer of C78. Its counterpart does not give the mono-adduct with a disilirane. The D3 isomer itself and the adduct of the C2v isomer of C78 with a disilirane are easily separated by a HPLC procedure. A facile oxidative desilylation of the adduct takes place, resulting in the formation of pristine C78(C2v). In this context, the separation and isolation of the two isomers of C78 were successfully accomplished by using a silylation–oxidative desilylation process.

Fullerenes violating the isolated pentagon rule (IPR) are only obtained in the form of their derivatives. Since the [5,5]-bond carbons are highly reactive, they are easily attacked by reagents to release the bond strains. Non-IPR endohedral metallofullerenes, however, still have unsaturated sp (2) carbons at the [5,5] bond junctions, which allow their chemical properties to be probed. In this work, La 2@C 72 was chosen as a representative non-IPR metallofullerene, since it has been experimentally proposed to have either the #10611 or #10958 non-IPR cage structure ( J. Am. Chem. Soc. 2003, 125, 7782 ), while theoretical calculations have suggested that the #10611 cage is more stable ( J. Phys. Chem. A 2006, 110, 2231 ). La 2@C 72 was modified by photolytic reaction with the carbene reagent 2-adamantane-2,3-[3H]-diazirine. Six isomers of adamantylidene monoadducts were isolated and characterized using various kinds of measurements, including high-performance liquid chromatography, matrix-assisted laser desorption ionization mass spectrometry, UV-vis-near-infrared spectroscopy, cyclic voltammetry, differential-pulse voltammetry, (13)C NMR spectroscopy, and single-crystal X-ray diffraction. Electronic spectra and electrochemical studies revealed that the essential electronic structures of La 2@C 72 are retained in the six isomers and the adamantylidene group acts as a weak electron-donating group toward La 2@C 72. X-ray structural results unambiguously elucidated that La 2@C 72 has the #10611 chiral cage (i.e., D 2 symmetry) with two pairs of fused pentagons at each pole of the cage and that the two La atoms reside close to the two fused-pentagon pairs. On the basis of these results and theoretical calculations, it is concluded that the fused-pentagon sites are very reactive toward carbene but that the carbons forming the [5,5] junctions are less reactive than the adjacent ones; this confirms that these carbons interact strongly with the encaged metals and thus are stabilized by them.

The reaction mechanism and origin of regioselectivity of (2 + 2) cycloadditions of benzyne to endohedral metallofullerenes M3N@C80 (M = Sc, Y) were investigated with density functional calculations. The reaction was demonstrated to follow a diradical mechanism rather than a carbene mechanism, in which the formation of the diradical intermediate is the rate-determining step. Through rotation of benzyne moiety on the fullerene surface, the diradical intermediate on 566 site could isomerize to two new diradical intermediates which give rise to two distinct [5,6] and [6,6] benzoadducts, respectively. However, the diradical intermediate on 666 site only produces the [6,6] benzoadduct. The nature of the endohedral cluster not only influences the regioselectivity, but also determines the cycloadduct geometry. For Sc3N@C80, the [5,6] benzoadduct is preferred kinetically and thermodynamically, whereas in the case of Y3N@C80, both [5,6] and [6,6] benzoadducts are favorable. In contrast to closed-cage benzoadducts of Sc3N@C80, Y3N@C80 affords open-cage benzoadducts, making it the first example that the endohedral cluster could alter cycloadducts from the closed cage to open cage. With further analysis, it is revealed that the origin of regioselectivity results from the local strain energy of the fullerene cage.

The reaction of La@C82 with diethyl bromomalonate in the presence of base (the Bingel reaction) generated five monoadducts which have been fully characterized. It was found that four of them (mono-A, -B, -C, and -D) are ESR-inactive, suggesting singly bonded regioisomers. In contrast, the fifth product (mono-E) is ESR-active, indicating that it possesses a cyclic moiety between the appended malonate group and the fullerene cage, analogous to conventional Bingel adducts. The differences in the molecular structures of mono-A, -B, -C, and -E result in varying thermal stabilities and electrochemical behavior. In particular, the singly bonded monoadducts undergo the retro-Bingel reaction either under thermal treatment or during electron transfer on the cyclic voltammetric timescale. However, mono-E shows remarkable thermal stability and perfect reversibility under the same experimental conditions.

Complexation of endohedral metallofullerene La@C(82)-A (1) with macrocyclic compounds, such as 1,4,7,10,13,16-hexaazacyclooctadecane (2), 1,4,7,10,13,16-hexamethyl-1,4,7,10,13,16-hexaazacyclooctadecane (3), mono-aza-18-crown-6 ether (4), 18-crown-6 ether (5), and p-tert-butylcalix[n]arenes (n = 4-8, 6-10), for the first time is examined. Among them, 1 forms a complex with azacrown ethers 2-4 while accompanying the electron transfer between them. This is characteristic of endohedral metallofullerene and caused by its low reduction potential. Activation energies, DeltaG(et), for the electron transfer from 2-4 to 1 are 4.6, 2.8, and 11 kcal/mol, respectively. These small DeltaG(et) values indicate that the electron transfer from the azacrown ethers to 1 is facile in the ground state. Furthermore, the selective isolation of lanthanum endohedral metallofullerenes from the extracts of soot is accomplished by utilizing the complexation of 1 with 2.