Ball and Socket Assembly of Binary Superatomic Solids Containing

Article pubs.acs.org/IC

Ball and Socket Assembly of Binary Superatomic Solids Containing Trinuclear Nickel Cluster Cations and Fulleride Anions Jessica L. Shott,† Matthew B. Freeman,† Nemah-Allah Saleh,† Daniel S. Jones,† Daniel W. Paley,‡,⊥ and Christopher Bejger*,† †

Department of Chemistry, The University of North Carolina at Charlotte, 9201 University City Boulevard, Charlotte, North Carolina 28223, United States ‡ Department of Chemistry and ⊥Columbia Nano Initiative, Columbia University, New York, New York 10027, United States S Supporting Information *

ABSTRACT: The superlattice structures of hierarchical cluster solids are dictated by short-range interactions between constituent building blocks. Here we show that shape complementary sites, as well as halogen and chalcogen bonding between exposed capping ligands and fullerides, govern the packing arrangement of the resulting binary solids. Four new superatomic solids, [Ni3(μ3-I)2(μ2-dppm)3+](C60•‑) (1·C60), [Ni 3(μ3 -I) 2 (μ2 -dppm)3 +](C70 −) 2 (1·C70 ), [Ni 3(μ 3-Te) 2 (μ2 dppm)3+](C60•‑) (2·C60), and [Ni3(μ3-Te)2(μ2‑dppm)3](C70−)2 (2·C70), (dppm = Ph2PCH2PPh2) were prepared and crystallized from solution. All four compounds were characterized by single crystal X-ray diffraction, IR spectroscopy, and SQUID magnetometry. Charge transfer between the molecular clusters is confirmed via optical spectroscopy and structural data. Compounds 1·C60 and 2·C60 are paramagnetic and 100 times more conductive than the constituent cluster precursors. The obtained solids exhibit close contacts, indicative of halogen/chalcogen bonds, between the fulleride anions and the nickel cluster capping ligands (I/Te) in the solid-state.

■

INTRODUCTION Atomically precise clusters such as metal nanoclusters,1 polyoxometalates,2 fullerenes,3 metal chalcogenide molecular clusters,4 and clusterfullerenes5 represent a class of nanoscale building blocks that pair the well-defined structures of molecular precursors with the size of nanoparticles.6 Many of these clusters are regarded as atom-like, owing to their delocalized electronic structures and ability to undergo reversible multielectron transfer while maintaining structural integrity.7 In fact, such clusters have been designated as superatoms due to these electronic features and their nanometer sizes.8 Several crystalline functional materials have been prepared through judicious selection and hierarchical assembly of such monodisperse superatoms.9 The physical properties of hierarchical cluster solids are directly related to the arrangement of the superatom ions within the superatomic lattice. However, the structure-directing forces within these solids are still largely unpredictable. Unlike traditional atoms, nanometer-sized cluster precursors are not always spherical in shape. Thus, many factors control their packing in the solid state. Beyond long-range Coulombic attraction, short-range interactions (hydrogen bonding, van der Waals interactions) between charged clusters have been shown to affect the crystal packing in these “intercluster” materials.10 Therefore, each class of cluster building block comes with its own design challenges arising from the wide variety of possible ligand and cluster combinations. © 2017 American Chemical Society

Recently, Nuckolls, Roy, and co-workers reported the synthesis and characterization of binary ionic solids, termed superatomic solids or superatomic crystals, from metal chalcogenide molecular clusters and fullerenes.9a,b,d These crystalline materials form when complementary superatoms in solution undergo spontaneous electron transfer. Specifically, both [Co6Se8(PEt3)6][C60]2 and [Cr6Se8(PEt3)6][C60]2 form hexagonal close packed lattices corresponding to nanoscale CdI2-type arrangements. The close fullerene−fullerene interactions in these structures lead to semiconductor behavior. Conversely, [Ni9Te6(PEt3)8][C60] is a rock-salt lattice that undergoes a ferromagnetic phase transition at low temperatures.9b Ostensibly, the different physical properties of superatomic solids depend on several factors: (1) the arrangement of the fullerenes around the metal chalcogenide clusters, (2) the degree of charge-transfer between the superatoms, and (3) the electronic structure of the metal chalcogenide cluster. Structurally, metal chalcogenide molecular clusters contain mixed-valence metal centers held together by bridging cappingligands (chalcogens) and passivating phosphine ligands.4 The electronic structures of these superatoms are dictated by the electron rich metal-capping-ligand “core”. We noticed that, upon charge-transfer to fullerene, the exposed polarizable Received: May 16, 2017 Published: August 28, 2017 10984

DOI: 10.1021/acs.inorgchem.7b01259 Inorg. Chem. 2017, 56, 10984−10990

Article

Inorganic Chemistry

Figure 1. (a) Space-filling structure of 1·C60 showing the crystal packing looking down the c-axis with a right and b down. The phenyl and methylene groups were removed to clarify the view. (b) Ball and stick view of capping ligand−fulleride interaction. Hydrogen atoms were removed for clarity. The space group contains a mirror plane perpendicular to b and bisecting the trinuclear nickel clusters, and thus fullerides are disordered by symmetry (not shown; see SI). Carbon, black; nickel, red; iodine, purple; phosphorus, orange.

capping ligands appear to act as “coordination sites” for the fulleride counteranions.9 Furthermore, certain phosphine ligands facilitate this coordination by producing shapecomplementary cavities that allow the fullerides to pack closer to the cationic cluster core. Therefore, we expect that providing accessible cavities through rational phosphine ligand and core choice will allow predictive structural control leading to superatomic solids by design. In this work, we used two trinuclear nickel clusters [Ni3(μ3I)2(μ2-dppm)3] (1) and [Ni3(μ3-Te)2(μ2-dppm)3] (2), to prepare superatomic solids with C60 and C70 fullerenes. We chose trinuclear superatoms, with only two exposed capping ligands, as model systems to study the relationship between the number of fulleride “coordination sites” and superatomic-lattice structure. Neutral iodide and telluride capped trinuclear nickel clusters were first synthesized and extensively studied by Kubiak and co-workers.11 These nickel clusters consist of two triple-bridging capping ligands, situated above and below the cyclic trimer of nickel atoms, 180° apart. Furthermore, these capping ligands are sterically unhindered despite the presence of bulky diphosphine ligands surrounding the periphery of the trinuclear center. The selection of bridging (as opposed to chelating) phosphine ligands provides a cavity that is approximately the same size as the diameter of C60 (7.1 Å). We hypothesized that ionic solids prepared using trinuclear superatoms and fullerenes would pack in a ball and socket orientation. Thus, clusters 1 and 2 would coordinate two fulleride anions in a one-dimensional linear arrangement whereas Ni9Te6(PEt3)8 accommodates six in a three-dimensional octahedral arrangement.

■

vis-NIR spectra were obtained using a Varian Cary 5000 UV−vis-NIR spectrophotometer. Conductivity measurements were taken under inert atmosphere using Kiethley 236 Source Measure Unit (SMU) and a homemade device comprised of a thick-walled glass capillary with two metal rods compressed with a small trigger clamp (see Supporting Information Section V). Magnetization data was obtained on a Quantum Design MPMS XL7 SQUID Magnetometer. Data for all compounds was collected on an Agilent Gemini diffractometer using mirror-monochromated Mo Kα radiation. Bright field images were obtained on a Phenom-World Phenom ProX desktop SEM instrument using a 10 kV accelerating voltage. Full synthetic and measurement details are presented in the Supporting Information. Preparation of [Ni3(μ3-I)2(μ2-dppm)3+](C60•‑) (1·C60). Filtered solutions (0.45 μm PTFE syringe filter) of Ni3(μ3-I)2(μ2-dppm)3 (25 mg, 0.016 mmol) dissolved in toluene (5 mL) and C60 (11.4 mg, 0.016 mmol) in 1-methylnaphthalene (5 mL) were layered into a 4-dram vial with a separating layer of toluene (1−2 mL) so that a clear region of toluene between the two precursor solutions remained visible. The vials were left undisturbed for 1 week while the layers slowly diffused resulting in the formation of single crystal rectangular blocks of [Ni3(μ3-I)2(μ2-dppm)3+](C60•‑). The supernatant was decanted from the vial and the crystals were washed with hexanes before they were dried in vacuo for ∼5 h (4.10 mg, 11%). Anal. Calcd for C160H92I2Ni3P6: C, 73.06; H, 3.53. Found: C, 73.31; H, 3.41. Preparation of [Ni3(μ3-I)2(μ2-dppm)3+](C70−)2 (1·C70). Preparation analogous to 1·C60. Ni3(μ3-I)2(μ2-dppm)3 (25 mg, 0.016 mmol), C70 (13.3 mg, 0.016 mmol) (9.5 mg, 26%). Anal. Calcd for C170H92I2Ni3P6: C, 74.24; H, 3.37; I, 9.23. Found: C, 74.40; H, 3.31; I, 9.05. Preparation of [Ni3(μ3-I)2(μ2-dppm)3+](TCNQ•‑) (1·TCNQ). A 0.016 M solution of 7,7,8,8-tetracyanoquinodimethane (TCNQ) (26.0 mg, 0.13 mmol) in toluene (8 mL) was prepared and filtered before use. An aliquot (1 mL) of this solution was added to a filtered solution of 1 (20.0 mg, 0.013 mmol) in toluene (5 mL). The resulting teal solution was shaken gently before it was left undisturbed for 12 h and isolated as dark purple crystalline solid. Preparation of [Ni3(μ3-Te)2(μ2-dppm)3+](C60•‑) (2·C60). Preparation analogous to 1·C60. Ni3(μ3-Te)2(μ2-dppm)3 (25 mg, 0.016 mmol), C60 (11.8 mg, 0.016 mmol) (14.9 mg, 40%). Preparation of [Ni3(μ3-Te)2(μ2-dppm)3+](C70−)2 (2·C70). Preparation analogous to 1·C60. Ni3(μ3-Te)2(μ2-dppm)3 (25 mg, 0.016 mmol), C70 (13.3 mg, 0.016 mmol) (18.8 mg, 48%).

EXPERIMENTAL SECTION

General Considerations. All metal cluster precursors were used as obtained from commercial suppliers as follows: nickel(II) iodide (anhydrous) (Strem, 99.5%), bis(cyclooctadiene)nickel(0) (Ni(COD)2) (Beantown Chemical, 96%), diphenylphosphinomethane (dppm) (Sigma-Aldrich, 97%), tellurium powder (Strem, 99.8%), trin-propylphosphine (PPrn3) (Aldrich, 97%). C60 and C70 fullerene were purchased from Sigma-Aldrich. 7,7,8,8-Tetracyanoquinodimethane (TCNQ) (98%) was used as purchased from Sigma-Aldrich. Solvents were obtained from commercial suppliers and purified before use. All manipulations were performed under inert atmosphere using standard Schlenk techniques or a nitrogen filled glovebox (M-Braun UNIlab Pro SP workstation). IR spectra were obtained neat under ambient conditions on a PerkinElmer Spectrum 100 FT-IR Spectrometer. UV−

■

RESULTS AND DISCUSSION

Trinuclear clusters 1 and 2 are redox active and electron rich, with 52 and 50 cluster valence electrons, respectively. Electrochemically, cluster 1 undergoes two reversible oxidations in 1,1,1-trichloroethane at −0.68 and −0.19 V vs SCE.11a 10985

DOI: 10.1021/acs.inorgchem.7b01259 Inorg. Chem. 2017, 56, 10984−10990

Article

Inorganic Chemistry Cluster 2 also exhibits two reversible oxidations at −0.52 and 0.28 V vs SCE in acetonitrile.11b Based on their superalkali character, we expected 1 and 2 to transfer charge to fullerene acceptors and form ionic solids. Interdiffusion of cluster 1 and C60 solutions for 1 week gives black needle-like crystals suitable for single-crystal X-ray diffraction (SCXRD). The binary solid 1·C60 crystallizes in the orthorhombic space group Pnma and comprises mixed zigzag chains of alternating superatoms along the b-axis (Figure 1a). The fullerene superatoms in each zigzag chain also align down the c-axis. The atom-to-atom distance between adjacent interchain C60 cages is 3.57 Å. The centroidto-centroid distance between the C60 is 10.39 Å. One 1methylnaphthalene and two toluene solvent molecules per formula unit are also incorporated in the lattice. Particularly noteworthy in the crystal structure of 1·C60 are the close contacts between the triple-bridging capping iodides, (μ3-I)2, and the fulleride counteranions in the zigzag chain. Compound 1·C60 is the first superatomic solid with halide capping ligands where each iodide forms short contacts to one fulleride. Specifically, the iodide atoms are centered over the face of the hexagons in adjacent fullerides. The average distance between the exposed capping iodides of 1 and each neighboring C60 is 3.36 Å (Figure 1b). These close contacts fall in the range of reported C−I···π halogen bonding interactions12 and denote an inorganic equivalent to this well-studied electrostatic force.13 Moreover, the angle θ between the centroid of the nickel trimer, capping iodide, and centroid of the Lewis basic hexagonal face is 177.89°, which is close to the commonly observed halogen bonding angle of 180°. Under similar conditions, trinuclear nickel telluride cluster 2 also reduces C60 to produce single crystals that are isostructural to 1·C60 (Figure S16, Supporting Information). Substituting the iodide capping-ligands with bridging telluride atoms yields binary solid 2·C60 with average Te−C60 contacts of 3.53 Å. These short contacts are expected given comparable size and polarizability of telluride and iodide atoms and may represent a form of chalcogen bonding. Chalcogen bonding, a relative of halogen bonding, has been shown to direct crystal packing in the solid state and anion binding in solution.14 The shape complementary phosphine cavities and these capping ligand− fulleride interactions direct the assembly of the two superatoms. It is worth noting that the face-to-face packing observed in 1· C60 and 2·C60 is similar to solid-state supramolecular networks formed between the cubic cluster Pd6Cl12 and π-conjugated molecules. For instance, various planar15a,b and curved15c polynuclear aromatic hydrocarbon molecules, as well as C60 fullerene,10a cocrystallize with Pd6Cl12. The packing structures of the resultant solids contain columns of alternating Pd6Cl12 clusters and aromatic molecules akin to the arrangement of superatoms in 1·C60 and 2·C60. Initial evidence for charge transfer in the new superatomic solids came from the Ni−Ni bond lengths in the nickel trimers. The metal triangle of nickel atoms in 1·C60 is isosceles with two sides of 2.51 Å and one side of 2.46 Å. This is in agreement with previously reported structural data of 1+ and contrasts with the nearly equilateral metal core of the neutral cluster 1.11 The trinuclear metal core in 2 undergoes a similar change in symmetry during the formation of 2·C60. The degree of charge transfer in 1·C60 was further corroborated using infrared (IR) spectroscopy. Specifically, the Tu(4) mode, which occurs at 1429 cm−1 in neutral C60, exhibits a diagnostic 34−40 cm−1 red shift upon transition to the radical anion.16 The Tu(4) mode in 1·C60 was observed at 1387 cm−1, which is consistent with the

values reported for known C60 fullerides and confirms the presence of single-reduced fulleride anions (Figure S4). Additionally, electronic absorption spectra were measured on crystalline samples of 1·C60 and 2·C60 dispersed in KBr and pressed into thin pellets. Both solids show distinct transitions in the NIR centered at 1100 nm, which are not observed in pressed KBr pellets of the precursors and correspond to the radical anion of fullerene C60•‑ (Figures S6−S9).16 Increasing the size of the fullerene from C60 to C70 in 1·C70 and 2·C70 produces binary solids with nearly identical structures as determined by SCXRD (Figures 2 and S14). The a and b

Figure 2. Structure of 1·C70 showing the crystal packing looking down the c-axis with a right and b down. The phenyl and methylene groups were removed to clarify the view. Hydrogen atoms were removed for clarity. The space group contains a mirror plane perpendicular to b and bisecting the trinuclear nickel clusters, and thus the dimer chains are disordered by symmetry (not shown; see SI). Carbon, black; nickel, red; iodine, purple; phosphorus, orange.

axes in 1·C70 and 2·C70 are slightly elongated compared to 1· C60 and 2·C60 due to the larger fullerene superatoms. Furthermore, for 1·C 70 and 2·C 70 , the fullerenes are predominantly (but not exclusively) found in C140 dimers (Figure S1). The C70 radical anions dimerize along the c-axis to form chains of single-bonded dimers (C70−)2. This dimerization is a common characteristic of many ionic C70 fulleride solids.17 Short contacts averaging 3.44 and 3.52 Å between the fullerides and the capping ligands are also present in 1·C70 and 2·C70, respectively. The similarity between these distances and the distances measured in 1·C60 and 2·C60 lead us to believe that there is a stabilizing interaction between the capping ligands and the fullerides. To investigate whether the shape of the acceptor affects this interaction, we cocrystallized 1 with tetracyanoquinodimethane (TCNQ). Solutions of cluster 1 and TCNQ in toluene were mixed and allowed to sit for 48 h. Block-like single crystals were isolated by removal of the supernatant and analyzed by SCXRD. Analysis of the bond lengths in TCNQ and the nickel trimer confirmed full charge transfer to form the ionic solid [1]+[TCNQ]•‑.18 The structure of 1·TCNQ obtained differs considerably from the fullerene-based solids, with mixed stacks of 1 and TCNQ along the b-axis (Figure 3). However, the TCNQ radical anion packs closely to the nickel cluster’s capping iodide, with the shortest contact (3.68 Å) between one iodide and the TCNQ methylidene carbon. The negative charge in singly reduced TCNQ localizes on this carbon, providing additional evidence that the capping iodides behave as charge-assisted halogen bond donors.18b 10986

DOI: 10.1021/acs.inorgchem.7b01259 Inorg. Chem. 2017, 56, 10984−10990

Article

Inorganic Chemistry

Figure 4. Temperature dependencies for effective magnetic moments of (a) 1·C60 and (b) 2·C60, and inverse magnetic susceptibility as a function of temperature for (c) 1·C60 and (d) 2·C60.

The spin-only effective magnetic moment for a material containing two noninteracting S = 1/2 spins is 2.45 μB at 300 K. For a similar system with only one S = 1/2 spin, μeff = 1.73 μB. Superatomic solids 1·C60 and 2·C60 were each expected to show effective magnetic moments in agreement with a two-spin system. However, respective moments of 2.06 and 3.39 μB were obtained at room temperature. This discrepancy can be attributed to the molecular orbital diagrams for the constituent clusters, 1 and 2. The reported electronic structure of 1, determined using MO calculations, contains a fully occupied, double-degenerate HOMO.10a Therefore, the HOMO of the 51 electron, oxidized nickel iodide species 1+ contains one unpaired electron, and the neutral 50 electron nickel telluride cluster 2 contains two unpaired electrons.11 Kubiak and coworkers report that 1+ exhibits no discernible EPR signal due to rapid spin relaxation in which the single unpaired electron occupies a pair of degenerate orbitals.11a This relaxation could cause 1·C60 to behave in a manner more closely related to a system containing one unpaired spin, lowering the observed effective magnetic moment. Removal of one electron to yield odd-electron species 2+ leaves one unpaired electron in the HOMO, which has been corroborated by magnetic susceptibility measurements using the Evans method. However, the effective magnetic moment of 2·C60 is higher than expected for a system of two unpaired electrons. We attribute this difference to the average oxidation states of the individual Ni atoms in the trinuclear cores. Specifically, 1+ consists of Ni33+ whereas the core of 2+ is Ni35+. For this reason, we expect the magnetic moment of 1·C60 to more closely resemble that observed for Ni(I) complexes (2.27 μB) and the moment of 2·C60 to be closer to the value of typical Ni(II) complexes (2.9−3.3 μB).20 Electrical conductivity measurements were performed on pressed pellets of the neutral clusters 1 and 2 as well as of the four binary solids. Ohmic I−V curves were obtained for all samples at room temperature. (Figures S10−13). A plot of current density versus electric field strength was used for direct comparison of material conductivities (Figure 5). Data indicates that the C60 fulleride-based superatomic solids (1·C60 = 2.0 ×

Figure 3. (a) Ball and stick structure of 1·TCNQ showing the crystal packing looking down the b-axis. (b) Ball and stick view of 1·TCNQ ion pair along the a-axis. The phenyl and methylene groups on 1 were removed to clarify the views. Hydrogen atoms were removed for clarity. Carbon, black; nickel, red; nitrogen, blue; iodine, purple; phosphorus, orange.

Although fullerene is diamagnetic, long-range spin coupling has been observed in fulleride salts.19 In fact, binary superatomic solids containing the larger nickel cluster Ni9Te6(PEt3)8 and fulleride show ferromagnetic ordering below 4 K.9a,b The ferromagnetic behavior of these earlier systems prompted us to study the magnetism of the trinuclear nickel cluster based fulleride solids. Superconducting quantum interference device (SQUID) magnetometry data were obtained for 1·C60 and 2·C60. The two fulleride-based superatomic solids adhere to paramagnetic Curie−Weiss behavior (Figure 3). The magnetic susceptibility of these systems can be modeled using the relationship χM(T) = [C/(T − Θ)] + χD + χTIC. A diamagnetic contribution (χD) of −0.0013 emu Oe1− (mol f.u.)−1 was estimated using the Pascal’s constants of the constituent atoms and was used to model both systems. The temperature independent contributions (χTIC) for 1·C60 and 2·C60, 0.0361 emu Oe1− (mol f.u.)−1 and 0.0791 emu Oe1− (mol f.u.)−1 were determined via extrapolation of the high temperature data and subsequent fitting to the expression Xm = C/(T − Θ) + χTIC. A Curie constant (C) of 114.94 emu K Oe1− (mol f.u.)−1and a Weiss constant (Θ) of −15.55 K provided the best fit for the susceptibility of 1·C60. For 2·C60, a good fit was obtained with C = 2941.12 emu K Oe1− (mol f.u.)−1 and Θ = −5.0 K. Plots of inverse molar magnetic susceptibility indicate that 1·C60 and 2· C60 exhibit weak antiferromagnetic interactions based on the slightly positive y-intercept (Figure 4). This is corroborated by the small negative Weiss constants and suggests that individual molecular clusters act as isolated magnetic moments with no long-range ordering. 10987

DOI: 10.1021/acs.inorgchem.7b01259 Inorg. Chem. 2017, 56, 10984−10990

Inorganic Chemistry

■

Article

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.7b01259. Synthesis, crystal data, IR, UV−vis-NIR, SQUID magnetometry, and I/V data (PDF) Accession Codes

CCDC 1548914−1548917 and 1550196 contain the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/ cif, or by emailing [email protected], or by contacting The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033.

■

Figure 5. Electrical properties of pressed pellets. Plots of current density versus electric field strength (J−E curves) for 1 and 2 as well as their corresponding C60 and C70 superatomic solids at 297 K.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Christopher Bejger: 0000-0001-9263-5414

10−6 S cm−1; 2·C60 = 8.0 × 10−7 S cm−1) are nearly 2 orders of magnitude more conductive than their superatom precursors 1 (2.0 × 10−8 S cm−1) and 2 (1.0 × 10−8 S cm−1). These increases in conductivity can be explained by the presence of linear fulleride chains in both 1·C60 and 2·C60. However, the centroid-to-centroid distances between adjacent fullerides (10.39 Å) in 1·C60 and 2·C60 are significantly farther apart than those reported in highly conductive ionic fulleride materials.21 Additionally, the C70 fulleride salts, 1·C70 and 2· C70, are more conductive than their constituent nickel clusters, both with electrical conductivity values of 1.0 × 10−7 S cm−1. These solids are less conductive than the analogous C60 materials. This decrease is likely due to the dimerization of the C70 radical anions in 1·C70 and 2·C70.

Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We gratefully acknowledge Dr. David Shultz at NCSU for assistance with SQUID magnetometry measurements and Dr. Michael Walter for help with conductivity measurements. J.S. thanks the Thomas D. Walsh Graduate Fellowship for support. This work was supported, in part, by funds provided by the University of North Carolina at Charlotte.

■

■

CONCLUSION Four new binary superatomic solids were synthesized from trinuclear nickel cluster and fullerene precursors. The fulleride anions pack in shape-complementary cavities provided by the passivating phosphine ligands, resulting in ball and socket arrangements. All four solids contain close contacts between the capping ligands (I/Te) and the Lewis basic fulleride anions. These contacts are consistent with halogen and chalcogen bonding interactions between the superatomic building blocks and emphasize the significance of directional noncovalent interactions in the assembly of hierarchical cluster solids. Furthermore, the physical properties of these materials differ from those observed for other metal chalcogenide-fulleride superatomic solids. Our C60 based solids, 1·C60 and 2·C60, are paramagnetic and exhibit ohmic conductivity at room temperature. We have further demonstrated that the properties of this class of solids can be tuned by manipulating the crystalline arrangement of the fulleride anions. Therefore, the physical properties of these binary superatomic solids can be influenced through selection of metal cluster precursors that contain shape complementary pockets for fullerene coordination and accessible capping ligand sites. Further investigations into the structure directing forces within these assemblies may lead to the synthesis of hierarchical materials with increased structural control.

REFERENCES

(1) Jin, R.; Zeng, C.; Zhou, M.; Chen, Y. Atomically Precise Colloidal Metal Nanoclusters and Nanoparticles: Fundamentals and Opportunities. Chem. Rev. 2016, 116, 10346−10413. (2) (a) Long, D. L.; Tsunashima, R.; Cronin, L. Polyoxometalates: Building Blocks for Functional Nanoscale Systems. Angew. Chem., Int. Ed. 2010, 49, 1736−1758. (b) Izzet, G.; Abécassis, B.; Brouri, D.; Piot, M.; Matt, B.; Serapian, S. A.; Bo, C.; Proust, A. Hierarchical SelfAssembly of Polyoxometalate-Based Hybrids Driven by Metal Coordination and Electrostatic Interactions: From Discrete Supramolecular Species to Dense Monodisperse Nanoparticles. J. Am. Chem. Soc. 2016, 138, 5093−5099. (c) Song, Y.-F.; Tsunashima, R. Recent Advances on Polyoxometalate-Based Molecular and Composite Materials. Chem. Soc. Rev. 2012, 41, 7384−7402. (d) Miras, H. N.; Yan, J.; Long, D.-L.; Cronin, L. Engineering Polyoxometalates with Emergent Properties. Chem. Soc. Rev. 2012, 41, 7403−7430. (3) (a) Pekker, S.; Kováts, E.; Oszlányi, G.; Bényei, G.; Klupp, G.; Bortel, G.; Jalsovszky, I.; Jakab, E.; Borondics, F.; Kamarás, K.; et al. Rotor-Stator Molecular Crystals of Fullerenes with Cubane. Nat. Mater. 2005, 4, 764−767. (b) Kong, X. J.; Long, L. S.; Zheng, Z.; Huang, R.-B.; Zheng, L. S. Keeping the Ball Rolling: Fullerene-like Molecular Clusters. Acc. Chem. Res. 2010, 43, 201−209. (4) (a) Steigerwald, M. L. Clusters as Small Solids. Polyhedron 1994, 13, 1245−1252. (b) Degroot, M. W.; Corrigan, J. F. High Nuclearity Clusters: Metal−Chalcogenide Polynuclear Complexes. Compr. Coord. Chem. II 2004, 7, 231−261. (c) Saito, T.; Imoto, H. Chalcogenide Cluster Complexes of Chromium, Molybdenum, Tungsten, and Rhenium. Bull. Chem. Soc. Jpn. 1996, 69, 2403−2417. (d) Lin, Z.; 10988

DOI: 10.1021/acs.inorgchem.7b01259 Inorg. Chem. 2017, 56, 10984−10990

Article

Inorganic Chemistry

Ed. 2008, 47, 2256−2259. (f) Turkiewicz, A.; Paley, D. W.; Besara, T.; Elbaz, G.; Pinkard, A.; Siegrist, T.; Roy, X. Assembling Hierarchical Cluster Solids with Atomic Precision. J. Am. Chem. Soc. 2014, 136, 15873−15876. (g) Konarev, D. V.; Kovalevsky, A. Y.; Khasanov, S. S.; Saito, G.; Lopatin, D. V.; Umrikhin, A. V.; Otsuka, A.; Lyubovskaya, R. N. Synthesis, Crystal Structures, Magnetic Properties and Photoconductivity of C60 and C70 Complexes with Metal Dialkyldithiocarbamates M(R2dtc)x, where M = CuII, CuI, AgI, ZnII, CdII, HgII, MnII, NiII, and PtII; R = Me, Et, and nPr. Eur. J. Inorg. Chem. 2006, 2006, 1881−1895. (11) (a) Morgenstern, D. A.; Ferrence, G. M.; Washington, J.; Henderson, J. I.; Rosenhein, L.; Heise, J. D.; Fanwick, P. E.; Kubiak, C. P. A Class of Halide-Supported Trinuclear Nickel Clusters [Ni3(μ3L)(μ3-X)(μ2-dppm)3]n+ (L = I−, Br−, CO, CNR; X= I−, Br−; n = 0, 1; dppm = Ph2PCH2PPh2): Novel Physical Properties and the Fermi Resonance of Symmetric μ3-η1 Bound Isocyanide Ligands. J. Am. Chem. Soc. 1996, 118, 2198−2207. (b) Ferrence, G. M.; Fanwick, P. E.; Kubiak, C. P. A Telluride Capped Trinuclear Nickel Cluster [Ni3(μ3-Te)2(μ2-PPh2CH2PPh2)3]n+ with Four Accessible Redox States (n = −1, 0, 1, 2). Chem. Commun. 1996, 3, 1575−1576. (12) (a) Cao, J.; Yan, X.; He, W.; Li, X.; Li, Z.; Mo, Y.; Liu, M.; Jiang, Y.-B. C−I···π Halogen Bonding Driven Supramolecular Helix of Bilateral N -Amidothioureas Bearing β-Turns. J. Am. Chem. Soc. 2017, 139, 6605. (b) Prasanna, M. D.; Guru Row, T. N. C−Halogen···π Interactions and Their Influence on Molecular Conformation and Crystal Packing: A Database Study. Cryst. Eng. 2000, 3, 135−154. (13) Cavallo, G.; Metrangolo, P.; Milani, R.; Pilati, T.; Priimagi, A.; Resnati, G.; Terraneo, G. The Halogen Bond. Chem. Rev. 2016, 116, 2478−260. (14) (a) Fanfrlík, J.; Přad́ a, A.; Padělková, Z.; Pecina, A.; Machácě k, J.; Lepšík, M.; Holub, J.; Růzǐ čka, A.; Hnyk, D.; Hobza, P. The Dominant Role of Chalcogen Bonding in the Crystal Packing of 2D/ 3D Aromatics. Angew. Chem., Int. Ed. 2014, 53, 10139−10142. (b) Gsänger, M.; Kirchner, E.; Stolte, M.; Burschka, C.; Stepanenko, V.; Pflaum, J.; Würthner, F. High-Performance Organic Thin-Film Transistors of J-Stacked Squaraine Dyes. J. Am. Chem. Soc. 2014, 136, 2351−2363. (c) Garrett, G. E.; Gibson, G. L.; Straus, R. N.; Seferos, D. S.; Taylor, M. S. Chalcogen Bonding in Solution: Interactions of Benzotelluradiazoles with Anionic and Uncharged Lewis Bases. J. Am. Chem. Soc. 2015, 137, 4126−4133. (d) Ho, P. C.; Szydlowski, P.; Sinclair, J.; Elder, P. J. W.; Kübel, J.; Gendy, C.; Lee, L. M.; Jenkins, H.; Britten, J. F.; Morim, D. R.; et al. Supramolecular Macrocycles Reversibly Assembled by Te···O Chalcogen Bonding. Nat. Commun. 2016, 7, 11299. (e) Benz, S.; Macchione, M.; Verolet, Q.; Mareda, J.; Sakai, N.; Matile, S. Anion Transport with Chalcogen Bonds. J. Am. Chem. Soc. 2016, 138, 9093−9096. (f) Garrett, G. E.; Carrera, E. I.; Seferos, D. S.; Taylor, M. S. Anion Recognition by a Bidentate Chalcogen Bond Donor. Chem. Commun. 2016, 52, 9881−9884. (g) Lim, J. Y. C.; Marques, I.; Thompson, A. L.; Christensen, K. E.; Felix, V.; Beer, P. D. Chalcogen Bonding Macrocycles and [2]Rotaxanes for Anion Recognition. J. Am. Chem. Soc. 2017, 139, 3122. (15) (a) Olmstead, M. M.; Wei, P.; Balch, A. L. Solid-State Architectures of Aggregates of the Cubic Cluster [Pd6Cl12] with Polynuclear Aromatic Hydrocarbons. Chem. - Eur. J. 1999, 5, 3136− 3142. (b) Olmstead, M. M.; Wei, P.; Ginwalla, A. S.; Balch, A. L. Bis (Benzonitrile) Palladium (II) Dihalides: Structures and Cocrystallization of the Cubic Cluster Pd6Cl12 with (E)-Stilbene and with Bis(Benzonitrile) Palladium (II) Dichloride. Inorg. Chem. 2000, 39, 4555−4559. (c) Spisak, S. N.; Filatov, A. S.; Petrukhina, M. A. Palladium π-Adduct of Corannulene. J. Organomet. Chem. 2011, 696, 1228−1231. (16) (a) Reed, C. A.; Bolskar, R. D. Discrete Fulleride Anions and Fullerenium Cations. Chem. Rev. 2000, 100, 1075−1120. (17) (a) Konarev, D. V.; Khasanov, S. S.; Vorontsov, I. I.; Saito, G.; Antipin, Y. The Formation of a Single-Bonded (C70−)2 Dimer in a New Ionic Multicomponent Complex of Cyclotriveratrylene. Chem. Commun. 2002, 2, 2548−2549. (b) Konarev, D. V.; Kuzmin, A. V.; Simonov, S. V.; Khasanov, S. S.; Otsuka, A.; Yamochi, H.; Saito, G.; Lyubovskaya, R. N. Ionic Compound Containing Iron Phthalocyanine

Williams, I. D. Structure and Bonding in Face- and Edge- bridged Octahedral Transition Metal Clusters. Polyhedron 1996, 15, 3277− 3287. (5) (a) Yang, S.; Liu, F.; Chen, C.; Jiao, M.; Wei, T. Fullerenes Encaging Metal ClustersClusterfullerenes. Chem. Commun. 2011, 47, 11822. (b) Stevenson, S.; Mackey, M. A.; Stuart, M. A.; Phillips, J. P.; Easterling, M. L.; Chancellor, C. J.; Olmstead, M. M.; Balch, A. L. A Distorted Tetrahedral Metal Oxide Cluster inside an Icosahedral Carbon Cage. Synthesis, Isolation, and Structural Characterization of Sc4(μ3-O)2-@Ih-C80. J. Am. Chem. Soc. 2008, 130, 11844−11845. (c) Dunsch, L.; Yang, S.; Zhang, L.; Svitova, A.; Oswald, S.; Popov, A. A. Metal Sulfide in a C82 Fullerene Cage: A New Form of Endohedral Clusterfullerenes. J. Am. Chem. Soc. 2010, 132, 5413−5421. (d) Popov, A. A.; Chen, N.; Pinzón, J. R.; Stevenson, S.; Echegoyen, L. A.; Dunsch, L. Redox-Active Scandium Oxide Cluster inside a Fullerene Cage: Spectroscopic, Voltammetric, Electron Spin Resonance Spectroelectrochemical, and Extended Density Functional Theory Study of Sc4O2@C80 and Its Ion Radicals. J. Am. Chem. Soc. 2012, 134, 19607−19618. (6) (a) Claridge, S. A.; Castleman, A. W.; Khanna, S. N.; Murray, C. B.; Sen, A.; Weiss, P. S. Cluster-Assembled Materials. ACS Nano 2009, 3, 244−255. (b) Cademartiri, L.; Kitaev, V. On the Nature and Importance of the Transition Between Molecules and Nanocrystals: Towards a Chemistry Of “Nanoscale Perfection. Nanoscale 2011, 3, 3435−3446. (7) Welch, E. J.; Long, J. R. Atomlike Building Units of Adjustable Character: Solid-State and Solution Routes to Manipulating Hexanuclear Transition Metal Chalcohalide Clusters. Prog. Inorg. Chem. 2005, 54, 1. (8) (a) Reber, A. C.; Khanna, S. N. Superatoms: Electronic and Geometric Effects on Reactivity. Acc. Chem. Res. 2017, 50, 255−263. (b) Chauhan, V.; Reber, A. C.; Khanna, S. N. Metal Chalcogenide Clusters with Closed Electronic Shells and the Electronic Properties of Alkalis and Halogens. J. Am. Chem. Soc. 2017, 139, 1871−1877. (c) Tomalia, D. A.; Khanna, S. N. Chem. Rev. 2016, 116, 2705−2774. (d) Champsaur, A. M.; Velian, A.; Paley, D. W.; Choi, B.; Roy, X.; Steigerwald, M. L.; Nuckolls, C. Building Diatomic and Triatomic Superatom Molecules. Nano Lett. 2016, 16, 5273−5277. (9) (a) Roy, X.; Lee, C.-H.; Crowther, A. C.; Schenck, C. L.; Besara, T.; Lalancette, R.; Siegrist, T.; Stephens, P. W.; Brus, L. E.; Kim, P.; et al. Nanoscale Atoms in Solid-State Chemistry. Science 2013, 341, 157−160. (b) Lee, C.; Liu, L.; Bejger, C.; Turkiewicz, A.; Goko, T.; Arguello, C. J.; Frandsen, B. A.; Cheung, S. C.; Medina, T.; Munsie, T. J. S.; et al. Ferromagnetic Ordering in Superatomic Solids. J. Am. Chem. Soc. 2014, 136, 16926. (c) Chauhan, V.; Sahoo, S.; Khanna, S. N. Ni9Te6(PEt3)8•C60 Is a Superatomic Superalkali Superparamagnetic Cluster Assembled Material (S3-CAM). J. Am. Chem. Soc. 2016, 138, 1916−1921. (d) Ong, W.-L.; O’Brien, E.; Dougherty, P. S. M.; Paley, D.; Higgs, C. F. I.; McGaughey, A. J.; Malen, A. J.; Roy, X. Orientational Order Controls Crystalline and Amorphous Thermal Transport in Superatomic Crystals. Nat. Mater. 2016, 16, 83−88. (10) (a) Olmstead, M. M.; Ginwalla, A. S.; Noll, B. C.; Tinti, D. S.; Balch, A. L. Supramolecular Aggregation of Pd6Cl12, a Cluster of Comparable Size to a Fullerene, with Aromatic Donors and with C60. J. Am. Chem. Soc. 1996, 118, 7737−7745. (b) Schulz-Dobrick, M.; Jansen, M. Structure-Directing Effects in the Supramolecular Intercluster Compound [Au9(PPh3)8]2[V10O28H3]2:Long-Range versus Short-Range Bonding Interactions. Inorg. Chem. 2007, 46, 4380− 4382. (c) Gruber, F.; Schulz-Dobrick, M.; Jansen, M. StructureDirecting Forces in Intercluster Compounds of Cationic [Ag14(C CtBu)12Cl]+ Building Blocks and Polyoxometalates: Long-Range versus Short-Range Bonding Interactions. Chem. - Eur. J. 2010, 16, 1464−1469. (d) Choi, B.; Yu, J.; Paley, D. W.; Trinh, M. T.; Paley, M. V.; Karch, J. M.; Crowther, A. C.; Lee, C. H.; Lalancette, R. A.; Zhu, X. Y.; Kim, P.; Steigerwald, M. L.; Nuckolls, C.; Roy, X. van der Waals Solids from Self-Assembled Nanoscale Building Blocks. Nano Lett. 2016, 16, 1445−1449. (e) Schulz-Dobrick, M.; Jansen, M. Intercluster Compounds Consisting of Gold Clusters and Fullerides: [Au7(PPh3)7]C60•THF and [Au 8(PPh3)8](C60)2. Angew. Chem., Int. 10989

DOI: 10.1021/acs.inorgchem.7b01259 Inorg. Chem. 2017, 56, 10984−10990

Article

Inorganic Chemistry (FeIPc)− Anions and (C70−)2 Dimers. Optical and Magnetic Properties of (FeIPc)− in the Solid State. Dalton Trans. 2012, 41, 13841−13847. (18) (a) Kistenmacher, T. J.; Emge, T. J.; Bloch, A. N.; Cowan, D. O. Structure of the Red, Semiconducting Form of 4,4′,5,5′-Tetramethyl△ 2,2 ′ -bi-1,3-diselenole-7,7,8,8-Tetracyano-p-quinodimethane, TMTSF-TCNQ. Acta Crystallogr., Sect. B: Struct. Crystallogr. Cryst. Chem. 1982, 38, 1193−1199. (b) Ward, M. D.; Johnson, D. C. Electrocrystallization and Structural and Physical Properties of ChargeTransfer Complexes Derived from [(η6-C6Me6)2M]2+ (M = Fe, Ru) and TCNQ (TCNQ = Tetracyanoquinodimethane. Inorg. Chem. 1987, 26, 4213−4227. (19) Lappas, A.; Prassides, K.; Vavekis, K.; Arcon, D.; Blinc, R.; Cevc, P.; Amato, A.; Feyerherm, R.; Gygax, F. N.; Schenck, A. Spontaneous Magnetic Ordering in the Fullerene Charge-Transfer Salt (TDAE)C60. Science 1995, 267, 1799−1802. (20) (a) Poulten, R. C.; Page, M. J.; Algarra, A. G.; Le Roy, J. J.; López, I.; Carter, E.; Llobet, A.; Macgregor, S. A.; Mahon, M. F.; Murphy, D. M.; Murugesu, M.; Whittlesey, M. K. Synthesis, Electronic Structure, and Magnetism of [Ni(6-Mes)2]+: A Two-Coordinate Nickel(I) Complex Stabilized by Bulky N-Heterocyclic Carbenes. J. Am. Chem. Soc. 2013, 135, 13640−13643. (b) Holland, P. L.; Cundari, T. R.; Perez, L. L.; Eckert, N. A.; Lachicotte, R. J. Electronically Unsaturated Three-Coordinate Chloride and Methyl Complexes of Iron, Cobalt, and Nickel. J. Am. Chem. Soc. 2002, 124, 14416−14424. (21) (a) Konarev, D. V.; Khasanov, S. S.; Otsuka, A.; Maesato, M.; Saito, G.; Lyubovskaya, R. N. A Two-Dimensional Organic Metal Based on Fullerene. Angew. Chem., Int. Ed. 2010, 49, 4829−4832. (b) Kromer, A.; Wedig, U.; Roduner, E.; Jansen, M.; Amsharov, K. Y. Counterintuitive Anisotropy of Electron Transport Properties in KC60(THF)5·2THF Fulleride. Angew. Chem., Int. Ed. 2013, 52, 12610.

10990

DOI: 10.1021/acs.inorgchem.7b01259 Inorg. Chem. 2017, 56, 10984−10990