J O U R N A L O F C H E M I S T R Y Materials Azulene-substituted TTF derivatives† Hiroshi M. Yamamoto, Jun-Ichi Yamaura and Reizo Kato* Institute for Solid State Physics, T he University of T okyo, RoppongiMinato-ku, T okyo 106, Japan In order to examine Little’s model for organic superconductors, several azulene-substituted TTF derivatives were synthesized. Measurements of the oxidation potentials using cyclic voltammetry (CV) provide their donor abilities.The molecular structure of AET (azulenoethylenedithiotetrathiafulvalene) was determined by X-ray analysis. Cation radical salts of synthesized donors with BF4-, ClO4-, PF6-, AsF6- and [Pt(dmit)2]n- (dmit=C3S52-=2-thioxo-1,3-dithiole-4,5-dithiolate) were prepared by galvanostatic electrolyses. Temperature-dependent electrical resistances indicate that these salts are all semiconductive.The crystal structure of (AET)2[Pt(dmit)2] was determined by X-ray analysis and its electronic structure is discussed. Organic molecular conductors have been the subject of considerable interest in physics and chemistry because of their stimulating properties such as their charge density wave,1 field induced spin density wave,2 spin-Peierls transition3 and superconductivity. 4 In order to reach the superconducting state, a finite value of the density of state should be maintained and thus the metal–insulator transition should be suppressed down to low temperature. Considering that the low-dimensional nature of the system induces a metal–insulator transition, one can understand that an increase of dimensionality of the electronic structure is the current focus in developing molecular superconductors.This strategy has led to the preparation of many molecular conducting materials that exhibit various types of Fermi surfaces.5 Some of these materials such as BEDT–TTF [bis(ethylenedithio)tetrathiafulvalene] salts show superconducting behaviour.6 It is obvious, however, that the properties of superconductivity, for example the critical temperature (Tc), are governed not only by the geometry of the Fermi surface but also by many other factors.7 Little has pointed out that the Tc of organic superconductors can rise to room temperature via a special mechanism.8 Fig. 1 illustrates his model, a polyacetylene chain substituted by cyanine dye. First, conduction electrons excite the positive charge of the side chains to approach close to the chain backbone. Second, the positive charge attracts another electron and thus provides an attractive force for electrons.Finally, this attractive force brings about the superconductivity. R R R R R N N Et N N Et N N Et R = + – – Therefore, an oscillation of positive charge on the side chain Fig. 1 Illustration of Little’s model for high-temperature supertakes the place of the phonon in the BCS theory.9 In other conductivity words, the conventional electron–phonon interaction mechanism is replaced by an electron–exiton mechanism, based on TTF derivatives. We selected the TTF moiety as a conduction electronic polarization eVects. path, and the azulene ring as a source of oscillating dipole.The high Tc in Little’s calculation is achieved because the Our choice of azulene was based on the following. (i ) Because mass of the electron is much smaller than that of nuclei which azulene consists of five- and seven-membered rings, the 6p can be regarded as an extension of the isotope eVect on the Tc aromatic stabilization eVect on each ring gives rise to a charge of superconductivity.We can, therefore, make use of the separation in the ground state10 [Fig. 2(a)]. Azulene therefore electronic polarization component of permeability instead of has a significant dipole moment with the negative charge on the ionic part. Electronic polarization has a higher energy than the five-membered ring in the ground state11 (1.08 Debye). In ionic polarization so Tc calculated using Little’s theory should addition, since the coeYcients of its LUMO have a tendency be higher than the conventional one.to gather on the seven-membered ring [Fig. 2(b)], azulene in Little’s polymer has not yet been prepared due to synthetic the excited state has a dipole moment in the opposite direc- diYculties, although we still consider the theory very attractive.tion12 (-0.42 Debye). Thus the excitation of an electron on We, therefore, planned to substitute an array of organic radicals an azulene ring will change the electric environment of the for the polyacetylene in order to make his proposal syntheticonduction carriers on the TTF moiety. (ii ) Excitation of an cally accessible. To this end, we designed azulene-substituted electron from the HOMO to the LUMO requires little energy (1.8 eV) because the excitation reduces the mutual repulsion between electrons.10 This small excitation energy enables the † Presented at the 58th Okazaki Conference, Recent Development conduction carriers on the TTF moiety to utilize the azulene and Future Prospects of Molecular Based Conductors, Okazaki, Japan, 7–9 March 1997.ring as a source of the oscillating dipole. J. Mater. Chem., 1998, 8(2), 289–294 289Table 1 Cyclic voltammetric data for new donors donor E1/Va E2/Va DE/V=E2-E1 BEDT TTF 0.12 0.43 0.31 5a 0.17 0.49 0.32 5b 0.15 0.44 0.29 5c 0.16b 0.47b 0.34 5d 0.24b 0.53b 0.29 HOMO LUMO ( a ) (b ) 5e 0.44c 0.84c 0.40 Fig. 2 (a) Charge separation in the azulene ring at its ground state; 5f 0.24c 0.63c 0.39 (b) schematic representation of the coeYcients of the azulene HOMO and LUMO aVs.Ag/AgNO3 , benzonitrile, 22 °C, Pt working and counter electrodes. bQuasi-reversible. cIrreversible (E1 and E2 values were determined by diVerential pulse polarography). The azulene ring attached to the TTF moiety can thus act as an oscillating dipole. The interaction between conduction carriers on the TTF and the oscillating dipole on the azulene probably owing to undesired reaction at the azulene ring.Donor 5a (AET) and 5b show oxidation potentials similar to ring should make carriers attractive to each other leading to superconductivity. With this in mind, we have examined those of BEDT–TTF, in spite of the introduction of the electron-withdrawing azulene ring.This result ensures the azulene-substituted TTF derivatives and this article describes the syntheses, electrochemistry, visible absorption, X-ray analy- donor ability of these donors. The substitution of selenium for outer sulfur (5c) does not aVect the potentials significantly, but sis and MO calculations of such TTF derivatives as well as the electrical properties and X-ray analysis of their cation replacement of inner sulfur (5d) positively shifts both E1 and E2 values.The high E1 and E2 values for 5e can be attributed radical salts. to the electron-withdrawing character of the thiadiazole ring. The molecular structure of AET was determined by single Results and Discussion crystal X-ray analysis (Fig. 3, Tables 2, 3). The CMC bond lengths in the azulene ring are almost equal (about 1.4 A ° ).This Synthesis and properties of neutral donors suggests that the aromaticity of the ring is retained even if the The synthetic route is presented in Scheme 1. 5,6- Dichloroazulene 1 is converted to trithiocarbonate 3 by the nucleophilic attack of potassium sulfide and subsequent intramolecular ring closure in potassium trithiocarbonate 2.Azulene-substituted TTF derivatives 5a–5f were synthesized by the cross-coupling reactions of units 4a–4f with azulenecontaining unit 3 using neat triethyl phosphite. The symmetrical donor, diazuleno–TTF, was also synthesized by the same method, but it was not characterized in detail because it is a mixture of cis and trans forms. The yields of coupling reactions were moderate except for the low yield of 5e.Replacement of the thioketone in the unit 4e with the ketone did not improve the reaction yield. This low yield was due to the decomposition of the thiadiazole ring. The electrochemical properties of donors 5a–5f have been studied by cyclic voltammetry (CV) and the results are listed in Table 1, along with the data for BEDT–TTF as a reference Fig. 3 (a) Side view of the AET; (b) top view of the AET; (c) crystal compound. Some of the observed redox peaks were irreversible packing of the AET Table 2 Crystallographic data for AET and (AET)2[Pt(dmit)2] AET (AET)2[Pt(dmit)2] empirical formula C16H10S6 C38H20PtS22 formula mass 394.61 1376.99 colour & shape dark brown brownish black plate block crystal system monoclinic monoclinic space group P21/n P21/n a/A° 19.104(4) 11.619(3) b/A ° 13.057(3) 30.222(4) c/A ° 6.515(2) 6.547(1) b (°) 93.16(2) 97.62(2) V /A ° 3 1622.5(6) 2278.7(7) Z 4 2 total no.of observed 4224 5218 reflections no. of unique data 2123 3640 with I3s(I) no. of variables 239 288 R; Rw 0.046; 0.032 0.042; 0.044 maximum peak in 0.35e 1.49e final diV. map/A ° 3 minimum peak in -0.39e -1.88e Cl Cl S S S S S X X X X Y R R R R Cl S S–K+ S K2S, CS2 1 3 (30%) 5a – f 4a – f P(OEt)3 H2O–DMF, 60 °C 2 a X = S, Y = O, R–R = SCH2CH2S 44% b X = S, Y = O, R–R = SCH2S 49% c X = S, Y = O, R–R = SeCH2CH2Se 34% d X = Se, Y = O, R–R = SCH2CH2S 38% e X = S, Y = S, R–R = 5% f X = S, Y = O, R = I 17% NSN final diV.map/A° 3 Scheme 1 290 J. Mater. Chem., 1998, 8(2), 289–294Table 3 Selected bond lengths (d/A° ) for (AET)2[Pt(dmit)2] AET Pt(dmit) C(1)MC(2) 1.303(9) PtMS(7) 2.270(2) S(1)MC(1) 1.763(6) PtMS(8) 2.273(2) S(2)MC(1) 1.759(7) S(7)MC(17) 1.716(6) S(1)MC(5) 1.742(7) S(8)MC(18) 1.694(7) S(2)MC(6) 1.780(7) C(17)MC(18) 1.366(8) S(3)MC(2) 1.749(7) S(9)MC(17) 1.735(6) S(4)MC(2) 1.759(6) S(10)MC(18) 1.746(6) S(3)MC(3) 1.751(7) S(9)MC(19) 1.737(7) S(4)MC(4) 1.751(7) S(10)MC(19) 1.726(7) C(3)MC(4) 1.339(9) S(11)MC(19) 1.640(7) Table 4 Resistivity of cation radical salts donor (D) anion (A) D5A rrt/V cm Ea/meV S S S S S S S S S S S S S ( a ) (b ) (c ) 5a BF4- — 0.6 45 Fig. 5 Schematic representation of the coeYcients of the AET for ClO4- — 0.3 52 (a) next-HOMO, (b) HOMO and (c) LUMO PF6- 251a 0.4 50 AsF6- 251a 2 62 Although this situation would reduce the dipole oscillating Pt(dmit)2- 251 10b 120 eVect of the azulene moiety, it is still possible that the excitation 40c 120 5c PF6- 251a 3 63 of an electron from the HOMO or the next-HOMO to the 5d PF6- 251a 1 23 LUMO causes a change of electric potential at the TTF moiety 5f PF6- 351a 10 67 and thus an attractive interaction between conduction electrons. aDetermined by EPMA measurement.bMeasured along the a axis. cMeasured along the c axis. Properties of cation radical salts Cation radical salts of new donors were prepared by galvano- ring is attached to the TTF moiety.13 This fact is also supported static oxidation with various kinds of counter anions such as by the absorption in the visible region, which is characteristic BF4-, ClO4-, PF6-, AsF6- and [Pt(dmit)2]n-.All these salts of the azulene ring14 (Fig. 4). The CNC bond lengths as well show rather small resistivities but are semiconductive above as the CMS bond lengths in the TTF moiety show normal room temperature with small activation energies (Table 4). values compared to those in the known neutral TTF Since the quality of the crystals is not so good in general, derivatives.15 The molecular plane bends slightly at S(3) and X-ray structural analysis was performed only for S(4), which is the same with the molecular structure of the (AET)2[Pt(dmit)2].Crystal data are summarized in Table 2. neutral BEDT–TTF.15 There is a positional disorder in the The unit cell contains two mixed columns which are crystallo- terminal ethylenedithio fragment as reflected by the elongated graphically equivalent (Fig. 6). In each column, a repeating thermal ellipsoid for C(7) and C(8) as well as slightly larger unit consists of two AET molecules and one Pt(dmit)2 mol- atomic displacement parameters (Beq=5.19 and 4.58, respectecule. The two AET molecules are interrelated by the inversion ively) of S(5) and S(6). AET molecules are dimerized and the centre and the Pt(dmit)2 molecule is located on the inversion molecules in the dimer are interrelated by the inversion centre centre.The AET molecule is almost planar and there is no as shown in Fig. 3(c). No S,S contact shorter than the sum positional disorder at the terminal ethylene group. Interplanar of the van der Waals radii (3.60 A ° ) is observed among the distances are 3.4 A ° between AET molecules and 3.7 A ° between donor molecules.AET and Pt(dmit)2 molecules. Short intermolecular S,S The molecular orbitals of AET have been calculated by a distances are observed only between AET molecules arranged semiempirical method (MOPAC: PM3 hamiltonian) using the in a side-by-side fashion (Fig. 7). coordinates of neutral AET obtained by X-ray analysis (Fig. 5). The bond lengths in the TTF moiety are known to be The coeYcients of the LUMO are distributed on the azulene sensitive to the formal charge, and oxidation of TTF should moiety, and its electronic structure is approximately the same result in an increase of the central CNC bond length.16 The as that of the simple azulene [cf. Fig. 2(b)]. On the other hand, central C(1)MC(2) bond length of AET is slightly shorter than the coeYcients of the HOMO and the next-HOMO are that of the neutral AET, and other bond lengths within the distributed on both azulene and TTF moieties, presumably TTF moiety are almost the same as those in the neutral one because their energy levels are very close to each other.(Table 3). This means that the AET molecule is nearly neutral in this crystal.As for the formal charge of [Pt(dmit)2]n-, it is diYcult to discuss a diVerence of the formal charge from its PtMS bond lengths,17 although a diVerence of the oxidation state from n=2 to n=0 is expected to result in a decrease of Fig. 4 Visible spectrum of (a) AET and (b) 5,6-dichloroazulene Fig. 6 Crystal packing in (AET)2[Pt(dmit)2] J. Mater. Chem., 1998, 8(2), 289–294 291Fig. 7 (a) Overlapping mode of donor molecules; (b) side-by-side contacts between donor molecules in the (AET)2[Pt(dmit)2] crystal Fig. 9 Band dispersion of (AET)2[Pt(dmit)2] calculated by the tightbinding method. The level of the AET HOMO is the origin of energy. be noted that the ordinary mixed-stack system is a semiconductor, even if there is enough charge transfer.If the transverse interactions are enhanced, however, the system can achieve a metallic band structure.19 Indeed, when small DE values are adopted, our system exhibits a Fermi surface. Therefore, it would be possible to obtain a metallic system by the choice of an appropriate acceptor. Conclusion Fig. 8 Molecular arrangement viewed along the b axis We have proposed that Little’s high-temperature superconductivity theory can be applied to molecular conductors.The this bond length. The PtMS bond lengths (about 2.27 A° ) are azulene moiety of the donor AET retains the electronic struc- shorter than those in TTF[Pt(dmit)2]3 (about 2.30 A ° ),18 ture of the simple azulene, although partial mixing of the which suggests that the Pt(dmit)2 molecule in our crystal is azulene HOMO and the TTF HOMO was suggested by approximately neutral.semiempirical MO calculations. At present, unfortunately, Intermolecular overlap integrals among frontier orbitals cation radical salts of the donors are all semiconductive. The [HOMO for AET and LUMO for Pt(dmit)2] illustrated in crystal and electronic structures of (AET)2[Pt(dmit)2] suggest Fig. 8 are shown in Table 5.One unknown band parameter is one condition for the metallic state. Research towards the the energy diVerence (DE) between the LUMO and HOMO. achievement of the metallic state is in progress. Considering that the amount of charge transfer from AET to Pt(dmit)2 is small and the system is a semiconductor with a band gap (Eg) of 0.24 eV [Eg is twice as large as the activation Experimental energy (Ea=0.12 eV)], the DE value is estimated to be about 0.3 eV so that Eg is consistent with the resistivity measurement.All reactions were carried out under an Ar atmosphere. 5,6- It has been assumed that the transfer integral (t) is proportional Dichloroazulene20 and units 4a–4f21–24 were synthesized to the overlap integral (S), t=eS (e=-10 eV, e is a constant according to the methods described in the literature. with the order of the orbital energies of the HOMO and LUMO).The band structure was calculated based on the Azuleno[5,6-d]-[1,3]-dithiole-2-thione 3 tight-binding approximation and is displayed in Fig. 9. The highest-lying (mainly) LUMO band is separated far from the A dimethylformamide (DMF) (300 ml) solution of 5,6-di- (mainly) HOMO bands.Since the degree of the band filling is chloroazulene (10.10 g, 51.3 mmol) was added dropwise into 2/3, the system is a semiconductor with an energy gap. In this an aqueous solution (30 ml ) of potassium sulfide (8.51 g, case, the large DE value (and thus small charge transfer) is one 77.2 mmol) at room temp. After 15 min, carbon disulfide of the main causes of the semiconductive behaviour.It should (50 ml ) was also added and the reaction mixture was stirred at 60 °C for 16 h. The resultant mixture was poured into Table 5 Intermolecular overlap integrals (S) of frontier orbitals in benzene (1 l ), and filtered through Celite. This benzene solution (AET)2[Pt(dmit)2] was washed with water (500, 300, 300 ml), and dried with MgSO4. The solvent was removed under reduced pressure, S/10-3 and the silica gel column chromatography with carbon disulfide as eluent gave pure 3 (3.57 g, 30%) as green fibres.Mp a1 -5.41 a2 -4.54 151–154 °C (decomp.) (Calc. for C11H6S3: C, 56.37; H, 2.58. c1 -0.079 Found: C, 56.29; H, 2.77%); (HRMS: Calc. 233.9632. Found c2 -1.31 233.9603); dH (CDCl3–CS2) 7.13 (1H, d, J 10.2), 7.35 (1H, d, p -0.62 J 4.0), 7.42 (1H, d, J 3.7), 7.86 (1H, t, J 3.8), 8.13 (1H, d, q -1.02 J 10.6), 8.20 (1H, s) (J values in Hz throughout). 292 J. Mater. Chem., 1998, 8(2), 289–294Table 6 Semiempirical parameters for Slater-type atomic orbitals Synthesis of donors 5a–5f: general procedure A triethyl phosphite (5 ml) solution of 3 (1 mmol) and unit S C H Pt 4a–4f was heated to 100 °C for 30 min.It was cooled to room 3s 3p 2s 2p 1s 6s 6p 5da temp., methanol (30 ml ) was added and the reaction mixture filtered through a glass filter. The brown solid obtained was f1 2.12 1.83 1.63 1.63 1.30 2.55 2.55 6.01(0.633) purified by silica gel chromatography with carbon disulfide as z2 2.70(0.551) eluent to aVord pure 5a–5f. -Ip/eV 1.47 0.79 1.57 0.84 1.0 0.67 0.40 0.93 aTwo Slater exponents were used for the 5d functions.Each is followed 2-(Azuleno[5,6-d][1,3]dithiol-2-ylidene)-5,6-dihydro-1,3-diin parentheses by the coeYcient in the double zeta expansion. thiolo[4,5-b][1,4]dithiine 5a. 44%, mp>300 °C (Calc. for C16H10S6: C, 48.69; H, 2.55. Found: C, 48.69; H, 2.73%) Electrical resistivity measurements (HRMS: Calc. 393.9107. Found 393.9127); dH (CDCl3–CS2) 3.27 (4H, d, J 1.3), 6.97 (1H, d, J 9.9), 7.14 (1H, d, J 4.3), 7.21 The direct current resistivity measurements were performed (1H), 7.70 (1H, t, J 3.8), 7.93 (1H, d, J 10.2), 8.04 (1H, s).with the standard four-probe method. Gold leads (15 mm diameter) were attached to the crystal with carbon paste. 2-(Azuleno[5,6-d][1,3]dithiol-2-ylidene)-1,3-dithiolo[4,5-d]- [1,3]dithiole 5b. 49%, mp>300 °C (Calc. for C15H8S6: C, 47.33; Crystal structure analysis H, 2.73. Found: C, 47.08; H, 2.28%) (HRMS: Calc. 379.8950. Found 379.8940); dH (CDCl3–CS2) 4.95 (2H, s), 6.97 (1H, d, J X-Ray diVraction data for AET were collected on a MAC 10.2), 7.14 (1H, d, J 4.3), 7.22 (1H), 7.69 (1H, t, J 3.8), 7.94 Science automatic four-circle diVractometer (MXC18) with (1H, d, J 10.2), 8.02 (1H, s).graphite-monochromated Mo-Ka radiation up to 2h=60°. The intensities were corrected for Lorenz and polarization eVects. 2-(Azuleno[5,6-d][1,3]dithiol-2-ylidene)-5,6-dihydro-1,3-di- The data for (AET)2[Pt(dmit)2] were collected on a MAC thiolo[4,5-b][1,4]diselenine 5c. 34%, mp 286–289 °C (decomp.) Science Weissenberg-type imaging plate system (DIP320S). (Calc.for C16H10S4Se2: C, 39.34; H, 2.06. Found: C, 39.19; The cell constants were refined by the four-circle diVractometer H, 2.19%); dH (CDCl3–CS2) 3.35 (4H, s), 6.96 (1H, d, J 10.2), with monochromated Mo-Ka radiation up to 2h=60°. 7.11 (1H, d, J 4.0), 7.20 (1H, d, J 3.6), 7.66 (1H, t, J 3.8), 7.92 The structures were solved by direct methods and refined (1H, d, J 9.5), 8.00 (1H, s).using full-matrix least-squares analysis using reflections with I3s(I). An analytical absorption correction was carried out 2-(Azuleno[5,6-d][1,3]dithiolo-2-ylidene)-5,6-dihydro-1,3- for AET. Anisotropic atomic displacement parameters were diselenolo[4,5-b][1,4]dithiine 5d. 38%, mp 225–228 °C used for non-hydrogen atoms. All calculations were performed (decomp.) (Calc. for C16H10 S4Se2: C, 39.34; H, 2.06.Found: using TEXSAN the crystallographic software package from C, 39.18; H, 2.16%); dH (CDCl3–CS2) 3.30 (4H, s), 6.95 (1H, Molecular Structure Co. d, J 10.3), 7.14 (1H, d, J 3.8), 7.23 (1H), 7.69 (1H, t, J 4.0), 7.94 (1H, d, J 9.6), 8.00 (1H, s). MO calculations The molecular orbital calculation was performed using 4-(Azuleno[5,6-d][1,3]dithiol-2-ylidene)-[1,3]dithiolo[4,5- MOPAC93 included in CHEM3D from Cambridge Science c][1,2,5]thiadiazole 5e. 5%, mp 268–272 °C (decomp.) (Calc. Co. The calculation was carried out with the option C.I.=4. for C14H6N2S5: C, 46.38; N, 7.73; H, 1.67. Found: C, 46.64; In order to calculate intermolecular overlap integrals, the N, 7.45; H, 1.96%) (HRMS: Calc. 361.9135. Found 361.9148); HOMO obtained from the extended Hu�ckel MO calculation dH (CDCl3–CS2) 7.02 (1H, d, J 10.2), 7.19 (1H, d, J 3.7), 7.27 was used.The calculation was carried out with the use of (1H, d, J 3.3), 7.73 (1H, t, J 3.6), 7.99 (1H, d, J 10.2), 8.06 semiempirical parameters for Slater-type atomic orbitals25 (1H, s). (Table 6). 2-(4,5-Diiodo-1,3-dithiol-2-ylidene)azuleno[5,6-d][1,3]- The authors are grateful to Takeda Chemical Industry Co.dithiole 5f. 17%, mp>300 °C (Calc. for C14H6I2S4: C, 30.23; Ltd. for the supply of 3,3,4,4-tetrachlorothiophene 1,1-dioxide H, 1.09. Found: C, 30.15; H, 1.16%); dH (CDCl3–CS2) 6.96 as a starting material for 5,6-dichloroazulene. (1H, d, J 10.3), 7.13 (1H, d, J 3.3), 7.20 (1H), 7.68 (1H, t, J 4.0), 7.93 (1H, d, J 10.9), 7.99 (1H, s). References Cyclic voltammetry measurements 1 K.Kagoshima, H. Anzai, K. Kajimura and T. Ishiguro, J. Phys. The cyclic voltammetry experiments were all performed under Soc. Jpn., 1975, 39, 1143; F. Denoyer, F. Come`s, A. F. Garito and argon atmosphere at room temp. A solution of tetra(n-butyl )- A. J. Heeger, Phys. Rev. L ett., 1975, 35, 445. ammonium perchlorate–benzonitrile (0.1 M), Pt working and 2 J.F. Kwak, J. E. Schirber, R. L. Greene and E. M. Engler, Phys. Rev. L ett., 1981, 46, 1296. auxiliary electrodes were used. Potentials were referenced vs. 3 J. W. Bray, H. R. Hart, Jr., L. V. Interrante, I. S. Jacobs, Ag/0.01 M AgNO3. Sweep rate was 100 mV s-1 in every J. S. Kasper, G. D.Watkins, S. H.Wei and J. C. Bonner, Phys. Rev. experiment. L ett., 1975, 35, 744. 4 D.Jerome, A. Mazaud, M. Ribault and K. Bechgaard, J. Phys. Preparations of cation radical salts L ett. (Paris), 1980, 41, 95. 5 J. Wosnitza, Fermi Surfaces of L ow-Dimensional Organic Metals Cation radical salts of 5a–5f were obtained by galvanostatic and Superconductors, Springer Tracts in Modern Physics vol. 134, oxidation of a solution containing the donor (ca. 8 mg) and Springer, Berlin, 1996. the corresponding supporting electrolyte (35–70 mg) as tetra(n- 6 J.M.Williams, J. R. Ferraro, R. J. Thorn, K. D. Carlson, U. Geiser, H. H. Wang, A. M. Kini and M. H. Whangbo, Organic butyl)ammonium salts in 1,1,2-trichloroethane (20 ml, contain- Superconductors (Including Fullerenes): Synthesis, Structure, ing 5% of ethanol as stabilizing reagent) under an argon Properties and T heory, Prentice Hall, Englewood CliVs, NJ, 1992.atmosphere at 20 °C. An H-shaped cell and platinum wire 7 Organic Superconductivity, ed. V. Z. Kresin and W. A. Little, electrodes (1 mm diameter) were employed and a constant Plenum Press, New York, 1990. current (0.5 mA) was applied for 1–4 weeks. Crystals formed 8 W. A.ittle, Phys. Rev., 1964, 134, A1416. in the anode compartment were collected and washed with 9 J.Bardeen, L. N. Cooper and J. R. SchrieVer, Phys. Rev., 1957, 108, 1175. acetone and n-hexane. J. Mater. Chem., 1998, 8(2), 289–294 29310 D. M. Lemal and G. D. Goldman, J. Chem. Educ., 1988, 65, 923. 18 M. Bousseau, L. Valade, J. P. Legros, P. Cassoux, M. Garbauskas and L. V. Interrante, J. Am. Chem. Soc., 1986, 108, 1908. 11 G. W. Wheland and D. E. Mann, J. Chem. Phys., 1949, 17, 264; 19 H. Kobayashi, R. Kato, A. Kobayashi and Y. Sasaki, Chem. L ett., A. G. Anderson and B. M. Steckler, J. Am. Chem. Soc., 1959, 1985, 191. 81, 4941. 20 S. E. Reiter, L. C. Dunn and K. N. Houk, J. Am. Chem. Soc., 1977, 12 R. M. Hochstrasser and L. J. Noe, J. Chem. Phys., 1969, 50, 1684. 99, 4199. 13 J. M. Robertson, H. M. M. Shearer, G. A. Sim and D. G. Watson, 21 N. Svensrup and J. Becher, Synthesis, 1995, 215. Acta Crystallogr., 1962, 15, 1. 22 R. Kato, H. Kobayashi and A. Kobayashi, Synth. Met., 1991, 14 J. Brunken, Chem. Ber., 1960, 93, 2572; E. W. Thulstrup, P. L. Case 41–43, 2093. and J. Michl, Chem. Phys., 1974, 6, 410. 23 A. E. Underhill, I. Hawkins, S. Edge, S. B. Wilkes, K. S. Varma, 15 H. Kobayashi, A. Kobayashi, Y. Sasaki and G. Saito, Bull. Chem. A. Kobayashi and H. Kobayashi, Synth.Met., 1993, 55–57, 1914. Soc. Jpn., 1986, 59, 301. 24 T. Imakubo, PhD Thesis, University of Tokyo, 1996. 16 H. Kobayashi, A. Kobayashi, Y. Sasaki, G. Saito and H. Inokuchi, 25 R. H. Summerville and R. HoVmann, J. Am. Chem. Soc., 1976, Chem. L ett., 1984, 183. 98, 7240. 17 L. Valade, J. P. Legros, M. Bousseau, P. Cassoux, M. Garbauskas and L. V. Interrante, J. Chem. Soc., Dalton T rans., 1985, 783; G. N. Schrauzer, Acc. Chem. Res., 1969, 2, 72. Paper 7/03737F; Received 29th May, 1997 294 J. Mater. Chem., 1998, 8(2), 289–294