Abstract
Organic neutral radicals have been predicted to exhibit various electronic functions such as electrical conduction. However, most organic neutral radicals are insulators, because they cannot form sufficient intermolecular interactions due to the bulky substituent groups required for stabilization. Here we report that one-dimensional assemblies of carbon-centered neutral π-radicals, namely 4,8,12,trioxotriangulene derivatives, possess effective conducting pathways as a result of strong intermolecular interactions based on two-electron-multicenter bonding. The columns of trioxotriangulene derivatives with weak π-dimerization and uniform π-stacking exhibit semiconducting behaviors, with high conductivities of ~10−3 S cm−1 as a single component purely organic molecular system. We exploit this general tendency to form one-dimensional assemblies, and the large 25 π-electronic system with a robust condensed polycyclic structure, to obtain mixed-valence salts consisting of neutral radicals and the corresponding anionic species with a higher room-temperature conductivity of 1–125 S cm−1.
Similar content being viewed by others
Introduction
Organic conductors have been a focus in the development of molecule-based electronic materials for more than 60 years1. Usually, organic compounds are electrical insulators, because the highest occupied molecular orbital (HOMO) band is completely filled and the electron must overcome the large HOMO–LUMO (lowest unoccupied MO) energy gap. The formation of a conduction pathway is also necessary through orbital overlaps between molecules; however, organic molecules are electronically discrete in the solid state. This has been overcome in plural component charge-transfer (CT) complexes and salts, where one-dimensional π-stacking columns with large intermolecular π-orbital overlaps provide a conduction pathway. Semiconducting perylene–bromine salt2 (room temperature conductivity (σRT) ~10–1 S cm–1) and the first organic metal, TTF–TCNQ complex3 (tetrathiafulvalene–tetracyanoquinodimethane, σRT = 102 S cm–1), are classic examples of such systems. In these cases, neutral molecules and radical ionic species of electron-donor and/or electron-acceptor molecules co-exist within a π-stacking column, generating the mixed valence (MV) state. A MV salt forms a partially filled band structure, providing the conduction carrier.
Besides plural component CT complexes and salts, single component neutral π-radicals are also studied for the development of organic conductors4,5,6. The singly occupied MO (SOMO) of a neutral π-radical gives a half-filled band, and the excitation from HOMO to LUMO bands is not necessary. However, most single component neutral π-radicals are insulators or poorly conductive semiconductors (σRT < 10–10 S cm–1) due to the conduction barrier originating from the on-site Coulomb repulsion (U, Mott insulator)7,8 and the steric hindrance needed to impart kinetic stability inhibiting intermolecular interactions9,10. High conductivities in single component neutral radicals with σRT > 10–6 S cm–1 have been reported in several systems, such as bisdithiazolyl and related radicals11,12,13,14,15,16,17 (σRT = 4 × 10–2 S cm–1 at most17, measured for single crystals), and betainic TCNQ18 and N,N’-dicyanoquinodiiminide19 radicals (σRT = 3.2 × 10–5 and 1.6 × 10–5 S cm–1 and at most, measured for compressed pellets, respectively), betainic TTF radicals20 (σRT = 1.0 × 10–1 S cm–1 at most, measured for compressed pellets), where the heteroatom effects and strong polarizability contribute to the reduction of U. It should be noted that carrier doping into neutral radical crystals generates the MV state and dramatically increases conductivity21,22. For an example, the conductivity of insulating crystals of 1,4-phenylene-bis(dithiadiazolyl) diradical increased to σRT = 100 S cm–1 by doping with iodine as an oxidant21. Furthermore, the connection of two TTF units through a π-electronic system effectively reduces U, and single component organic metals were realized in zwitterionic TTF dimer dicarboxylate radical23 (σRT = 530 S cm–1, measured for a bulk sheet) and a transition metal complex with two TTF ligands24 ([Ni(tmdt)2], σRT = 400 S cm–1, measured for single crystal). High conductivities over several S cm–1 have been also realized in several single component bis(dithiolene)-type metal complexes25,26,27 (σRT = 750 S cm–1 at most27, measured for single crystal). Highly conductive single component neutral radicals are also reported on several TTF derivatives having neutral radical units, such as nitronyl nitroxide28 (σRT = 9 × 10–4 S cm–1, measured for single crystal) and triphenylmethyl29,30,31 (σRT = 4 × 10–1 S cm–1 at most29, measured for single crystal under 15.2 GPa).
Phenalenyl (PNL), a neutral π-radical based on polycyclic aromatic hydrocarbons, possesses a delocalized charge and electronic spin structure around the π-system4,32,33,34. Our recent studies on PNL-based π-radicals disclosed their intriguing electronic and dynamic functions originating from self-assembling ability by the π-stacks through intermolecular overlap of SOMO33,34. PNL-based radicals are also regarded as potential candidates of organic conductors, and high conductivities of zwitterionic bisPNL boron complexes35,36,37,38,39,40,41,42 (σRT = 0.3 S cm–1 at most38,39, measured for single crystal) and bisPNL singlet-biradicals43 (σRT = 5.0 × 10–5 S cm–1, measured for single crystals) have been disclosed. Conductivities of monomeric PNL neutral radicals have been also investigated, however, they were insulators as seen in 1,9-dithioPNL44 (σRT < ~10–6 S cm–1, measured for compressed pellets), perchloroPNL45 (σRT = 10–10 S cm–1, measured for single crystals) and 2,5,8-tris(pentafluorophenyl)PNL46 (σRT < 1 × 10–10 S cm–1, measured for compressed pellets).
4,8,12-Trioxotriangulene (TOT, Fig. 1) can be designed by the 2D π-expansion and oxygen-atom substitution of PNL while retaining a three-fold molecular symmetry, and this carbon-centered neutral radical exhibits a peculiarly high stability even without steric protection due to the delocalization of electronic spin around the 25π-electronic system47,48,49,50. Our previous study on TOT derivatives having various substituent groups at the 2,6,10-positions (R3TOT, respectively, Fig. 1) disclosed the self-assembling ability based on two-electron-multicenter bonding forming 1D π-stacking columns (π-stacked radical polymers)47,48,49. The strong intermolecular orbital overlap within the column demonstrated the near-infrared photo-absorption48 and the high mobility n-type FET performance51.
We propose that the 1D self-assembling ability and the large 25π-electronic system with a robust condensed polycyclic structure of TOT may also provide high electrical conductivities in the solid state. More importantly, these peculiar electronic and structural properties may give a unique chance to synthesize MV salts by formal charge injection into the 1D stacks of TOT neutral radical crystals. Here we report high electrical conductivities of TOT neutral radicals (σRT = 10–4–10–3 S cm–1, measured for single crystals) as a single component purely organic molecular system, and a general tendency to form MV salts composed of neutral radicals and the corresponding anion species with further higher conductivities of σRT = 1–125 S cm–1.
Results
Stacking patterns and conductivities of TOT neutral radicals
Figures 2 and 3 illustrates the π-stacking columns of R3TOT derivatives (R = t-Bu, n-BuO, F, Br) and π-stacking patterns in the crystal structures, where the π-stacking patterns are modified with the aid of substituent effect47,48. The theoretical calculation on (t-Bu)3TOT and Br3TOT performed by Kinoshita et al. suggested that the electrical conductivity of π-stacking columns of TOT would be influenced by the stacking patterns within the 1D columns52.
The π-stacking column of (t-Bu)3TOT is formed by the 1D alignment of π-dimers with a staggered stacking pattern (Figs. 2a and 3a)47,48, where intermolecular overlap integral within the π-dimer (s1) was ~15 times larger than that of interdimer one (s2) (s1/s2 = 18.4 × 10–3/1.15 × 10–3), indicating that a strong spin–spin coupling within the π-dimer reduces the carrier concentration and also cut off the conduction pathway. In the tight-binding band calculation based on the crystal structure, the energy dispersion of SOMOs is large along the stacking direction (Fig. 4a and Supplementary Figure 1a) due to significant face-to-face π–π interactions. The strong π-dimerization in the 1D column causes a large band splitting of 0.30 eV between upper and lower bands with narrow bandwidths of ~0.06 eV, showing the characteristics of a typical band insulator of (t-Bu)3TOT (σRT < 10–8 S cm–1//c-axis). The insulating behavior of (t-Bu)3TOT due to strong π-dimerization can be also seen in the solid state electronic spectrum, where the λmax of intermolecular transition was observed at 8800 cm–1 (Fig. 5)49. In the case of (n-BuO)3TOT, a weakly dimerized π-stacking column was formed, where the TOT skeleton stacked with twisting of ~32° and overlapping central carbon atoms (Figs. 2b and 3b). The interdimer overlap integral s2 (–11.4 × 10–3) was much larger than that in (t-Bu)3TOT (Fig. 2b), causing the larger band dispersion and a smaller band gap of 0.16 eV in (n-BuO)3TOT (Fig. 4b). Such situation well coincides with the lower energy intermolecular transition, λmax = 6700 cm–1 in the solid-state electronic spectrum (Fig. 5)48, suggesting the higher concentration of carrier than that of (t-Bu)3TOT. (n-BuO)3TOT showed a high conductivity of σRT = 1.2 × 10–3 and a semiconducting behavior with an activation energy (Ea) = 296 meV (Fig. 6). In the π-stacking column of F3TOT, the staggered π-dimer stacked with a fan-shaped stacking, where the central carbon atoms possessing the largest spin density did not overlap each other (Figs. 2c, 3a, and 3c). The interdimer interaction s2 = –9.7 × 10–3 S cm–1 was slightly smaller than that of (n-BuO)3TOT. Due to the difference in the interdimer interaction, the band dispersion of F3TOT became smaller than that of (n-BuO)3TOT (Fig. 4c), resulting in the slightly lower conductivity of σRT = 8.1 × 10–5 S cm–1 with Ea = 304 meV (Fig. 6).
Br3TOT forms a uniform π-stacking column with a large face-to-face distance of 3.43 Å and a slip-stacked pattern in the crystal structure (Figs. 2d and 3d)47. The overlap integral of SOMO within the column (s = 2.0 × 10–3) was ~103 times larger than those of intercolumnar directions. The band calculation gave a half-filled band with a bandwidth of 0.20 eV along the a-axis and a 1D Fermi surface (Fig. 4d and Supplementary Figure 1d). The electrical conductivity of the Br3TOT single crystal measured along the stacking direction (a-axis) was σRT = 1.8 × 10–3 S cm–1. The resistivity became larger with lowering temperature indicating the semiconducting nature with Ea = 308 meV (Fig. 6). These solid-state properties of Br3TOT are well explained as a Mott-type insulator. It should be noted that the σRT values of (n-BuO)3TOT and Br3TOT are very high as single component neutral radicals.
The high conductivity of Br3TOT is achieved by the stable uniform π-stacked radical array forming conduction pathway. For the reduction of U, conduction barrier, extensive π-electronic system with the delocalization of charge and electronic spin is also essential. U of TOT derivatives can be experimentally evaluated by the difference between oxidation and reduction potentials in the electrochemical measurement. That value of (t-Bu)3TOT in dichloromethane is 1.10 V (Supplementary Figure 3), and is much smaller than PNL radical (1.52 V)53 and 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO) radical (2.3 V, measured in our laboratory). The strong CT interaction within the column and reduction of U in Br3TOT is also indicated by a low-energy absorption around 8000 cm–1 in the solid-state electronic spectrum (Fig. 5) that can be assigned as an intermolecular CT band48,54.
The intermolecular overlap integral in the Br3TOT column was one order smaller than those of other derivatives, because the central carbon atoms having the largest SOMO coefficient were not close in the slipped stack (Fig. 3d). Due to the small overlap integral within the column, the bandwidth of Br3TOT was 0.20 eV, and smaller than those of (n-BuO)3TOT and F3TOT (Fig. 2b–d). Since the conduction barrier in a 1D array of neutral radicals is defined as Ueff–4t, where Ueff and t are the effective on-site Coulomb repulsive energy and the transfer integral of SOMOs, respectively, the strong orbital overlap is essential for the improvement of conductivity. A uniform π-stacking manner by strong face-to-face interaction with the overlapping of central carbon atoms would be the most effective for further high conductivity in single component TOT neutral radicals.
Crystal structures and conductivities of MV salts
These findings strongly encouraged us to prepare MV salts of TOT by formally injecting excess electrons into the 1D stacks of TOT neutral radical crystals. As a result, our continuous synthetic efforts finally found out a general preparation method for MV salts of TOTs formulated as (n-Bu4N+)(Br3TOT–)(Br3TOT)2(THF)2 1, (n-Bu4N+)(Cl3TOT–)(Cl3TOT) 2, and (Li+)(Cl3TOT–)(Cl3TOT)x(H2O)y 3 were obtained by the electrocrystallization of R3TOT– salts in appropriate solvents (Supplementary Figure 2). In the case of 3, the component ratios x and y cannot be determined due to the severe disorder of Li+ and crystalline water molecules. The MV state of 1–3 can be also confirmed by the solid-state electronic spectra (Fig. 5), where intermolecular CT bands between monoanion and neutral radical species are observed around 3000 cm–154. These bands suggest the high carrier concentration and high conductivity by the injection of excess electron into the column of neutral radicals.
Figure 7a shows the 1D π-stacking column in the crystal structure of 1. The column was surrounded by n-Bu4N+ and THF molecules, resulting in the 1D electronic structure. There were five crystallographically independent Br3TOT molecules (labeled as A–E in Fig. 7a) having a three-fold symmetry, and the E molecule also possessed a two-fold axis on the molecular plane. These molecules were aligned in the order of D–C–B–A–E–A–B–C–D as a repeating unit. The face-to-face stacking distances varied from 3.21 to 3.35 Å (Supplementary Figure 4), and smaller than the sum of the van der Waals radii of two carbon atoms (3.40 Å)55. In each stack of Br3TOT molecules, the central carbon atoms, which possessed the largest SOMO coefficient and spin distribution49,50, were close to each other, causing strong orbital overlaps (Supplementary Figure 4). The charge-distribution in the column was discussed by the C–O bond lengths of the TOT skeleton. The C–O bond lengths of Br3TOT neutral radical is 1.223 Å and slightly shorter than that in the anion salt (1.244 Å) (averaged values, Supplementary Table 1)47. The similar relationship can be also seen in the neutral radical and monoanion species of (t-Bu)3TOT (1.229 vs. 1.245 Å, respectively, Supplementary Table 1)47,48,49. The slightly longer C–O bond lengths in monoanion species is caused by the large electronegativity of the oxygen atom, stabilizing the resonance structures having C–O single bond character, where the minus charge locates on the oxygen atom. Since the differences in C–O bond lengths between neutral radical and monoanion species were small in comparison with experimental errors, the precise estimation of charge on each molecule could not be done. However, the longer C–O bond lengths in C and E molecules (1.251(9) and 1.240(17) Å, respectively) than the others (1.211(9)–1.214(9) Å) suggest the charge-separated state within the column (Supplementary Table 1). Accordingly, the column is constructed by the alternation of neutral radical dimer and anionic species as shown in Fig. 5a. The dimerization of neutral radicals in the column was supported by the strong antiferromagnetic coupling observed in the temperature-variable ESR measurement (2J/kB = –694 K) (Supplementary Figure 6a)56. The anionic C and E molecules were arranged in a layered structure perpendicular to the stacking direction (blue molecules in Fig. 7b). The anion layer also included the n-Bu4N+ molecules (green molecules in Fig. 7b), indicating that the electrostatic interaction between the cation layer and Br3TOT molecules induces the charge-separated state within the MV column. Due to the charge-separation and non-uniform overlap integrals (Supplementary Figure 4) in the column, 1 showed a semiconducting behavior with Ea = 127 meV (Fig. 6). The σRT value, 1.3 S cm–1, was three orders larger than that of Br3TOT neutral radical crystal.
Similarly to 1, the salt 2 was constructed by the 1D columns of Cl3TOT molecules (A–B and C–F columns in Fig. 7c, d) that were surrounded by n-Bu4N+ molecules. The C–O bond lengths of each molecules varied in a range of 1.224(5)–1.240(4) Å (Supplementary Table 1), suggesting a charge-separated state also in these columns, although the distinct difference as seen in 1 was not found. The overlap integrals within both columns were not uniform (Supplementary Figure 5), causing the small band gap of ~0.02 meV of the SOMO band (Fig. 7e) at the Fermi level. The salt 2 exhibited a semiconducting behavior with σRT = 2.5 S cm–1 and Ea = 177 meV (Fig. 6). The temperature dependence of the magnetic susceptibility showed a strong antiferromagnetic coupling within the columns (Supplementary Figure 6b, 2J/kB = –1220 K with the singlet–triplet model)56.
The variation of counteraction highly affected the stacking pattern and the conductivity of the MV TOT columns, and the highest conductivity was obtained in the combination of Li+ and Cl3TOT. The crystal of 3 contained crystallographically independent two Cl3TOT molecules (A and B) both of which possessed a three-fold symmetry. Molecules A and B individually formed 1D columns with the uniform stacking distance of 3.29 Å (A and B columns, Fig. 8a). The overlapping patterns were staggered (A, Fig. 8b) and twisted (B, Fig. 8c) modes, and the central carbons were close to each other, resulting in the strong orbital overlaps (overlap integrals of A and B columns: sA = 8.0 × 10–3 and sB = 15.3 × 10–3, respectively, Fig. 8a). The tight binding approximated calculation showed six SOMO/HOMO bands from six Cl3TOT molecules in a unit cell, and indicated a 1D electronic structure along the stacking direction (Fig. 8d). The band dispersion of B column, 0.48 eV, was larger than that of A column (green and blue lines in Fig. 8d, respectively), reflecting the stronger orbital overlap within the B column. These calculations suggest that the B column can exhibit more effective charge-transport ability. The electrical conductivity of 3, σRT = 125 S cm–1, was two orders higher than those of n-Bu4N+ salts 1 and 2, and the activation energy was 12 meV (Fig. 6). In the magnetic measurement, 3 showed a superimposition of two kinds of 1D antifferomagnetic chains of Bleaney–Bowers56 and Bonner–Fisher57 models (Supplementary Figure 7), suggesting that A and B columns exhibited different magnetic and conducting behaviors. The origin of the semiconducting behavior has not been clarified, however, the electrostatic perturbation due to Li+ and H2O molecules might cause the conduction barrier (Fig. 8e). The temperature dependence of ESR intensity of 3 did not show any abrupt change due to phase transition even under the low temperature of ~4 K.
Discussion
In conclusion, we have investigated electrical conductivities of π-stacked radical polymers of TOT derivatives. The extended π-electronic system with a charge-delocalization and strong π-stacks realized high conductivities of σRT ~ 10–3 S cm–1 as a single component, purely organic molecular system. The conductivity further increased in the MV salts, and the Li+ salt 3 exhibited a notably high conductivity of 102 S cm–1 order, which is close to that of the TTF–TCNQ complex. Control of stacking patterns within the columns by diversifying counter cation and chemical modifications on the TOT skeleton will realize the further higher conductivity and metallic conduction in TOT-based π-stacked radical polymers. Furthermore, the application of high pressure to neutral radical and MV salts of TOT may dramatically increase the conductivity and realize a metallic behavior as reported in bisdithazolyl neutral radicals13,15,16,17.
Material exploration for organic conductors revealed a number of intriguing functions and phenomena1. However, most previous studies are based on limited molecular skeletons, such as TTF, TCNQ, dithiolate-type complexes, C60, etc. In the present study, TOT showed a general tendency to form 1D columns and MV salts exhibiting high conductivities. This result suggests the potential for TOT to serve as a molecular building block of organic conductors, and may provide a new milestone in the material exploration in the development of organic electronic materials. It should be noted that various chemical modifications on TOT skeleton, such as introduction of various substituent groups at R positions (Fig. 1) and oxo-groups58,59, can be carried out, giving a chance to discover new phenomena and functions. Our current interest is aimed to the development of multi-functional organic conductors by the introduction of chirality60,61 and coordination to metal ions.
Methods
Materials
TOT derivatives and the monoanion salts were synthesized according to our previous papers47,48,49. Tetrabutylammonium tetrafluoroborate ((n-Bu4N+)(BF4–)) was purified by the recrystallization from ethanol. Solvents were dried (drying reagent in parenthesis) and distilled under an argon atmosphere prior to use: THF (Na–benzophenone ketyl); acetonitrile (P2O5/CaH2); and methanol (Mg).
Measurements
Electronic spectra were measured for KBr pellets or solutions on a Shimadzu UV/vis scanning spectrophotometer UV-3100PC. Infrared spectrum was recorded on a JASCO FT/IR-660 Plus spectrometer using a KBr plate (resolution 2 cm–1). Cyclic voltammetric measurement was made with an ALS Electrochemical Analyzer model 630A. Cyclic voltammogram was recorded with 3-mm-diameter carbon plate and Pt wire counter electrodes in dry CH2Cl2 containing 0.1 M n-Bu4NClO4 as supporting electrolyte at room temperature. The experiment employed a Ag/AgNO3 reference electrode, and the final result was calibrated with ferrocene/ferrocenium couple. The magnetic properties were measured on a single crystal through an ESR Spectrometer Bruker E500 in the temperature range of 4–300 K. The DC magnetic susceptibility was measured at a field of 0.3 or 4.0 T with an MPMS-XL Quantum Design SQUID magnetometer in a temperature range of 2–350 K in an atmosphere of helium. DC conductivities were measured by a standard four-probe technique using gold wires of 10 µm diameter using carbon paste.
X-ray crystallography
X-ray crystallographic measurements were made on a Rigaku Raxis-Rapid imaging plate with graphite monochromated Mo Kα (λ = 0.71075 Å). Structures were determined by a direct method using SIR200462. Refinements were performed by a full-matrix least-squares on F2 using the SHELXL-9763. All non-hydrogen atoms were refined anisotropically, and all hydrogen atoms were included but not refined. Empirical absorption corrections were applied. Selected crystal data collection parameters are given in Supplementary Table 2.
Preparation of (n-Bu4N+)(Br3TOT–)(Br3TOT)2(THF)2 (salt 1)
In an H-shaped cell with a glass frit separating two chambers, (n-Bu4N+)(Br3TOT–) (7.5 mg, 0.01 mmol) and (n-Bu4N+)(BF4–) (33.5 mg, 0.10 mmol) were placed in the anodic and both chambers, respectively. After dissolving the salt and supporting electrolyte in THF (18 mL) under Ar atmosphere, a constant current of 1 µA was applied for 10 days between the platinum electrodes of the 1 mm diameter, yielding salt 1 (3.8 mg) as black needle crystals. Due to the efflorescence of the salt, elemental analysis could not be performed. In the measurements of ESR and electrical conductivity, the samples were coated with epoxy resin to prevent from decomposition of the crystals. The X-ray diffraction measurement was performed for a sample soaked in mineral oil and then frozen with a cold N2 gas (200 K).
Preparation of (n-Bu4N+)(Cl3TOT–)(Cl3TOT) (2)
In an H-shaped cell with a glass frit separating two chambers, (n-Bu4N+)(Cl3TOT–) (3.4 mg, 0.005 mmol) and (n-Bu4N+)(BF4–) (34.6 mg, 0.10 mmol) were placed in the anodic and both chambers, respectively. After dissolving the donor and supporting electrolyte in acetonitrile (18 mL) under Ar atmosphere, a constant current of 0.5 µA was applied for 5 days between the platinum electrodes of the 1 mm diameter, yielding salt 2 (1.5 mg) as black needle crystals.
Preparation of (Li+)(Cl3TOT–)(Cl3TOT)x(H2O)y (3)
In an H-shaped cell with a glass frit separating two chambers, (Li+)(Cl3TOT–) (9.5 mg, 0.022 mmol) and LiClO4 (83.3 mg, 0.78 mmol) were placed in the anodic and both chambers, respectively. After dissolving the salt and supporting electrolyte in methanol (18 mL) under Ar atmosphere, a constant current of 1.0 µA was applied for 14 days between the platinum electrodes of the 1 mm diameter, yielding salt 3 (1.2 mg) as black needle crystals.
Data availability
The X-ray crystallographic coordinates for the structures of MV salts 1–3 are available as Supplementary Data 1–3, respectively. These data have also been deposited at Cambridge Crystallographic Data Centre (CCDC), under deposition numbers CCDC1856410, CCDC1856411, and CCDC1856412. These data can be obtained free of charge from the CCDC via http://www.ccdc.cam.ac.uk/data_request/cif. The other data that support the findings of this study are available from the corresponding author upon reasonable request.
References
Batail, P. Introduction: molecular conductors. Chem. Rev. 104, 4887–4890 (2004).
Akamatu, H., Inokuchi, H. & Matsunaga, Y. Electrical conductivity of the perylene-bromine complex. Nature 173, 168–169 (1954).
Ferraris, J., Cowan, D. O., Walatka, V. & Perlstein, J. H. Electron transfer in a new highly conducting donor-acceptor complex. J. Am. Chem. Soc. 95, 948–945 (1973).
Haddon, R. C. Design of organic metals and superconductors. Nature 256, 394–396 (1975).
Haddon, R. C. Quantum chemical studies in the design of organic metals. II. The equilibrium geometries and electronic structure of tetrathiofulvalene (ttf) and tetracyanoquinodimethane (tcnq) and their uni- and di-valent ions. Aust. J. Chem. 28, 2333–2342 (1975).
Haddon, R. C. Metals and superconductors: molecular analogs of atomic hydrogen. Chemphyschem 13, 3581–3583 (2012).
Torrance, J. B. The Difference between metallic and insulating salts of tetracyanoquinodimethane (TCNQ): how to design an organic metal. Acc. Chem. Res. 12, 79–86 (1979).
Garito, A. F. & Heeger, A. J. The design and synthesis of organic metals. Acc. Chem. Res. 7, 232–240 (1974).
Eley, D. D., Jones, K. W., Littler, J. G. F. & Willis, M. R. Semiconductivity of organic substances. Part 10. Electrical properties of galvinoxyl, α,α-diphenyl-β-picryl hydrazyl and related solid free radicals. Trans. Faraday Soc. 62, 3192–3200 (1966).
Inokuchi, H., Harada, Y. & Maruyama, Y. Electric properties of the single-crystal and thin film of α,α´-diphenyl-β-picrylhydrazyl. Bull. Chem. Soc. Jpn. 35, 1559–1561 (1962).
Hicks, R. G. The Synthesis and characterization of stable radicals containing the thiazyl (SN) fragment and their use as building blocks for advanced functional materials. In Stable Radicals: Fundamentals and Applied Aspects of Odd-electron Compounds (ed. Hicks, R. G.) 317−380 (John Wiley & Sons, Chichester, 2010).
Yu, X. et al. Semiquinone-bridged bisdithiazolyl radicals as neutral radical conductors. J. Am. Chem. Soc. 134, 2264–2275 (2012).
Mailman, A. et al. Crossing the insulator-to-metal barrier with a thiazyl radical conductor. J. Am. Chem. Soc. 134, 9886–9889 (2012).
Wong, J. W. L. et al. Supramolecular architecture, crystal structure and transport properties of the prototypal oxobenzene-bridged bisdithiazolyl radical conductor. Chem. Commun. 50, 785–787 (2014).
Wong, J. W. L. et al. Pressure induced phase transitions and metallization of a neutral radical conductor. J. Am. Chem. Soc. 136, 1070–1081 (2014).
Tian, D. et al. The metallic state in neutral radical conductors: dimensionality, pressure and multiple orbital effects. J. Am. Chem. Soc. 137, 14136–14148 (2015).
Mailman, A. et al. Fine tuning the performance of multiorbital radical conductors by substituent effects. J. Am. Chem. Soc. 139, 1625–1635 (2017).
Tsubata, Y., Suzuki, T. & Miyashi, T. Single-component organic conductors based on neutral radicals containing the pyrazino-TCNQ skeleton. J. Org. Chem. 57, 6749–6755 (1992).
Suzuki, T., Yamashita, Y., Imaeda, K., Ishida, T. & Nogami, T. Single-component organic semiconductors based on novel radicals that exhibit electrochemical amphotericity: preparation, crystal structures, and solid-state properties of N,N’-dicyanopyrazinonaphthoquinodiiminides substituted with an N-alkylpyridinium unit. J. Org. Chem. 66, 216–224 (2001).
Neilands, O. Dioxo-and aminooxopyrimido-fused tetrathiafulvalenes-base compounds for novel organic semiconductors and for design of sensors for recognition of nucleic acid components. Mol. Cryst. Liq. Cryst. 335, 331–349 (2001).
Bryan, C. D. et al. Conducting charge-transfer salts based on neutral π-radicals. Nature 365, 821–823 (1993).
Awaga, K. et al. Multi-dimensional crystal structures and unique solid-state properties of heterocyclic thiazyl radicals and related materials. Bull. Chem. Soc. Jpn. 79, 25–34 (2006).
Kobayashi, Y., Terauchi, T., Sumi, S. & Matsushita, Y. Carrier generation and electronic properties of a single-component pure organic metal. Nat. Mater. 16, 109–115 (2017).
Tanaka, H., Okano, H., Kobayashi, H., Suzuki, W. & Kobayashi, A. A three-dimensional synthetic metallic crystal composed of single-component molecules. Science 291, 285–287 (2001).
Belo, D. et al. Gold complexes with dithiothiophene ligands: a metal based on a neutral molecule. Chem. Eur. J. 7, 511–519 (2001).
Filatre-Furcate, A. et al. Single-component conductors: a study electronic structure generated by bulky substituents. Inorg. Chem. 55, 6036–6046 (2016).
Gal, Y. L. et al. Stable metallic state of a neutral-radical single-component conductor at ambient pressure. J. Am. Chem. Soc. 140, 6998–7004 (2018).
Komatsu, H. et al. Influence of magnetic field upon the conductance of a unicomponent crystal of a tetrathiafulvalene-based nitronyl nitroxide. J. Am. Chem. Soc. 132, 4528–4529 (2010).
Souto, M. et al. Pressure-induced conductivity in a neutral nonplanar spin-localized radical. J. Am. Chem. Soc. 138, 11517–11525 (2016).
Souto, M., Rovira, C., Ratera, I. & Veciana, J. TTF–PTM dyads: from switched molecular self assembly in solution to radical conductors in solid state. CrystEngComm 19, 197–206 (2017).
Souto, M. et al. Role of the open-shell character on the pressure-induced conductivity of an organic donor–acceptor radical dyad. Chem. Eur. J. 24, 5500–5505 (2018).
Reid, D. H. The chemistry of the phenalenes. Q. Rev. 19, 274–302 (1965).
Morita, Y., Suzuki, S., Sato, K. & Takui, T. Synthetic organic spin chemistry for structurally well-defined open-shell graphene fragments. Nat. Chem. 3, 197–204 (2011).
Morita, Y. & Nishida, S. Phenalenyls, cyclopentadienyls, and other carbon-centered radicals. In Stable radicals: Fundamentals and Applied Aspects of Odd-electron Compounds (ed. Hicks, R. G.), 81–145 (John Wiley & Sons, Chichester, 2010).
Chi, X. et al. The first phenalenyl-based neutral radical molecular conductor. J. Am. Chem. Soc. 121, 10395–10402 (1999).
Chi, X. et al. Dimeric phenalenyl-based neutral radical molecular conductors. J. Am. Chem. Soc. 123, 4041–4048 (2001).
Pal, S. K. et al. Synthesis, structure and physical properties of the first one-dimensional phenalenyl-based neutral radical molecular conductor. J. Am. Chem. Soc. 126, 1478–1484 (2004).
Pal, S. K. et al. Resonating valence-bond ground state in a phenalenyl-based neutral radical conductor. Science 309, 281–284 (2005).
Mandal, S. K. et al. Resonating valence bond ground state in oxygen-functionalized phenalenyl-based neutral radical molecular conductors. J. Am. Chem. Soc. 128, 1982–1994 (2006).
Bag, P. et al. Resonating valence bond and σ-charge density wave phases in a benzannulated phenalenyl radical. J. Am. Chem. Soc. 132, 2684–2694 (2010).
Sarkar, A. et al. Synthesis, structure, and physical properties of a partial π-stacked phenalenyl-based neutral radical molecular conductor. Chem. Eur. J. 17, 11576–11584 (2011).
Sarkar, A. et al. Sulfur and selenium substituted spiro-biphenalenyl-boron neutral radicals. J. Mater. Chem. 22, 8245–8256 (2012).
Kubo, T. et al. Synthesis, intermolecular interaction, and semiconductive behavior of a delocalized singlet biradical hydrocarbon. Angew. Chem. Int. Ed. 44, 6564–6568 (2005).
Beer, L. et al. The first electronically stabilized phenalenyl radical: effect of substituents on solution chemistry and solid-state structure. Cryst. Growth Des. 7, 802–809 (2007).
Koutentis, P. A. et al. Perchlorophenalenyl radical. J. Am. Chem. Soc. 123, 3864–3871 (2001).
Uchida, K. et al. Dual association modes of the 2,5,8-tris(pentafluorophenyl)phenalenyl radical. Chem. Asian J. 9, 1823–1829 (2014).
Morita, Y. et al. Organic tailored batteries materials using stable open-shell molecules with degenerate frontier orbitals. Nat. Mater. 10, 947–951 (2011).
Ikabata, Y. et al. Near-infrared absorption of π-stacking columns composed of trioxotriangulene neutral radicals. npj Quantum Mater. 2, 27 (2017).
Morita, Y. et al. Trioxotriangulene: air- and thermally stable organic polycyclic carbon-centered neutral π-radical without steric protection. Bull. Chem. Soc. Jpn. 91, 922–931 (2018).
Ikabata, Y., Akiba, K. & Nakai, H. Superphenalenyl: theoretical design of a π-conjugated planar hydrocarbon radical. Chem. Lett. 42, 1386–1387 (2013).
Morita, Y. et al. Organic semiconductor material with neutral radical compound of a trioxotriangulene derivative as a semiconductor material. US Patent 8933437 B2 (2009).
Kinoshita, K. et al. Theoretical studies on the magnetic and conductive properties of crystals containing open-shell trioxotriangulene radicals. Bull. Chem. Soc. Jpn. 89, 315–333 (2016).
Goto, K. et al. A stable neutral hydrocarbon radical: Synthesis, crystal structure, and physical properties of 2,5,8-tri-tert-butyl-phenalenyl. J. Am. Chem. Soc. 121, 1619–1620 (1999).
Torrance, J. B., Scott, B. A., Welber, B., Kaufman, F. B. & Seiden, P. E. Optical properties of the radical cation tetrathiafulvalenium (TTF+) in its mixed-valence and monovalence halide salts. Phys. Rev. B 19, 730–741 (1979).
Bondi, A. van der Waals volumes and radii. J. Phys. Chem. 68, 441–451 (1964).
Bleany, B. & Bowers, D. K. Anomalous paramagnetism of copper acetate. Proc. R. Soc. Lond. Sec. A 214, 451–465 (1952).
Bonner, J. C. & Fisher, M. E. Linear magnetic chains with anisotropic coupling. Phys. Rev. 135, A640–A658 (1964).
Ueda, A. et al. An extremely redox-active air-stable neutral π radical: dicyanomethylene-substituted triangulene with a threefold symmetry. Chem. Eur. J. 18, 16272–16276 (2012).
Ueda, A. et al. A dicyanomethylene-substituted triangulene: effects of molecular- symmetry reduction and electron-accepting substituents on a fused polycyclic neutral π-radical system. Chem. Asian J. 8, 2057–2063 (2013).
Avarvari, N. & Wallis, J. D. Strategies towards chiral molecular conductors. J. Mater. Chem. 19, 4061–4076 (2009).
Pop, F., Auban-Senzier, P., Canadell, F., Rikken, G. L. J. A. & Avarvari, N. Electrical magnetochiral anisotropy in a bulk chiral molecular conductor. Nat. Commun. 5, 4757 (2014).
Burla, M. C. et al. SIR2004: an improved tool for crystal structure determination and refinement. J. Appl. Crystallogr. 38, 381–388 (2005).
Sheldrick, G. M. A short history of SHELX. Acta Crystallogr. Sect. A 64, 112–122 (2008).
Acknowledgements
We thank Dr. M. Ishikawa and Prof. H. Yamochi, Kyoto University for their technical assistances in the electrical conductivity measurements. This work was supported by the Canon Foundation, the Grants-in-Aid for Scientific Research B (Nos. 25288022 and 16H04114) and Elements Science and Technology Project from the Ministry of Education, Culture, Sports, Science and Technology, Japan, and Core Research for Evolutional Science and Technology (CREST) Basic Research Program “Creation of Innovative Functions of Intelligent Materials on the Basis of Element Strategy” of Japan Science and Technology Agency (JST).
Author information
Authors and Affiliations
Contributions
Y.M. planned this project. T.M. and C.Y. prepared the materials and demonstrated spectroscopic analyses and measurement of transport properties. T.M. carried out X-ray crystal structure analyses and the tight-binding band calculation. K.F. measured ESR and magnetic susceptibilities of 1–3. All authors wrote and reviewed the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic supplementary material
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Murata, T., Yamada, C., Furukawa, K. et al. Mixed valence salts based on carbon-centered neutral radical crystals. Commun Chem 1, 47 (2018). https://doi.org/10.1038/s42004-018-0048-5
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s42004-018-0048-5
- Springer Nature Limited
This article is cited by
-
Metal-free electrocatalysts for oxygen reduction reaction based on trioxotriangulene
Communications Chemistry (2019)