Development of Conductive Organic Molecular Assemblies

8/10/2019 Development of Conductive Organic Molecular Assemblies

1/40

1

1

Development of Conductive Organic Molecular Assemblies:Organic Metals, Superconductors, and Exotic Functional Materials

We aim at developing the following functional molecular assemblies (A~ E).A) Molecular conductors: organic superconductors, organic metals, and organic semiconductors.B) Phase transitions: switching-memory materials, metal-insulator, neutral-ionic, proton transfer-chargetransfer, photo-induced phase transition (PIPT), etc.C) Single component functional materials: fastener molecules, D--A, mesomeric zwitter ions (betaines).D) Ionic liquids and conductive liquids including soft conductors.E) Organic-inorganic functional hybrids.

The research work is performed by the sequence of the following steps 1 ~ 5.1: Design of the molecules, molecular assemblies, and functions based on the real and energy spaces.2: Preparation of molecules and molecular assemblies.3: Structural analysis.4: Measurement of physical properties.5: Understanding of the structures and functions of the molecular assemblies based on the intermolecular

interactions.

Important objective is the development of new functions based on the establishment of relationbetween structure and functionalityand control of the physical and chemical parameters relevant to thetargeted structures and functions.

Our research interests have straddled a number of conductive and magnetic organic assembliesincluding organic metals, superconductors, spin-ladders, spin-liquid, etc. The control of the chemical and

physical parameters for electric transport, such as ionization potential (IP), electron affinity (EA), Madelungenergy (M), on-site (U) and off-site (V) Coulomb energies, density of state (D(F)), Fermi level (F: chemicalpotential), bandwidth (W), transfer interaction (t), basicity (pKb), and acidity (pKa), which are controllablethrough the molecular and crystal designs, is fundamental to the progress of organic (super)conductors.

The control of the self-assembling ability, dimensionality, electron correlation, and geometry of thespin lattice is very important to design the ground state of the electronic structure (e.g., superconductor, metal,insulator, antiferromagnet, and spin-liquid). The anisotropic electronic structures of organic assemblieshaving very deformable molecular and crystal structures with strong electronphonon coupling have affordedmany fascinating electronic phase transitions caused by molecular and/or lattice deformation, chargeseparation (charge ordering, charge disproportionation), density wave formation, etc., which can be controlled

by external stimuli: light, temperature (T), pressure (P, hydrostatic or uni-axial), magnetic (H), and electric (E)

fields. These features are demonstrated in Fig. 1and the following main review articles and textbook 1 ~ 4.Main review articles and textbook for our research1) Development of Conductive Organic Molecular Assemblies: Organic Metals, Superconductors, and Exotic Functional Materials,

G. Saito, Y. Yoshida,Bull. Chem. Soc. Jpn., 80, 1-137 (2007) [BCSJ Vol.80 Commemorative Account]2) "Organic Superconductors 2nd Edition", T.Ishiguro, K.Yamaji, G.Saito, Springer, pp1-522 (1998)3) Design of Organic (Super)Conductors and Study of Their Physical Properties, G. Saito, H. Yamochi, M. Maesato, Y. Yoshida, A.

Ota, Y. Shimizu, inNATO ASI Ser.: Organic Conductors, Superconductors and Magnets: From Synthesis to Molecular Electronics,ed. by L.Ouahab, E. Yagubskii, Kluwer Academic Pub., Boston, pp.19-44 (2004)

4) Mixed Valency in Organic Charge Transfer Complexes, G. Saito, T. Murata,Phil. Trans. R. Soc. London A, 366, 139-150 (2008)

5) Frontiers of Organic Conductors and Superconductors, G. Saito, Y. Yoshida, Top. Curr. Chem.,312, 67-126(2012) in

Unimolecular and Supramolecular Electronics I. Chemistry and Physics Meet at Metal-Molecule Interfacesed. By

R.M.Metzger,

Springer.


2/40

2

2

Figure 1.Concept of the development of organic metals, superconductors, and exotic magnets is based on the dynamics of electronsin a soft, anisotropic, and nano-scaled molecular world. For inorganic electronics, charge, spin, and orbital are the three key issues.While, for organic materials, softness (lattice and molecular) and internal freedom (molecular vibration, deformation, and

polarizability) should be taken into account additionally.

Recent information

Refer to Research Abstract (2008

) (http://saitolab.meijo-u.ac.jp/research_abstract.html )
http://saitolab.meijo-u.ac.jp/research_abstract.htmlhttp://saitolab.meijo-u.ac.jp/research_abstract.htmlhttp://saitolab.meijo-u.ac.jp/research_abstract.htmlhttp://saitolab.meijo-u.ac.jp/research_abstract.html


3/40

3

3

A) How to Develop Molecular Conductors:

semiconductors, metals, and superconductors

A-1) Ionicity Diagram and Functions (metal, magnet, phase transition system)

A-1-1Partial CT (Mixed Valency) & an Ionicity-DiagramA metallic band structure is realized when the CT solids have the partial CT state and molecules form

uniform segregated columns or layers. The partial CT state can be predicted and controlled by (IDEA) orE(DA) [E(DA) =E1(D) E1(A):E1, first redox potential] for a combination of specific donor (D) andacceptor (A), and the complex DA exhibits a low lying CT band below 5103cm-1. Figure 2is a plot ofelectrical conductivity data of 1:1 low-dimensional TTFTCNQ system, vs.redox potentials.[1]The two lines aand b are related to the equation expressing the relationship amongIP,EA, and the Madelung energyM(:degree of CT) between partially charged component molecules (Eq. 1).[2]The mixed valence region (M) is

located between fully ionic (I) and neutral (N) regions. In the region M, the CT solids are either highlyconductive () or metallic () when they have segregated stacks. The solids in the regions I and Nareinsulators (), in general.

IPEA=M() (1)

Figure 2. Ionicity diagram for TTFTCNQ system plotted as E1(A) vs. E1(D) vs. SCE after modification of the originaldiagram in Ref.1. = insulators or semiconductors; = highly conducting in compaction studies; = organic metals.Some donors are A:TTN, B:TMTTF, D:TTF, E:HMTTF, F:HMTSF, G:TMTSF, H:TSF, J:TTC1-TTF, K:ET, L:DBTTF.Some TCNQs are a:2,5-(CN)2, b:F4, e:2,5-I2, i:F, m:TCNQ, p:2,5-Et2. Complexes 1-7 are HMTTFF4TCNQ,HMTSFF4TCNQ, TTFTCNQ, TMTSFTCNQ, TSFEt2TCNQ, ETTCNQ, and DBTTFCl2TCNQ, respectively. (E1(A)andE1(D) in this figure are the peak values). Region N: neutral, M: partial CT, I: fully ionic. Line a: E(DA) = 0.02 V, b:E(DA) = 0.34 V.

Important Remarks:

1) Mixed valence state for low-D CT solids0.02 VE1(DA) +0.34 V

2) Complex isomerization (bistable CT solids)TMTSFTCNQ (4), TSFEt2TCNQ(5)BEDT-TTF(ET)TCNQ(6)ET two-dimensional conductor

3) Region I(= 1)

Mott insulator(1, 2), antiferromagnet, spin-Peierls,FET channel, switching (memory) system

4) Region M(0.5 < 1)metal(3, 7, LB films), electrode (for FET), transparent metal ofBO complex, switching, capacitor

5) Region N(< 0.5)non-linear optics, FET channel, solar cell

S

SX

X S

S

X

X

X=S: ET, X=O: BO

S

S

S

S

TTF

NC CN

CNNC

TCNQ

NC CN

CNNCF4TCNQ

FF

F F

Te

Te

Te

Te

HMTTFe

Scheme 1


4/40

4

4

A-1-2Two Kinds of Ionicity-Diagrams and How to Find Functional CT Solids

Figure 3shows the relationship between Fig. 2and other kind of ionicity phase diagram (V shaped linein Fig. 3b) proposed by Torrance for neutral-ionic (NI) phase transition system for the alternating stacks.[3]Figure 3a was made by rotating Fig. 2 (E1(D) and E1(A) have the same scale here) so as to have two

borderlines a and b vertical. Then, every CT solid allocated on the horizontal line in Fig. 3ahave the samechemical potential (IP+EA= constant). The organic metals in the region Mresiding on the varied horizontallines in Fig. 3a were employed as the source and drain electrodes to control the Fermi level alignment

between electrodes and channel for FETs making the injection of carriers smooth and giving varied polarityin FET behavior (see Section A-4-5).[4a]

In the neutral region near the bottom of the V-shaped line (region MA), an enantiotropic phase

transition system (N

I transition) is allocated.

[3]

The CT solids that have hCTbands below 5 10

3

cm

1

Figure 3. (a) The same figure as Fig. 2except the scale ofE1(D or A). Organic metals on the bluedotted line have the same chemical potential (Ip+EA = constant). (b) Schematic phase diagram ofionicity, conductivity and stacking of DA CT solids. The first transition energy in solid (hCT) is

plotted against theE(DA) value. Left- and right-hand sides of the V-shaped line correspond tohCTN=IPEA C Eq. 2hCT

I= IP+EA+ (2 1)C Eq.3,respectively, where C and C are the Coulomb attractive energy between D+ and A and theMadelung energy, respectively.IA: ionic alternating, MA: mixed valence alternating, NA: neutral alternating, NS: neutral segregated,MS: mixed valence segregated (MS-1: non 1:1 (one component is fully ionic), MS-2: 1:1low-dimensional,MS: 1:1 high-dimensional). An appropriate V-shaped line for thep-quinone systemwas obtained by a parallel shift of the V-shaped line for the TCNQ system towards the lower side by0.130.16 V.

(a)

(b)


5/40

5

5

(horizontal green dotted line) belong to a different class (region MS) that usually includes (super)conductorsand narrow-gap semiconductors having the mixed valence segregated stacks or layers. Important point here isthat MSis for high-dimensional 1:1 CT solids, which usually extend their metallic regime toward lower values (higher E(DA) values) like the HMTTeF and BEDO-TTF (BO) systems owing to their strongself-assembling ability. Even with = 1/3, some BO complexes have segregated stack and show metallic

behavior. Figures 2 and 3gave us a prediction on the modes of molecular stacking: alternatingand segregated,and information on the chemical potentials and other functionalities as described below.

A-1-3Variety of Functional CT Solids Derived from Ionicity Diagrams

Ionicity diagrams as depicted in Figs. 2and3are clues to explore functional conductors of CT type suchas molecular metals, Mott insulators, NI system, complex isomers, and self-assembled two-dimensionalconductors.

1) The fully ionic solids (region I) afforded band insulators, 1:1 Mott insulatorswith ground states ofantiferromagnets(Eb and Fb) or spin-Peierls, ferroelectrics, ferromagnets, spin-ladders, and non-linear

transport materials (switchingand memory).2) The mixed valence (region M in Fig. 1a, and MA, MS-1, MS-2 and M'S in Fig. 3b) afforded(super)conductors and the following various kinds of insulators: a) a (nearly) uniform segregated stackhaving spin density wave (SDW), and anion or charge ordered(= charge disproportionation) state, b) anon-uniform segregated stack showing Peierls-type distortion, spin-Peierls distortion, and dimer-type Mottinsulators including antiferromagnets, spin-liquidand spin-ladders, and c) an alternating stack includingN

I systems, ferroelectrics, and highly-conductive semiconductors. There are hybrids by the combination offerro-, ferri- or paramagnetic species based on transition metal compound as one component and mixedvalence counterpart to form magnetic CT conductors.

3) The neutral solids (regions Nand NA+ NS), in general, exhibit a CT band represented by Eq. 2regardless of the stacking modes. Since most of CT solids in the region Nprefer alternating stackswith a

few exceptions, they are band insulators with low ionicity. Hydrogen-bondand proton-transferbetweenthe components manifest many interesting functions; switching, ferroelectrics, etc.

4) Near the borderline b in Fig. 2, the bistabilityconcerning the ionicity between the neutral and partialCT states is realized, i.e., the monotropic complex isomers Gm, Hp, and Km. Even though Km(ETTCNQ) is expected to afford a neutral insulator based on its E(DA) value, a highly conductive complexisomer has been prepared. This result indicates that the ET molecule has a significant self-assembling abilityto form a segregated column with increased dimensionality, which is a nature of the solids in the regionMSin Fig. 3b(see Section A-2-1). The insulating CT solids residing near the borderline a or b have the

potential to exhibit a phase transitioninto a highly conducting phase induced by external stimuli (electricfield, photo, etc.) with smaller threshold than those allocating far from the borderlines.

Similar diagrams for other systems have been proposed for BO,

[4b]

EOET,

[4c]

HMTTeF,

[4d]

EDO,

[4e,f]

p-phenylenediamine,[5] benzidine,[4g] 1,4,6,8-tetrakis(dimethylamino)pyrene (TDAP),[4h] C60,[4i]

andCF3TCNQ

[4j] concerning the charge transfer interaction, and for anilinepicric acid,[4k,l]2,2'-bi-1H-imidazole(H2BIM),

[4m] and dihydrotetracyanodiphenoquinodimethane(4,4-bis(dicyanomethyl)biphenyl) (H2TCNDQ),

[4n]concerning the charge and proton transfer interactions.

Scheme 2

NH2

NH2

NH2H2NOH

O2N NO2

NO2

NMe2Me2N

NMe2Me2N

CH(CN)2(NC)2HCNHN

NHN

TDAP

H2BIM H2TCNDQ

S

S

S

S

S

SO

O

EOET


6/40

6

6

Reference A-11. G. Saito, J. P. Ferraris,Bull. Chem. Soc. Jpn., 53, 2141-2145 (1980)2. H. M. McConnell, B. M. Hoffman, R. M. Metzger,Proc. Natl. Acad. Sci., USA, 53, 46-50(1965).3. J. B. Torrance, J. E. Vazquez, J. J. Mayerle, V. Y. Lee,Phys. Rev. Lett.,46, 253-257(1981).

4. a) Y. Takahashi, T. Hasegawa, Y. Abe, T. Tokura, G. Saito,Appl. Phys. Lett., 88, 073504/1-4 (2006)b) S. Horiuchi, H. Yamochi, G. Saito, K. Sakaguchi, M. Kusunoki,J. Am. Chem. Soc.,118, 8604-8622 (1996)c)G. Saito, H. Sasaki, T. Aoki, Y. Yoshida, A. Otsuka, H. Yamochi, O. O. Drozdova, K. Yakushi, H. Kitagawa, T. Mitani, J. Mater. Chem.,12, 1640-1649 (2002)d) S. S. Pac, G. Saito,J. Solid State Chem.,168, 486-496 (2002)e) A. Ota, H. Yamochi, G. Saito,J. Mater. Chem., 12, 2600-2602 (2002)f) A. Ota, H. Yamochi, G. Saito,Mol. Cryst. Liq. Cryst.,376, 177-182 (2002)g) Y. Matsunaga, G. Saito,Bull. Chem. Soc. Jpn.,44, 958-963(1971).h) G. Saito, S. Hirate, K. Nishimura, H. Yamochi,J. Mater. Chem.,11, 723-735(2001).i) G. Saito, T. Teramoto, A. Otsuka, Y. Sugita, T. Ban, M. Kusunoki, K. Sakaguchi, Synth. Metals, 64, 359-368(1994).j) G. Saito, H. Ikegami, Y. Yoshida, O. O. Drozdova, K. Nishimura, S. Horiuchi, H. Yamochi, A. Otsuka, T. Hiramatsu, M. Maesato, T. Nakamura, T. Akutagawa, T. Yum,

Bull. Chem. Soc. Jpn., 83, 1462(2010)k) G. Saito, T. Inukai, Crystal Growth(Japanese), 16, 2-16(1989).l) G. Saito, Y. Matsunaga,Bull. Chem. Soc. Jpn.,44, 3328-3335(1971).m) T. Akutagawa, G. Saito, M. Kusunoki, K. Sakaguchi,Bull. Chem. Soc. Jpn., 69, 2487-2511(1996).n) K. Nishimura, S. S. Khasanov, G. Saito,J. Mater. Chem., 12, 1693-1702(2002).

5. Y. Matsunaga,Bull. Chem. Soc. Jpn.,42, 2490-2493(1969)


7/40

7

7

A-2) Two-dimensional Organic Metals & Superconductors

A-2-1 Molecular Design for Dimensionality: Self-Assembling AbilitySince the metallic state in the one-dimensional electronic system is unstable, an increase in the

electronic dimensionality is necessary to prevent the nesting of Fermi surfaces. Several attempts have beenmade through "pressure", "heavy atom substitution" or "peripheral addition of alkylchalcogeno groups"(Fig.4). The latter two correspond to the enhancement of the self-assembling ability of the molecules andhence increase the electronic dimensionality of molecular assemblies. The HMTTeF molecules affordedstable metallic phase without any trace of superconductivity.[A-1-4d]The BO molecules also afforded stabletwo-dimensional metals having two-dimensional Fermi surface owing to the strong self-assembling ability(see Section A-4-2).[A-1-4b] To destabilize the metallic state of the BO complexes, the substitution of oneethylenedioxy group with ethyleneditio group (BOEOET) or the elimination of one ethylenedioxy group(BOEDO) was found to be very efficient to provide localized (magnetic) phase or metal-insulator (MI)transition.[A-1-4e,f]Several superconductors have been prepared based on TMTSF, ET, BO and a variety ofanalogues of TTF even though TTF itself did not afford superconductors. EDO and DIETSe afforded

intriguing MI transition originated partly from their electronic one-dimensionality (see Sections B-2 and B-4).

Figure 4. Strategy for chemical modification of the TTF molecule to increase (red arrows) or decrease (blue arrows) theelectronic dimensionality by the aid of enhance or suppress the self-assembling ability of the donor molecules, respectively.Typical Fermi surfaces of TMTSF (a: (TMTSF)2NbF6), BO (b: (BO)5HCTMM(benzonitrile)2), ET (c:-(ET)2I3), EDO (d:(EDO)2PF6), and DIETSe (e: (DIETSe)2FeCl4) CT solids are depicted.

S

SO

O S

S

EDO S

SH2n+1CnS

H2n+1CnSS

S

SCnH2n+1

SCnH2n+1

TTCn-TTF

S

S

S

S

Se

Se

Se

Se Me

Me

Me

MeTTFTMTSF

Te

Te

Te

Te

HMTTeF

S

SO

O S

S

O

O

BEDO-TTF(BO)

S

SS

S S

S

S

S

BEDT-TTF(ET)

1DSuper 2DStable Metal

2D Stable Metal

1D Metal

Ultrafast

T

2DUnstable Metal2D Super

High Mobility Fastener Effect

Atomic-Wire Effect

Se

SeS

S Se

Se

DIETSe

1D Metal

Strong d-

I

I

H

H

Heavy AtomSubstitution

AlkylchalcogenoPeripheralAddition

Uncapped

Increase ofself-assemblingability

Decrease ofself-assemblingability

(a)

(b) (c)

(d)(e)

CNNC

CN CN

CNNC

HCTMM2-


8/40

8

8

A-2-2Two-Dimensional ET Conductors and Superconductors

We have proposed an idea to increase dimensionality of CT solids in orderto suppress the Peierls type MI transition by using TTF derivative with peripheraladdition of alkylchalcogeno groups and found the first two-dimensional organicmetal (ET)2(ClO4)(1,1,2-trichloroetnane).

[1] Since then, hundreds of ET solidshave been prepared. Different kinds of ETET (-, SS) and ETanion(hydrogen bonds) intermolecular interactions, large conformational freedom ofethylene groups, flexible molecular framework, fairly narrow bandwidth (W) andstrong electron correlations (Ueff) gave a rich variety of complexes with differentcrystal and electronic structures ranging from insulators to superconductors.About 80 ET superconductors are so far known. Currently'-(h8-ET)2ICl2(on-set Tc= 14.2 K at 8.2 GPa,

[2]

Table 1.Four typical

-type superconductors withTcabove 10 K (2 5) and a Mott insulator (7)

Ligand L1forms infinite chain by the coordination to Cu(I). Ligand L2coordinates to Cu(I) as pendant.

midpoint Tcof 13.4 K is estimated) and -(d8-ET)2Cu[N(CN)2]Cl (Tc= 13.1 K at 0.03 GPa)[3] show the

highest Tc under pressure, while both are Mott insulators at ambient pressure. At ambient pressure,-(d8-ET)2Cu(CN)[N(CN)2] shows the highest Tcof 12.3 K

[4] followed by -(h8-ET)2Cu[N(CN)2]Br (Tc=11.8 K).[5] The four -type superconductors -(ET)2CuL1L2 (L1, L2 = Cl, Br, NCS, N(CN)2) share somecommon structural and physical properties. Table 1summarizes the two kinds of ligand in a salt, Tcof H- andD-salts (salt using h8-ET and d8-ET, respectively), ratio of transfer interactions t/tfor triangle geometry of ET

Number in Fig. 6and Ligand

LigandL1 L2

Tc/KH-salt D-salt

t'/t U/W year, group

3) Cu(NCS)2 SCN NCS 10.4 11.2 0.81-0.86 0.94 1988, Saito4) Cu[N(CN)2]Br N(CN)2 Br 11.8 11.2 0.67 0.92 1990, Argonne5) Cu[N(CN)2]Cl N(CN)2 Cl 12.8(0.03GPa) 13.1 0.75 0.90 1990, Argonne2) Cu(CN)[N(CN)2] CN N(CN)2 11.2 12.3 0.66-0.71 0.87 1991, Saito7) Cu2(CN)3 CN CN 6.8-7.3 1.06 0.9 1991,Argonne, Saito

(a)

Figure 5-(ET)2Cu(NCS)2: a) Crystal structure: two-dimensional conducting ET layer is sandwiched by the insulating anion layersalong the a-axis (bc-plane is two-dimensional conducting plane). Two kinds of layers are Josephson coupled. b) Anion structure:CuSCNCuSCNforms zigzag infinite chain along the b-axis and other ligand SCN (Ligand L2in Table 1) coordinates to Cu(I) by

N atom to make a space (indicated by red ellipsoid) to which an ET dimer fits. Picture is the dextrorotatory form. c) Reflecting thecrystal symmetry, the calculated Fermi surfaces of theP21salts (-(ET)2Cu(NCS)2, -(ET)2Cu(CN)[N(CN)2]) showed the certain

energy gap between a one-dimensional electron like Fermi surface and a two-dimensional cylindrical hole-like one (-orbit), whilesuch a gap is absent in thePnma salts (-(ET)2Cu[N(CN)2]Br, -(ET)2Cu[N(CN)2]Cl). For -(ET)2Cu(NCS)2, electrons move alongthe closed ellipsoid (-orbit) then at higher magnetic field (> 20T) electron hops from the ellipsoid to open Fermi surface to showcircular trajectory (-orbit, Magnetic breakdown). d) Single crystals by electrooxidation.

(b) (c) (d)

S

SS

S S

S

S

S

h8-ET

H

H

HH

HH

HH

S

SS

S S

S

S

S

d8-ET

D

D

D

D

DD

D

D

Scheme 3


9/40

9

9

molecules (see Section A-3-2), U/W, and years and group of discovery. Figure 5shows the crystal structure[6]of the prototype -(ET)2Cu(NCS)2, anion structures, calculated Fermi surface, and micrograph of singlecrystals. Although these ET salts have similar structural aspects, their transport properties differ apparently(Fig. 6). -(ET)2Cu(CN)[N(CN)2] (2) showed a monotonous decrease of resistivity with upper curvaturedown to Tc. -(ET)2Cu[N(CN)2Br] (4) exhibited a similar behavior to that of -(ET)2Cu(NCS)2(3) without ametallic regime near RT. -(ET)2Cu[N(CN)2]Cl (5) showed a semiconductor (g= 24 meV)semiconductor(g = 104 meV) transition at ca. 42 K due to an antiferromagnetic (AF) fluctuation resulting in a weakferromagnet below 22 K.[7,8]Under a weak pressure, it showed a similar temperature dependence to that of-(ET)2Cu[N(CN)2]Br. -(h8-ET)2Cu[N(CN)2]Cl shows a complicated T-Pphase diagram as elucidated byIshiguro and Ito et al.[9] Thoroughgoing studies under He gas pressure shows a firm evidence of thecoexistence of superconducting (I-SC-2 phase: I-SC = incomplete superconducting) and AF phases,[9d,10,11]

where the radical electrons of ET molecules play both roles of localized and itinerant ones. Under a pressureof ca. 20-30 MPa another incomplete superconducting phase (I-SC-1) appears and the completesuperconducting (C-SC) phase neighbors to this phase at higher pressures. Below these SC phases, reentrant

nonmetallic (RN) phase was observed. Similar T-P phase diagrams were obtained for -(d8-ET)2X (X =Cu[N(CN)2]Cl[9f] and X = Cu[N(CN)2]Br

[9g-i] with a parallel shift of pressure. They are allocated at thehigher and lower pressure sides of the -(h8-ET)2Cu[N(CN)2]Cl for the Br and Cl salts, respectively. Contraryto the H salt, -(d8-ET)2Cu[N(CN)2]Cl did not exhibit the coexistence of the superconducting andantiferromagnetic phases and hence afforded antiferromagnetic resonance.

With increasing the distance between the ET dimers in Fig. 5a,b, the transfer interactions between ETdimers decrease that may correspond to the decrease of band width and to increase of D(F), andconsequently Tcis expected to increase. According to this line, higher Tcis expected for the salt having largeranion spacing. Such a -type salt may be allocated near the border between poor metals and Mott insulators.

Fermi surfaces of them has been studied by the quantum oscillations such as SdH(Shubnikov-de Haas),dHvA(de Haas-van Alphen), and AMRO(angular dependent magnetoresistance oscillation).[12]Fermi surfaceof -(ET)2Cu(NCS)2 (Fig. 5) calculated based on the crystal structure is in good agreement with thoseobserved data.[13]

Strange behavior of 3 and4: The semiconductive region of 3 5 and7, which is called as paramagneticnon-metallic phase, is postulated to be a Mott insulating state. However the enhancement in the magneticsusceptibility (4.4 4.7 104emu mol1at RT) due to the electron correlation is not significant compared tothose of the typical Mott insulators (9 10 104emu mol1,-(ET)2X X=ICl2, AuCl2) and comparable tothose of good metal (-(ET)2X, X=I3, AuI2, 3.4 4.6 104emu mol1). The magnetic susceptibility, optical,and thermopower measurements confirmed that the semiconductive-like region of 3 and 4 is neither the

typical Mott insulator nor typical metal. The origin of the regime is still controversial, but the strong electron

Figure 6.Temperature dependences of resistivity of 10 K class superconductors -(ET)2Cu(CN)[N(CN)2] (2), -(ET)2Cu(SCN)2(3), -(ET)2Cu[N(CN)2]Br (4), -(ET)2Cu[N(CN)2]Cl] (5)) are compared with those of a good metal with low Tc-(ET)2AuI2(1), strongly electron correlated insulator -(ET)2Cu2(CN)[N(CN)2]2(6) and a Mott insulator -(ET)2Cu2(CN)3(7) at AP.


10/40

10

10

correlation is the most plausible cause. 3shows metallic temperature dependence above 270 K.

A-2-3 Superconducting Characteristics

1)HC2: -(ET)2Cu(NCS)2gave higherHc2values in the two-dimensional plane thanHPauli.[14]

2) Symmetry of superconducting state: No Hebel-Slichter coherence peak was observed in both-(ET)2Cu(NCS)2and -(ET)2Cu [N(CN)2]Br in

1H NMR measurements, ruling out the BCSs-wavestate. The symmetry of the superconducting state of -(ET)2Cu(NCS)2had been controversially describedas normal BCS-type or non-BCS type, however, STM spectroscopy showed the d-wave symmetry withline nodes along the direction near /4 from the ka- and kc-axes (dx2y2)[15], and thermal conductivitymeasurements were consistent with that.[16a]STM on -(ET)2Cu[N(CN)2]Bralso showed the same symmetry.

A recent specific heat measurement on -(ET)2Cu(NCS)2and -(ET)2Cu [N(CN)2]Br was consistent withthese results.[16b]

3) Inverse isotope effect: Inverse isotope effect has so far been observed for -(ET)2Cu(NCS)2,[17]-(ET)2

Cu(CN)[N(CN)2][4]and-(ET)2Cu [N(CN)2]Cl

[5]and normal isotope effect for -(ET)2Cu[N(CN)2]Br.[18]

The reason of the observed isotope effects is not fully understood yet consistently.4) Phase diagram:A very simplified T-Pphase diagram for -(ET)2X was proposed (Fig. 7b), where only

the parameter U/W is taken into account. Fig. 7b includes the salts -(ET)2Cu(NCS)2, -(ET)2Cu[N(CN)2]Br and -(ET)2Cu[N(CN)2]Cl,-(ET)2I3, and-(ET)2ICl2,

[19]however, the metallic behavior of-(ET)2Cu(NCS)2 above 270 K and that of -(ET)2Cu(CN)[N(CN)2], whole behavior of-(ET)2Cu2(CN)3 , and low-temperature reentrant behavior of -(ET)2Cu[N(CN)2]Br and -(ET)2Cu[N(CN)2]Cl (Fig 7a) cannot be allocated in this diagram. The-(ET)2I3in Fig. 7b should beH-phase (Tc~8 K) and other two-(ET)2I3 salts of Tc~ 1.5 K and ~2 K phases cannot be allocated though they shouldhave the same U/W values. Tcof -(ET)2Cu[N(CN)2]Cl is higher than that of-(ET)2ICl2 in Fig. 7bcontrary to the experimentally observed Tcresults. This phase diagram and "geometrical isotope effect"

[20]point out that Tcs of-(ET)2I3, -(ET)2Cu(NCS)2, and -(ET)2Cu [N(CN)2]Brdecrease with increasing

pressure if only the parameter U/WorD(F) is taken into account. This tendency has been observed underhydrostatic pressure but not under uniaxial pressure. Thus the phase diagram in Fig. 7b remainsincomplete, despite it is frequently used to explain the general trends for these salts.

Figure 7. a) Phase diagram of -(h8-ET)2Cu[N(CN)2]Cl determined conductivity and magnetic measurements.[9a,e,f]

N1-N4:non-metallic phase, M: metallic phase, RN: reentrant non-metallic phase, I-SC-I, II: incomplete superconducting phase,S-SC:complete superconducting phase. N2shows the low-dimensional AFfluctuation. N3shows growth of three-dimensional AFordered phase. N4: weal ferromagnetic phase. B) Proposed phase diagram.[19] a: -(ET)2I3, b: -(ET)2Cu(NCS)2, c:-(h8-ET)2Cu[N(CN)2]Br, d: -(d8-ET)2Cu[N(CN)2]Br, e:-(ET)2Cu[N(CN)2]Cl, f:-(ET)2ICl2.

(a)a b c d e f(b)


11/40

11

11

References A-21. G. Saito, T. Enoki, K. Toriumi, H. Inokuchi,Solid State Commun., 42, 557-560 (1982)2. H. Taniguchi, M. Miyashita, K. Uchiyama, K. Satoh, N. Mori, H. Okamoto, K. Miyagawa, K. Kanoda, M. Hedo, Y. Uwatoko,J. Phys. Soc. Jpn., 72, 468-471(2003).

They reported Tc(on-set) = 14.2 K and Tc(midpoint) = 13.4 K at 8.2 GPa.3. J. E. Schirber, D. L. Overmyer, K. D. Carlson, J. M. Williams, A. M. Kini, H. H. Wang, H. A. Charlier, B. J. Love, D. M. Watkins, G. A. Yaconi, Phys. Rev.,B44,

4666-4669(1991).4. a) G. Saito, H. Yamochi, T. Nakamura, T. Komatsu, T. Inoue, H. Ito, T. Ishiguro, M. Kusunoki, K. Sakaguchi, T. Mori, Synth. Metals., 55-57, 2883-2890(1993).

b) Fig. 54 in G. Saito, Y. Yoshida,Bull. Chem. Soc. Jpn., 80, 1(2007).5. a) M. Kini, U. Geiser, H. H. Wang, K. D. Carlson, J. M. Williams, W. K. Kwok, K. G. Vandervoort, J. E. Thompson, D. L. Stupka, D. Jung, M. -H. Wangbo,Inorg. Chem.,

29, 2555-2557(1990).b) J. E. Schirber, D. L. Overmyer, K. D. Carlson, J. M. Williams, A. M. Kini, H. H. Wang, H. A. Charlier, B. J. Love, D. M. Watkins, G. A. Yaconi,Phys. Rev. B, 44,4666-4669(1991).

6. H. Urayama, H. Yamochi, G. Saito, K. Nozawa, T. Sugano, M. Kinoshita, S. Sato, K. Oshima, A. Kawamoto, J. Tanaka, Chem. Lett.,1988, 55-58(1988).7. J. M. Williams, A. M. Kini, H. H. Wang, K. D. Carlson, U. Geiser, L. K. Montgomery, G. J. Pyrka, D. M. Watkins, J. M. Kommers, S. J. Boryschuk, A. V. Crouch, W. K.

Kwok, J. E. Schirber, D. L. Overmyer, D. Jung, M.-H. Whangbo,Inorg. Chem., 29, 3272-3274(1990).8. U. Welp, S. Fleshler, W. K. Kwok, G. W. Crabtree, K. D. Carlson, H. H. Wang, U. Geiser, J. M. Williams, and V. M. Hitsman,Phys. Rev. Lett.,69, 840-843(1992).9. a) T. Ishiguro, H. Ito, Y. Yamochi, E. Ohmichi, M. Kubota, H. Yamochi, G. Saito, M. V. Kartsovnik, M. A. Tanatar, Yu. V. Sushko, and G. Yu. Logvenov, Synth. Metals, 85,

1471-1478(1997).b) Y. V. Sushko, H. Ito, T. Ishiguro, S. Horiuchi, and G. Saito,J. Phys. Soc. Jpn.,62, 3372-3375(1993).c) Y. V. Sushko, H. Ito, T. Ishiguro, S. Horiuchi, and G. Saito, Solid State Commun.,87, 997-1000(1993).

d) Yu. V. Shshko, K. Murata, H. Ito, T. Ishiguro, and G. Saito, Synth. Metals, 70, 907-910(1995).e) H. Ito, T. Ishiguro, M. Kubota, and G. Saito,J. Phys. Soc. Jpn., 65, 2987-2993(1996).f) H. Ito, M. Kubota, T. Ishiguro, and G. Saito, Synth. Metals, 85, 1517-1518(1997).g) H. Ito, T. Kondo, H. Sasaki, G. Saito, and T. Ishiguro, Synth. Metals, 103,1818-1819(1999).h) H. Ito, M. Watanabe, Y. Nogami, T. Ishiguro, T. Komatsu, G. Saito, and N. Hosoito,J. Phys. Soc. Jpn.,60, 3230-3233(1991).i) H. Ito, T. Ishiguro, T. Kondo, and G. Saito,J. Phys. Soc. Jpn., 69, 290-291(2000).

10. a) H. Posselt, H. Muller, K. Andres, and G. Saito,Phys. Rev.,B49, 15849-15852(1994).b) H. Posselt, K. Andres, and G. Saito,Physica B, 204, 159-161(1995).

11. S. Lefebvre, P. Wzietek, S. Brown, C. Bourbonnais, D. Jrome, C. Mzire, M. Fourmigu, and P. Batial,Phys. Rev. Lett., 85, 5420-5423(2000).12. Appendix (pp455-pp458) in T.Ishiguro, K.Yamaji, G.Saito,"Organic Superconductors 2nd Edition", Springer, pp1-522 (1998)13. K. Oshima, T. Mori, H. Inokuchi, H. Urayama, H. Yamochi, G. Saito,Phys. Rev. B,38, 938-941 (1988).14. K. Oshima, H. Urayama, H. Yamochi, G. Saito,J. Phys. Soc. Jpn.,57, 730-733 (1988).15. K. Ichimura, M. Takami, K. Nomura,J. Phys. Soc. Jpn. 77, 114707/1-6(2008).16. a) K. Izawa, H. Yamaguchi, T. Sasaki, Y. Matsuda,Phys. Rev. Lett.,88, 27002/1-4 (2002).

b) L. Malone, O. J. Taylor, J. A. Schlueter, A. Carrington,Phys. Rev. B, 82: 014522/1-5(2010)17. G. Saito, H. Urayama, H. Yamochi, K. Oshima, Synth. Met., 27, A331 (1988). Fig. 53 in G. Saito, Y. Yoshida,Bull. Chem. Soc. Jpn., 80, 1(2007).18. a) A. M. Kini, J. D. Dudek, K. D. Carlson, U. Geiser, R. A. Klemm, J. M. Williams, K. R. Lykke, J. A. Schlueter, H. H. Wang, P. Wurz, J. R. Ferraro, G. A. Yaconi, P.

Stout,Physica C,204, 399-405(1993).b) T. Komatsu, N. Matsukawa, T. Nakamura, H. Yamochi, G. Saito, Phosphorus, Sulfur Silicon Relat. Elem. 67, 295-300(1992), and Section 3-5-4-2 in G. Saito, Y.Yoshida,Bull. Chem. Soc. Jpn., 80, 1(2007).

19. K. Kanoda,J. Phys. Soc. Jpn.,75, 051007/1-16(2006).20. N. Toyota,Physical Phenomena at high magnetic fields-II(World Sci.) pp282293, 1996.


12/40

12

12

A-3) Peculiarities in a Mott Insulator -(ET)2Cu2(CN)3

We have been working to understand the peculiarities of the structural and physical properties of a Mott

insulator-(ET)2Cu2(CN)37which has a larger anion space than those of four 10 K class superconductors 2

5in Table 1,however it showed superconductivity at 23 K under hydrostatic pressure of 0.35 GPa.[1,2]

A-3-1Control ofU/Wand Band Filling : '-(ET)2Cu2(CN)3

The crystal structure of a Mott insulator -(ET)2Cu2(CN)3 7 in Fig. 6 revealed the disorder in theposition of C and N atoms in part of the CN groups (Fig. 7a).[1a,b]That kind of disorder was thought to be thereason for the low Tcof this salt. Under pressure it showed oscillations relevant to the Fermi surface (SdHand AMRO).[1c-e]-(ET)2Cu2(CN)3was converted to a metal and superconductor (on-set Tcof 23 K) byapplying hydrostatic pressure of 0.35 GPa through a Mott insulator-metal transition at 1314 K with aresistivity drop by 105.[1f]While, an isostructural salt'-(ET)2Cu2(CN)3, which was at early stage thought tocontain a small amount of Cu2+, exhibited a metal-superconducting transition without Mott insulatingstate.[1f,g] Later, it was found that exact chemical formula of '-(ET)2Cu(CN)3 was-(ET)2(Cu

1+2-x-yCu

2+x){(CN)3-2y[N(CN)2]y} and its transport natures were governed by the amount of Cu

2+(x)and ligand [NC-N-CN](y).[1h]Owing to the very similar geometrical shape, size and equal charge between[Cu(CN)2: NC-Cu-CN and/or CN-Cu-NC, dihedral angle 120.1] and [N(CN)2: NC-N-CN, dihedralangle 116.7], they were nearly freely replaceable to each other in the anion layer resulting in the comparablelattice parameters among -(ET)2Cu(CN)3, -(ET)2Cu(CN)[N(CN)]2and their alloy, '-salt (Fig. 7a-c). Atx=0 andy= 0, the salt is a Mott insulator [-(ET)2Cu(CN)3] (point ain Fig. 7d), while the other extreme side (x= 0,y = 1) is [-(ET)2Cu(CN)[N(CN)2]] with Tc= 11.2 K at AP (point e). By changing bothx(80 1200

ppm) andy(preferential values ofyare 0.05 (point b), 0.3 0.4 (c), 0.8 (d)), the Tcwas tuned from 3 to 11K.Aty= 0.3 0.4, the Tc covered from 3 to 10 K and the crystals with different Tc had differentxvaluesindicating that the charge of ET was modified from +0.5 to +0.5(1x), that corresponds to the change ofchemical potential, i.e. band-filling. Tcincreased with increasingx(= the content of Cu

2+) up to 400 ppm and

then Tcdecreased.These experimental facts indicate that this system can be an excellent model of band filling control andwill be a good candidate for making superlattice composed of Mott insulator/superconductor hetero-junctions.It should be emphasized that their lattice parameters are nearly kept constant through such an anionmodification, which is the most essential feature for achieving the successful tuning of Tc in an organicsuperconductor.

A-3-2 New Spin State Originated from Strong Spin Frustrations : Spin-Liquid State

It has been generally observed that the superconducting state is located in close proximity to themagnetic-ordered state in oxide, C60, TMTSF and ET systems (SDW or AF). However, the ground state of 7

differs from them distinctly.Figure 8ashows the donor packing of 7where an ET dimer is a unit with S = 1/2

Figure 7 The anion structures of (a) -(ET)2Cu2(CN)3 and (c) -(ET)2Cu(CN)[N(CN)2]. (b) A schematic figure of-(ET)2(Cu

+2-x-yCu

2+x){(CN)3-2y[N(CN)2]y} withy~ 0.1. (d) Relation between the content of N(CN)2, y and Tcin several crystals

of' salt: -(ET)2(Cu1+)2-x-y(Cu

2+)x(CN)3-2y[N(CN)2]y. As for points a-e, see text. Dashed line indicates the samples of y 0.3.[1]

(d)


13/40

13

13

spin to form the triangular lattice with two kinds of transfer integrals t= (|tp| + |tq|)/2 and t= tb2/2 (Fig. 8b).[3,4]

Figures 8c and8dcompare the line shape of 1H-NMR absorption of 7and 5, respectively. Salt 5exhibited adrastic change below 27 K owing to the formation of three-dimensional AF ordering. On the other hand, theabsorption band of 7remains almost invariant down to 32 mK indicating non-spin-ordered state: spin-liquidstate.[5]The appearance of spin-liquid state in 7can be explained in terms of the spin geometry of -(ET)2Xwhich forms an anisotropic triangular lattice with varied t'/t (Table 1). Although the Mott insulators 5anddeuterated 4have nearly the same W/Ueffas that of 7, the less frustrated spins in 5(t/t~ 0.75) and D-salt of 4(t'/t=0.68 for the H-salt) can lead to an AF ordered state. On the other hand, since the spin frustration is muchsignificant in 7because of the equilateral triangle spin geometry (t'/t= 1.06), the formation of the AF andsuperconducting states was suppressed at ambient pressure and the unprecedented spin-liquid state appeared.

Strange behavior of 7:1) gap or no gap: Controversial discussions have been presented concerning with the magnitude of the gap ofthe spin-liquid state. Specific heat measurements suggested gapless nature,[7]while thermal conductivitymeasurements suggested small gap.[8]2) abnormality near 56 K: 13C-NMR[5j]and thermal expansion[9]measurements indicated abnormality andsuggests that the lattice is not frozen even at 56 K.

A-3-3 Emergence of Superconducting State Next to Spin Liquid State

Figure9 shows the T-Pphase diagram of -(ET)2Cu2(CN)3 at low temperature regionby applying

uni-axial strain along b- (right figure: t/tincreases in this direction)andc-(left figure: t/tdecreases in thisdirection) axes with epoxy-method.[10]In both cases, a superconducting state readily appeared nearly above0.1 GPa since the t'/tdeviates from unity; i.e. strong spin frustrations were released. It is very noteworthy thatthe disappearance of the semiconductor-like region, appearance of the superconducting state and its Tc's arefairly anisotropic as shown in Fig.10. Compared to the hydrostatic results, which extinguish superconducting

phase above 0.3 GPa, the uni-axial method afforded 1) a much higher Tcvalue, 2) an increase of Tcat theinitial pressure region, 3) an anisotropic pressure dependence and 4) superconducting phase remains at higher

pressure. Within thebc-plane, the superconducting state appeared above 0.1 GPa (Tc= 3.8 K (//b), 5.8 K (//c))and Tcincreased up to 6.8 K (//b, 0.5 GPa) and 7.2 K (//c, 0.3 GPa). Along the a*-axis, the superconductingstate appeared above 0.3 GPa.

Figure 8. (a) Donor packing pattern of-(ET)2Cu2(CN)3along the a-axis (transfer integrals; tb1= 22 meV, tb2= 12 meV, tp= 8 meV and tq= 3 meV) and (b) triangular spin lattice (t'/t= 1.06;t' = tb2, t = (tp+tq)/2) composed of the ET dimer.White and black circles represent an ET molecule and an ET dimer (spin site), respectively. Line shape of 1H-NMR of (c)

-(ET)2Cu2(CN)3[5a]

and (d) -(ET)2Cu[N(CN)2]Cl.

[6]

(b)


14/40

14

14

The appearance of the superconducting state is interpreted by both the increase of W/Ueff and thedeviation of t'/tfrom unity. The increase of Tcin the initial pressure regime is ascribed to the reduction of thespin frustration. The following decrease of Tcin whole measured directions is explained by the decrease of

D(F) owing to the increase of W. The appearance of superconducting state immediately after the release of

the spin frustration in the spin-liquid state is an indication of the importance of the magnetic mediation forsuperconductivity.

The uni-axial strain experiments including other -type superconductors clearly revealed that the Tcincreased as the U/Wapproaches unity and as the t/tdeparts from unity (Fig. 11).[11]Strange behavior: Hydrostaticvs.Uni-axial stressWhy the hydrostatic pressure results of Tcin Fig. 10donot agree with any of those along the principal axes or averaged one ?

Figure 9. Temperature-uniaxial pressure phase diagram in the low temperature region.[10]

Figure 10. Pressure dependence of on-set Tc of

-(ET)2Cu2(CN)3 by the uni-axial strain (colored) and

hydrostatic pressure methods(triangle). TIM: Mott

insulator-metal transition temperature.[10c]

0.60.7

0.80.9

1.01.1

0

2

4

6

8

10

12

0.8

0.9

1.0

Cu[N(CN)2]Br

Cu2(CN)

3

Cu(NCS)2

(d)

(c)

(b)

(a)

TC/K

U/W

t'/t

t/tU/W

Figure 11. Tcof-(ET)2X salts are plotted as function of t'/tand U/W.[11] X = I3(a), Ag(CN)2H2O (b), Cu(CN)[N(CN)2] (c), and

Cu[N(CN)2]Cl (d). Blue and yellow arrows indicate the direction of t'/tdecreases and U/Wincreases, respectively. Red arrows

correspond to the change of Tcby applying uni-axial strain.


15/40

15

15

Figure 12shows a plot of U/tagainst t/tof calculated based on the crystal structure of several -(ET)2Xby extended Hckel (EH) or DFT method.[10c]A linear relation between U/tand t/tis seen in both calculatedby EH and DFT methods. Figure 12also demonstrates that the -(ET)2X salts with t/tin the range of 0.86(X=Cu(NCS)2) and 1.06 (X=Cu2(CN)3) have not been explored yet (between two red lines in Fig. 12).

It has long been predicted that the geometrical spin frustration of antiferromagnetscaused by the spin correlation in particular spin geometry (triangle, tetrahedral, Kagome(Scheme 4), etc.) prevents the permanent ordering of spins. So the spins of Ising systemwith AF interaction in the equilateral triangle spin lattice will not show any long-rangeorder even at 0 K, and hence the phase, namely quantum spin liquid phase, has highdegeneracy.[12]Such spin liquid state has only been predicted theoretically,[13] and avariety of ideal materials have been designed and examined for long.[14] Since thediscovery of the spin liquid state in -(ET)2Cu2(CN)3, several materials have beenreported to have such exotic spin state: EtMe3Sb[Pd(dmit)2]2,

[15a] ZnCu3(OH)6Cl2,[15b,c] Na4Ir3O8,

[15d] andBaCu3V2O8(OH)2.

[15e,f]Some inorganic materials reported as spin liquid candidates were eliminated owing tothe spin ordering at extremely low temperatures, etc.[16] Na4Ir3O8 and two organic compounds(-(ET)2Cu2(CN)3,EtMe3Sb[Pd(dmit)2]2)may be recognized as softMott insulators and have metallic stateunder pressure. Only -(ET)2Cu2(CN)3 has the superconducting phase next to spin-liquid state, though allhighly correlated superconductors so far known; TMTSF, ET, C60

[17] families and also electron-correlated

cuprate and iron pnictide high Tcsystems,[18]

indicate that a magnetic ordered state (SDW, AF) is allocatednext to the superconducting state.

References A-31. a) T. Komatsu, T. Nakamura, N. Matsukawa, H. Yamochi, G. Saito, H. Ito, T. Ishiguro, M. Kusunoki, K. Sakaguchi, Solid State Commun., 80, 843-847 (1991).

b) H. Yamochi, T. Nakamura, T. Komatsu, N. Matsukawa, T. Inoue, G. Saito, T. Mori, M. Kusunoki, K. Sakaguchi,Solid State Commun.,82, 101-105 (1992)c) E. Ohmichi, H. Ito, T. Ishiguro, T. Komatsu, G. Saito,J. Phys. Soc. Jpn.,66, 310-313 (1997).d) E. Ohmichi, H. Ito, T. Ishiguro, T. Komatsu, G. Saito,Rev. High Pressure Sci. Tech., 7, 523-525 (1998).e) E. Ohmichi, H. Ito, T. Ishiguro, G. Saito,Phys. Rev. B, 57, 7481-7484 (1988).f) T. Komatsu, G. Saito,Mol. Cryst. Liq. Cryst.,285, 51-56 (1996)g) T. Komatsu, N. Matsukawa, T. Inoue, G. Saito,J. Phys. Soc. Jpn.,65, 1340-1354(1996).h) T. Komatsu, N. Kojima, G. Saito, Synth. Met., 85, 1519-1520 (1997)i) O. Drozdova, G. Saito, H. Yamochi, K. Okubo, K. Yakushi, M. Uruichi, L. Ouahab,Inorg. Chem.,40, 3265-3266(2001).

2.-(ET)2Cu2(CN)3 was first prepared and reported by a few groups including ref. 1a.

a) U. Geiser, H. H. Wang, K. D. Carlson, J. M. Williams, H. A. Charlier Jr., J. E. Heindl, G. A. Yaconi, B. H. Love, M. W. Lathrop, J. E. Schirber, D. L. Overmyer, J. Ren,

Kagome lattice

Scheme 4

Figure 12.U/tplotted against t0/tfor -(ET)2X. Closed circles: results of the extended Hckel calculation using the crystal structure

at 290K in the series of -(ET)2X for X = A: Cu2(CN)3, B: Cu[N(CN)2]Cl, C: Cu[N(CN)2]Br, D: Cu(CN)[N(CN)2], E: Cu(NCS)2,

F: Ag(CN)2H2O, G: I3,H: -(ET)4Hg2:89Br8,and I: -(ET)4Hg2:89Cl8.Open circles: those obtained for the crystal structure at 120

K.Averages are used for the two inequivalent t and t0in D and E without the inversion symmetry. H and I are metals due to theband filling deviating by 10% from the half-filling. Open triangles: DFT calculations plotted in the top and right axes.[10c]


16/40

16

16

M. -H. Whangbo,Inorg. Chem., 30, 2586-2588(1991).b) X. Bu, A. Frost-Jensen, R. Allendoerfer, P. Coppens, B. Lederle, M. Naughton, Solid State Commun.,79, 1053-1057(1991).c) G. C. D. Papavassiliou, J. Lagouvardos, V. C. Kakoussis, A. Terzis, A. Hountas, B. Hilti, C. Mayer, J. S. Zambounis, J. Pfeiffer, M. -H. Whangbo, J. Ren, and D. B.

Kang,Mater. Res. Soc. Symp. Proc.,247, 535-540(1992).3. R. H. McKenzie,Comments Condens. Matter Phys., 18, 309-337(1998).

4. H. Kino, H. Fukuyama,J. Phys. Soc. Jpn., 64, 2726-2729(1995).5. a) Y. Shimizu, K. Miyagawa, K. Kanoda, M. Maesato, G. Saito,Phys. Rev. Lett., 91, 107001/1-4 (2003)

b) Y. Shimizu, M. Maesato, G. Saito, K. Miyagawa, K. Kanoda, Synth. Met., 137, 1247-1248 (2003)c) Y. Shimizu, M. Maesato, G. Saito, O. Drozdova, L. Ouahab, Synth. Met., 133-134, 225-226 (2003)d) Y. Shimizu, K. Miyagawa, K. Oda, K. Kanoda, M. Maesato, G. Saito,J. Phys. IV France, 114, 377-378 (2004)e) Y. Shimizu, Y. Kurosaki, K. Miyagawa, K. Kanoda, M. Maesato, G. Saito, Synth. Met., 152, 393-396 (2005)f) Y. Shimizu, K. Miyagawa, K. Kanoda, M. Maesato, G. Saito,Prog. Theor. Phys. Suppl., 159, 52-60 (2005)g) Y. Kurosaki, Y. Shimizu, K. Miyagawa, K. Kanoda, G. Saito,Phys. Rev. Lett., 95, 177001/1-4 (2005)h) I. Kzsmrki, Y. Shimizu, G. Mihly, Y. Tokura, K. Kanoda, G. Saito,Phys. Rev. B, 74, 201101/1-4 (2006)i) S. Ohira, Y. Shimizu, K. Kanoda, G. Saito, J. Low Temp. Phys.,142, 153-158 (2006)j) Y. Shimizu, K. Miyagawa, K. Kanoda, M. Maesato, G. Saito,Phys. Rev. B,73, 140407/1-4 (2006)

6. K. Miyagawa, A. Kawamoto, Y. Nakazawa, and K. Kanoda,Phys. Rev. Lett., 75, 1174-1177(1995).7. S. Yamashita, Y. Nakazawa, M. Oguni, Y. Oshima, H. Nojiri, Y. Shimizu, K. Miyagawa, K. Kanoda, Nature Phys.4, 459-462(2008)8. M. Yamashita, N. Nakata, Y. Kasahara, T. Sasaki, N. Yoneyama, N. Kobayashi, S. Fujimoto, T. Shibauchi, Y. Matsuda, Nature Phys. 5, 44-47(2009)

9. R. S. Manna, M. de Souza, A. Bruhl, J. A. Schulueter, M. Lang, Phys. Rev. Lett., 104, 016403/1-4(2010)10. a) Y. Shimizu, M. Maesato, G. Saito, O. Drozdova, L. Ouahab, Synth. Met., 133-134, 225-226 (2003)

b) M. Maesato, Y. Shimizu, T. Ishikawa, G. Saito,Synth. Met., 137, 1243-1244 (2003)c) Y. Shimizu, M. Maesato, G. Saito,J. Phys. Soc. Jpn.,80, 074702/1-7(2011).

11.M. Maesato, Y. Shimizu, T. Ishikawa, G. Saito, K. Miyagawa, K. Kanoda,J. Phys. IV France, 114, 227-231 (2004).12. G.H.Wannier,Phys. Rev. 79, 357-364(1950)13. P. W. Anderson,Mater. Res. Bull.8, 153-164(1973).14. a) L. Balents,Nature, 464, 199-208(2010).

b) M. R. Norman, Science, 332, 196-200(2011)c) Y. Furukawa, Y. Sumida, K. Kumagai, F. Borsa, H. Nojiri, Y. Shimizu, H. Amitsuka, K. tenya, P. Kogerler, L. Cronin,J. Phys.: Conf. ser.320012047d) G. Seeber, P. Kogerler, B. M. Kariuki, L. Cronin, Chem. Commun., 1580-1581(2004)

15. a) T. Itou, A. Oyamada, S. Maegawa, M.Tamura, R. Kato,J. Phys. Condens. Matter19, 145247/1-5(2007).b) M. P. Shores, E. A. Nytko, B. M. Bartlett, D. G. Nocera,J. Am. Chem. Soc., 127, 13462-13463 (2005).c) P. Mendels, F. Bert,J. Phys. Soc. Jpn., 79, 011001/1-10 (2010)

d) Y. Okamoto, M. Nohara, H. Aruga-Katori, H. Takagi,Phys. Rev. Lett.,99, 137207/1-4 (2007)e) Y. Okamoto, H. Yoshida, Z. Hiroi,J. Phys. Soc. Jpn.,78, 033701/1-4 (2009).f) R. H. Colman, F. Bert, D. Boldrin, A. D. Hiller, P. Manuel, P. Mendels, A. S. Wills,Phys. Rev. B.,83, 180416/1-4 (2011).

16. a) Cs2CuCl4: R. Coldea, D.A. Tennante, Z. Tylczynski,Phys. Rev.B,68, 134424/1-16 (2003).b) NiGa2S4: S. Nakatsuji, Y. Nambu, H. Tonomura, O. Sakai, C. Broholm, H. Tsunetsugu, Y. Qiu, Y. Maeno, Science, 309, 1697-1700 (2005).c) NaCrO2: A. Olariu, P. Mendels, F. Bert, B. G. Ueland, P. Schiffer, R. F. Berger, R. J. Cava,Phys. Rev. Lett.,97, 167203/1-4 (2006).d) FeSc2S4: A. Kimmel, M. Mucksch, V. Tsurkan, M. M. Koza, H. Mutka, A. Loidl,Phys. Rev. Lett., 94, 237402/1-4 (2005).e) Cu3V2O7(OH)22H2O: H. Yoshida, Y. Okamoto, T. Tayama, T. Sakakibara, M. Tikunaga, A. Matsuo, Y. Narumi, K. Kindo, M. Yoshida, M. Takigawa, Z. Hiroi,J.

Phys. Soc. Jpn.,78, 043704/1-4 (2009).17. Y. Iwasa, T. Takenobu,J. Phys.: Condens. Matter, 15, R495-R519(2003).18. C. W. Chu,Nat. Phys., 5, 787-789(2009).


17/40

17

17

A-4) Various Conductors and Semiconductors of CT Solids

A-4-1 Control of Structure, Dimensionality, Band-Filling, Phase Transition of CT

Solids by Hydrogen-Bond and Proton-TransferThe related elements, proton (H+), hydrogen (H), and hydride (H) change their physical propertiesincluding size drastically by the change of the number of electrons. Hydrogen-bond and proton-transferinteractions are the key to understand many chemical reactions, biological activities, structure of molecularassemblies and supramolecules, functionalities in solid state, etc. We have develop systems where

proton-transfer and electron-transfer interactions are competing and/or cooperating using the combinations ofnitrophenol derivatives (picric acid, etc) and aromatic amines (mono- and diamine) (see SectionB-1),[A-1-4k,4l,A-1-5]quinhidrone which is possible to provide degenerate condensed solid composed of neutralradical (QH) by the reaction between quinone (Q) and hydroquinone (H2Q),

[1]band-filling control by H+in3,3,5,5-tetranitro-4,4-biphenyldiol (H2TNBP),

[2] 2,2'-bi-1H-imidazole (H2BIM),[A-1-4m] and

dihydrotetracyanodiphenoquinodimethane (4,4-bis(dicyanomethyl)biphenyl) (H2TCNDQ),[A-1-4n] and

formation of conductors through the Htransfer between NADH derivatives and TCNQ.[3]

A-4-2 Two-dimensional BO Stable Metals in Various ShapesAs shown in Figure 4the peripheral addition of alkylchalcogeno groups to TTF skeleton increases the

self-assembling ability of molecule and hence the electronic dimensionality of molecular assemblies increases.

With increasing the self-assembling ability of the component molecules (mainly donor molecules) theE(DA) range for conductors extends into MSregion in Fig. 3band most of the DA complexes in thisregion with a low hCTband (


18/40

18

18

complexes (reticulate doped polymer films) some of which are transparent,[5] compressed pellets withferrimagnetic behavior,[6]films sensitive to the moisture in air,[7]etc., regardless of the sort, shape, and size ofacceptor or anion molecules. As a result, the BO complexes hardly exhibit any phase transition including thesuperconducting one (only two superconducting salts with Tc1.5 K).

A-4-3 Cytosine Complexes with TCNQ Derivatives

A variety of transport studies have been performed on the biological materials, such as hemoglobin,amino acids, proteins, polypeptides and so on by Eley et al.[8]and others.Most of them are insulating (< 1017S cm1at RT)[8]except cytochrome-c3.

[9]In the recent studies on biomolecular conductors, DNA is one of themost active target molecules, and numerous experiments concerning the transport properties of DNAmolecule have been examined.[10]The origin of the electrical conduction of DNA wire is regarded as the holetransport within a one-dimensional -stacks of nucleobases. Plenty of controversial reports have forwardedconcerning with the metallic or highly conductive properties of pure DNA molecules. Some reported highconductivity[10b,c,e]with RTat most 10

4S cm1[10b]or even superconducting properties,[10f]and others claimed

that the carefully deionized DNA molecules are insulating[10d,g]

in agreement with the old reports[10h,i]

with RTless than 106 S cm1, suggesting thesurroundings of DNA are critical for itstransport property. It seems that the DNAmolecules turned out to be insulating.Therefore, investigation of CT complexesof nucleobases may give key clue tounderstand the electrical conduction ofDNA.

Several attempts have been undertaken to investigate the CT complexes in a variety of biochemicalsystems, especially using nucleobases.[11]Estimation ofIPof the nucleobases, as potential components in CT

complexes, indicate that they are reasonably effective -donors particularly in the case of guanine; IP =7.647.85 eV vs. adenine (7.808.26 eV), cytosine (8.458.74 eV) and thymine (8.748.87 eV).[12]

We have started the CT complex formation of cytosine (C) because of its moderate solubility in organicsolvent.[13]Chas a weak electron donating ability (Ep

ox= 1.90 V vs. SCE in water) as well as mediumproton donating/accepting abilities (pKa = 4.55 and 12.2). Reaction between C in methanol and severalTCNQ derivatives (RTCNQ) in acetonitrile yielded three kinds of ionic solids; (I)insulators composed of methoxy substituted RTCNQ anions such as(CHC+)[F4TCNQ-OMe

(right figure)](H2O),[13a](II) semiconducting CT solids with

fully ionic RTCNQ radical anions such as (CHC+)(TCNQ),[13c] and (III)conducting CT solids of partially ionic or mixed valent RTCNQ radical anions such

as (CHC+

)(MeTCNQ

0.5

)2.

[13c]

Cation units in all products were found to beprotonated cytosine species. Crystal structures were determined formethoxy-substituted anion salts (R = F4and H) in Group Iand RTCNQ radical anionsalts (R = H and Et2) in Group IIwith hemiprotonated cytosine pairs, which wereformed by triple self-complementary hydrogen bonds(right figure).

Crystal structural analysis of (CHC+)(TCNQ) revealed the segregated structurewith uniform stacking pattern (Figure 14a). The interplanar separations of TCNQ andCHC

+ columns were 3.14 and 3.32 , respectively. The CHC+ pairs formed aone-dimensional ribbon structure (Figure 14b). The hydrogen bonds between CHC+ribbon and TCNQ molecules constructed the layered structure. In addition, the self-assembling ability of Cstrengthened the uniform arrangement of the crystal resulting in both the high conductivity (RT= 3.2 102

S cm1

on single crystals), which is one of the best among the conventional Mott type TCNQ salts, and the

N

N NH

N

NH2

Adenine

HN

N NH

N

Guanine

O

H2N

N

NH

NH2

O

Cytosine

HN

NH

O

O

Thymine

Me

Scheme 6

Hemiprotonated

Cytosine pair CHC+

F4TCNQ-OMe

Scheme 7


19/40

19

19

absence of spin-Peierls type structural distortion down to 10 K. (CHC+)TCNQ was examined under highpressures up to about 7 GPa using a diamond anvil cell. The activation energy aof 0.14 eV at AP decreasedmonotonically by a rate of 0.013 eV/GPa.[13f]The partially ionic salt of MeTCNQ in Group IIIexhibited thehighest conductivity of 2 S cm1so far observed for CT complexes based on biological molecules. Our studyrevealed that the protonated states of Cespecially CHC+species are extraordinary stable and furthermore thecharacteristic pattern of the complementary hydrogen bonds between cytosine molecules contribute toconstruct effective molecular packing and to control the electronic structure of TCNQ molecules forelectronic conductors.

A-4-4 Two-dimensional Metal Based on C60

We have developed a multicomponent approachfor synthesizing ionic fullerene compoundsDI

+DII(fullerene)of various structures including -and -type dimer of fullerenes, type complex,etc.(Fig. 15), where DI

+ is a small, strong donor orcation that ionizes fullerene and determines its chargedstate, whereas DII is a large, neutral molecule thatdefines the crystal packing of the complex.[14]We have

been working to develop two-dimensional metalsbased on C60

. In order to exhibit metallic properties,the C60sublattice should have a close-packed structure.However, C60

radical anions have a strong tendencyfor dimerization, and when they are allowed to

approach each other they normally form diamagnetic single-bonded (C60

)2dimers. Furthermore when the

a

b

Figure 15. (a) -type C60dimer, (b) -type C70dimer,(c) -type C60 dimer, (d)

1-coordination of CoIITPPwith C60(CN)2

c

d

(c)

Figure 14.Crystal structure of (CHC+)(TCNQ) salt. (a) Uniform segregated stacks of CHC+and TCNQ. (b) CHC+ribbons bycomplementary hydrogen bonds and layered structure of this salt. Dotted lines show hydrogen bonds. (c) Three-dimensional structurealong the a-axis.


20/40

20

20

transfer interactions are small in the honeycomb structure of C60, strong spin frustration will be created

based on the trilateral triangle spin geometry of C60 just like -(ET)2Cu2(CN)3 in Section A-3-2. By

choosing DIImolecules with suitable spatial geometry and size, we are able to synthesize a complex with aclose-packed fullerene two-dimensional sublattice in which the C60

monomers preferentially form atwo-dimensional honeycomb network of C60rather than undergo dimerization. The TPC molecule (Fig. 16)afforded a suitable geometrical space and spatial regulation as the DIIcomponent for C60

ions. The TPCmolecules form hexagonal layers with voids that accommodate foreign cations, such as MDABCO+(DI

+)[MDABCO:N-methyldiazabicyclooctane], in the first keykeyhole relationship (Fig. 16a). DockingC60

into the periodic hollow sites in the (MDABCO+)TPC network (second keykeyhole relationship; Fig.16b,c) leads to hexagonal packing of the fullerene layers without dimerization of the C60

monomers,resulting in the first fullerene-based two-dimensional organic metal (MDABCO+)TPC(C60).[15]It exhibits ametallic state down to 1.9 K, which can be explained by the two-dimensional character of the electronicstructure in accordance with the calculated Fermi surfaces (Fig. 16 d, e). Since the hexagonal packing of C60

formed nearly equilateral triangle spin geometry (Fig. 16f), a two-dimensional layer of C60

exhibits either

strong spin frustration or itinerancy depending on the C60C60 distance and rotational disorder of C60.Actually there are two kinds of two-dimensional layers of C60, layerAandB, C60molecules in the latterlayer is rotationally disordered above 200 K and the layerBis not metallic and exhibits spin frustration. Theordering of C60

in the type B layers with MDABCO+ at around 200 K triggers a transition from anonmetallic and antiferromagnetically frustrated state to a metallic state in layerB, whilst the ordered C60

inlayerAkept its two-dimensional itinerancy over the entire temperature range. This compound is a fascinatingexample of a material composed of only light elements (C, H, N) that exhibits a metallic state down to 1.9 K.

Figure 16. Molecular structures of TPC (DII) and MDABCO+ (DI

+). Crystalstructure packing in (MDABCO+)TPC(60

): (a) TPC molecules form ahexagonal hollow and the MDABCO+ cation fits into the hollow; the 60

molecules are docked into the hollow in TPC layer in a keykeyhole relationship(top view (b) and side view (c)) to form DI

+DIIC60(see figur below). Colours: C,

dark yellow: H, pale blue; and N, dark blue. Calculated Fermi surface at 160 K in

(d) layerAand (e) layerB. (f) Projection of the (DABCO

+

)TPC layer on the C60layerA.

(a)

(b) (c)

X

Y

V

V

C

X

Y

V

V

C

(d) (e) (f)


21/40

21

21

A-4-5 Tuning of Fermi level of FET ElectrodesFine control of p-, n-, and ambipolar-type field-effect transistor (FET) operations is successfully

demonstrated in prototypical single-crystal organic FETs with use of chemically tunable nature of Fermienergy (see Section A-1-2) in TTFTCNQ-based organic metal electrodes. Carrier-type preference andrectifying nature in the organic-organic contacts are revealed in terms of the FET operations as well as of theall-organic Schottky diode characteristics.[A-1-4a] For the electrodes whose chemical potentials are allocatedwithin the conduction band of the channel material (DBTTFTCNQ), FET exhibited n-type behavior (Fig. 16,A, B). While the chemical potentials of organic metals are allocated within the valence band of the channel,

p-type behaviors were observed (Fig. 17, C, D). When the chemical potentials of the electrodes are within thegap of the channel, FET exhibited ambipolar-type behavior (Fig. 17, E, F).

References A-41. T. Mitani, G. Saito, and H. Urayama,Phys. Rev. Lett.,60, 2299-2302(1988).2. K. Nishimura, T. Kondo, O. O. Drozdova, H. Yamochi, and G. Saito,J. Mater. Chem., 10, 911-919(2000).3. G. Saito and A. K. Colter,Tetrahedron Lett.,18, 3325-3328(1977).4. a) T. Nakamura, G. Yunome, R. Azumi, M. Tanaka, H. Tachibana, M. Matsumoto, S. Horiuchi, H. Yamochi, and G. Saito,J. Phys. Chem., 98, 1882-1887(1994).

b) K. Ogasawara, K. Ishiguro, S. Horiuchi, H. Yamochi, and G. Saito,Jpn. J. Appl. Phys., 35, L571-L573(1996).c) M. Izumi, V.M. Yartsev, H. Ohnuki, L. Vignau, and P. Delhaes,Recent Res. Dev. Phys. Chem., 5, 37-75(2001)..

5.a) J. K. Jeszka, A. Tracz, A. Sroczynska, M. Kryszewski, H. Yamochi, S. Horiuchi, G. Saito, and J. Ulanski, Synth. Metals, 106, 75-83(1999).b) S. Horiuchi, H. Yamochi, G. Saito, J. K. Jeszka, A. Tracz, A. Sroczynska, and J. Ulanski,Mol. Cryst. Liq. Cryst.,296, 365-382(1997).

Figure 17. OFETs composed of CT-complex-based organic metal electrodes. (a) An illustration of the device. (b) Interface banddiagram of metal/semiconductor contact in DBTTFTCNQ single crystal OFET with a variety of organic metal electrodes. (c)Modified figure of Fig. 3a. The conductive complexes AF are drawn as functions of both ionization potentials of the donors IP

D,

upper-left axis and electron affinities of the acceptorsEAA, upper-right axis. (d) Transfer characteristics at VD=5 V of DBTTFTCNQ

single-crystal field effect transistors with the source and drain electrodes, composed of A: TTF TCNQ, B: TTFFTCNQ, C:TTFF2TCNQ, D: TSFFTCNQ, E: TSFF2TCNQ, and F: DBTTFF4TCNQ, measured along the crystal long axes.

(d)

(a)

AB

CD

E

F

Channel

Electrode

(b)(c)


22/40

22

22

6. H. Yamochi, T. Kawasaki, Y. Nagata, M. Maesato, and G. Saito,Mol. Cryst. Liq. Cryst., 376, 113-120(2002)7. A. Tracz,J. Appl. Polym. Sci.,86, 1465-1472(2002)8. D. D. Eley and D. I. Spivey,Discuss. Faraday Soc., 56, 1432-1442(1960).9. a) Y. Nakahara, K. Kimura, H. Inokuchi, and T. Yagi, Chem. Lett., 877-880(1979).

b) K. Kimura, K., Y. Nakahara, T. Yagi, and H. Inokuchi,J. Chem. Phys.,70, 3317-3323(1979).

c) K. Kimura, S. Nakajima, K. Niki, and H. Inokuchi,Bull. Chem. Soc. Jpn., 58, 1010-1012(1985).10. For a review (a) B. Giese,Acc. Chem. Res., 33, 631-636(2000) and references therein. For individual worksb) H.W. Fink, C. Schnenberger,Nature, 398, 407-410(1999).(c) D. Porath, A.Bezryadin, S. Vries, C. Dekker,Nature, 403, 635-638(2000).d) P. J. de Pablo, F. Moreno-Herrero, J. Colchero, J. Gmez-Herrero, P. Herrero, A. M. Bar, P. Ordejn, J. M. Soler, E. Artacho,Phys. Rev. Lett.,85,

4992-4995(2000).e) P. Tran, B. Alavi, G. Grner,Phys. Rev. Lett. 85, 1564(2000).f) A. Y. Kasumov, M. Kociak, S. Guron, B. Reulet, V. T. Volkov, D. V. Klinov, H. Bouchiat, Science,291, 280(2001).g) Y. Zhang, R. H. Austin, J. Kraeft, E. C. Cox, N. P. Ong,Phys. Rev. Lett.,89, 198102/1-4(2002).h) J. Ladik,Acta Phys. Acad. Sci. Hung., 1960, 11, 239.i) M. E. Burnel, D. D. Eley, V. Subramanyan,Ann. N. Y. Acad., Sci.,158, 191(1961).

11. R. Foster, Organic Charge-Transfer Complexes,Academic Press, New York (1969). Chapter 1212. a) Y. M. Orlov, A. N. Smirnov, and Y. M. Varshavsky,Tetrahedron Lett.,18, 4377-4378(1976).

b) D. Dougherty, E. S. Younathan, R. Voll, S. Abdulnur, and S. P. McGlynn,J. Electron Spectrosc. Relat. Phenom., 13, 379-393(1978).c) S. G. Lias, J. E. Bartmess, J. F. Liebman, L. J. Holms, R. D. Levin, and W. G. Mallard,J. Phys. Chem.Ref. Data, 17, 861(1988).d) S. D. Wetmore, R. J. Boyd, and L. A. Eriksson, Chem. Phys. Lett.,322, 129-135(2000) and references therein.

13. a) T. Murata, G. Saito, Chem. Lett., 35, 1342-1343 (2006)b) T. Murata, K. Nishimura, G. Saito,Mol. Cryst. Liq. Cryst., 466, 101-112 (2007).c) T. Murata, G. Saito, K. Nishimura, Y. Enomoto, G. Honda, Y. Shimizu, S. Matsui, M. Sakata, O. O. Drozdova, K. Yakushi, Bull. Chem. Soc. Jpn.,81, 331-344 (2008).d) T. Murata, Y. Enomoto, G. Saito, Solid State Sciences, 10, 1364-1368 (2008)e) T. Murata, K. Nakamura, H. Yamochi, G. Saito, Synthetic Metals, 159(21-22), 2375-2377(2009)f) M. Sakata, M. Maesato, T. Miyazaki, K. Nishimura, T. Murata, H. Yamochi, G. Saito,J. Phys.: Conf. Ser., 132, 012011/1-4 (2008)

14. a) D. V. Konarev, S. S. Khasanov, A. Otsuka, Y. Yoshida, G. Saito,J. Am. Chem. Soc.,124, 7648-7649 (2002)b) D. V. Konarev, S. S. Khasanov, A. Otsuka, G. Saito,J. Am. Chem. Soc.,124, 8520-8521 (2002)c) D. V. Konarev, S. S. Khasanov, G. Saito, A. Otsuka, Y. Yoshida, R. N. Lyubovskaya,J. Am. Chem. Soc., 125, 10074-10083 (2003)d) D. V. Konarev, S. S. Khasanov, G. Saito, R. N. Lyubovskaya,Recent Res. Devel. Chem.,2, 105-140 (2004)e) D. V. Konarev, A. Y. Kovalevsky, A. Otsuka, G. Saito, R. N. Lyubovskaya,Inorg. Chem.,44, 9547-9553 (2005)f) D. V. Konarev, S. S. Khasanov, A. Otsuka, G. Saito, R. N. Lyubovskaya,J. Am. Chem. Soc., 128, 9292-9293 (2006)g) D. V. Konarev, S. S. Khasanov, G. Saito, A. Otsuka, R. N. Lyubovskaya,Inorg. Chem., 46, 7601-7609 (2007)

15. D. V. Konarev, S. S. Khasanov, A. Otsuka, M. Maesato, G. Saito, R.N. Lyubovskaya,Angew. Chem. Inter. Ed, 49(28), 4829-4832 (2010)


23/40

23

23

B) How to Find & Develop Systems Showing Phase Transitions:

B-1: Basic Aspects

We aim to develop the bi- or multi-stable systems which exhibit competitive or cooperative interactionsconcerning with 1) protons and electrons, 2) the itinerant and/or localized electrons, 3) monomerdimer, 4)phonons, molecular conformations or molecular vibrations and electrons, and 5) electrons and molecularions.

Figure 18 shows the iconicity diagram concerning with the proton transfer (PT) and charge transfer(CT) interactions between aromatic amines and polynitrophenols.[A-1-4k,l] pKa [= pKa(acid) (14

pKb(amine))] value is a governing parameter for the ground state of the complexes: proton-transferred solid ismore stable than CT one for negative pKa combination (complex with strong base), while CT solid is

preferred for positive pKa(complex with weak base) and isomeric complexes are obtained at the boundary(ex. 32 in Fig. 18).

All of the organic molecules have multi-functional natures and provide plural intermolecularinteractions depending on the nature of counter component, morphology (solid, films, uni-molecule, etc.) andexternal circumstances. For example, the CT interaction between D and A molecules in solid are brokendown into two kinds of interactions; Interaction I: electron transfer from neutral D to A molecules that costs

Figure 18. Complexes between aromatic amines (1~38)

and picric acid (pKa= 0.96), 2,4-dinitrophenol (pKa=4.09), or 2,6-dinitrophenol (pKa = 3.58). Complexes ofPT type and of CT type are plotted above and below thehorizontal line, respectively. Some aromatic amines are 1:

N,N-diethyl-m-toluidine, 14: aniline, 18: m-anisidine, 19:-chloroaniline, 23: m-chloroaniline, 25: o-chloroaniline,

32: 2,5-dichloroaniline, 36: skatole, 38: indole.

Figure 19. (a) A schematic balance between ionization energy (IDEA) and Madelung cohesive energyM. (b) Model doubleminimum potential for the N-I system for switching (upper, potential barrier between stable and metastable states is small andthermally accessible) and memory (lower, potential barrier is rather high E >> kBT). one- (c) and two-dimensional (d)diagrams for searching the boundary zone and functional materials. (d) is a schematic diagram corresponding to Fig. 2.

EA = ID M

EA

M

Ionic

Neu

tral ID

Strong D

StrongA

(d)(ID-EA) M(ID-EA) M

(a)

IDEA=M

(b)

Switching

MemoryE

N

I

M

Neutral SolidIonic Solid

(ID EA)

(c)

I

II


24/40

24

24

(IDEA) and Interaction II: Madelung energyM, as described in Section A-1-1. The situation is schematicallyillustrated in Fig. 19. Figure19arepresents the system with double-minimum potential as a balance like aYajirobe of Japanese toy, where the boundary condition is (IDEA) M. The condition is experimentallydetermined by a one-dimensional (Fig.19c, i.e., Figs.3b,18) or two-dimensional plot (Fig.2, 3a) for thecombination of D and A. One can expect that the balance between the interactions I and II can be controlledeasily by the external stimuli. The controllability increases as the system approaches to the boundary area.The system shows a variety of phase transitions (i.e., metal insulator, Mott insulator metal, spin-liquidsuperconductor, and neutral ionic, valence tautomerization), monotropic (ex. complexes 46in Fig. 2,32 in Fig. 18) and enantiotropic (ex. TTFp-chloranil) complex isomerizations, and switching or memoryeffect depending on the potential depths and barrier height E in Fig. 19b. Also it shows multiplestoichiometry, polymorphism, or phase transition in solid depending on the number of potential minima.

The Mott insulators, charge ordered insulators, and N-I transition systems exhibit a fast phase transition(switching or memory phenomenon) induced by electric field application or photo irradiation. For the formertwo systems, the phase transitions caused the pronounced change in reflectance and conductivity from

insulating to metallic features. The third system also exhibits a change in conductivity and dielectric responseconnected with the transports of solitons and/or domain walls, dynamic dimerization, and ferroelectricity.For these transition systems, the following 5 parameters are important for the development of materials.

1) Response timewhich can be controlled by taking into account the origin of the transition; a) fs for pureelectronic transition, b) addition of molecular deformation leads ps response time, c) further addition of latticedeformation leads slower response time >ps.2) Coherency (transition efficiency) depends on the degree of electronphonon coupling,electronmolecular vibration coupling, or electron-phonon coupling.3) Response temperature (operating temperature) may decrease in the following order; bondformationcleavage molecular deformation lattice deformation electronic deformation such asSDW and charge-order melting.

4) Thresholdof the external stimuli should not be zero to have clear switching and memory.5) Durabilityof the system.

Table 2 summarizes CT solids having switching behavior based on the transitions such as Mott(spin-Peierls) insulator mixed valency (metal), Mott or Peierls insulator metal, charge-ordered (CO)insulator metal, and neutral ionic, and are compared with superconductor (TMTSF)2ClO4 showingsliding SDW.

Table 2. Selected CT solids having switching behavior under electric field and/or photon irradiationare compared with the sliding SDW.Mechanism CT solid Electric field Photon irradiation

EthV/cm

ToperK

Photondensity

Sensitivity Responsetime

Toper

Mott or spin-PeierlsInsulator

Mixed valency orMetal

CuTCNQ 410 RT 1500W/cm RTCuTNAP 810 RT KTCNQ >10


25/40

25

25

It should be emphasized that most of -molecules we use are soft acid or soft base in Pearson's term.Thus they have a huge electronic polarizability, and are susceptible to external stimuli when the ionicity andmutual orientation of the molecules are appropriately designed. In order to find a system which affords suchintriguing phenomena, one should elucidate the nature of the essential intermolecular interactions governingthe phenomena by breaking down them into the physical parameters, followed by the translation from the

physical ones to chemical parameters. Then it is easy to find and develop the system being close to theproximity of the boundary zone.B-2: Ultrafast Photo-Induced Phase Transition in (EDO)2X

To destabilize the metallic state of the BO complexes, the elimination of one ethylenedioxy group(BOEDO)[A-1-4e,f]was very efficient owing to the weakened self-assembling ability (Fig.4). The (EDO)2Xsalts (X=PF6, AsF6, and SbF6) are 3/4-filled band conductors with a quasi-one-dimensional Fermi surface andexhibit a first-order MI transition (Fig. 20) at rather high temperatures (240280 K).[1]The phase transitionconsists of cooperative mechanism with charge-ordering, anion order-disorder and Peierls-like instability,which induces a doubled lattice periodicity giving rise 2kFnesting. The high temperature metallic phase iscomposed of flat EDO molecules with +0.5 charge, while the low temperature insulating phase is composedof both flat monocations and bent neutral EDO molecules with charge-ordered stripe (+1, +1, 0, 0).[1b,c]Thisstripe is different from those so far known (0,+1, 0, +1) stripe for -(ET)2MM'(SCN)4,

[2]indicating that theneighbor-site Coulomb repulsion energy is not dominant compared to the transfer energy within the (EDO1+)2dimer.

Laser irradiation onto the insulating (EDO)2PF6 crystal induces a phase transition to the highlyconductive state within a few pico-second.[1e,f] The crystal surface was excited by laser irradiation with the

pulse width of 0.12 ps. The excitation photon energy (1.55 eV) was nearly resonant to the CT band at11.1103cm1(1.37 eV) directly reflecting the excitation of the charge ordered state. The reflectance changeR/Rfrom insulating to conductive states exhibits negative and positive maxima at the probe photon energyof 1.38 and 1.72eV, respectively (Fig. 21). The life-time of the photo-induced conductive phase stronglydepends on the excitation intensity. In case of 21018 cm3 excitation condition, the reflectance changecompleted within only about 1.5 ps. Therefore, it is said that the melting of charge ordered state accompanied

by the insulator-to-conductor phase conversion occurs within 1.5 ps just after excitation with the

Figure 20.Temperature dependence of (a) resistivity and (b) magnetic susceptibility of (EDO)2PF6. Blue shadow in(a) and arrows in (b) indicate the MI transition. (c) Molecular structures of neutral and monocationic EDOmolecules.[A-1-e] Calculated Fermi surface of (EDO)2PF6is depicted in Fig. 4d.

(b) (c)(a)

O

O

Cl

Cl

Cl

Cl

QCl4: p-chloranil

NC

NC

CN

CN

TNAP

N

N

Me

Me

NC

CN

DMDCNQI

N

SS

N

SS

NT

S

S

S

S TeMe

TeMe

MeTe

MeTeTTeC1-TTF

Scheme 8


26/40

26

26

threshold-like behavior (threshold photon density is 1018 cm3). The excitation intensity corresponds to asingle excitation photon for ca. 500 molecules. Within the resolution time (1 s), the electric conductivity waslargely enhanced (more than 5 orders of magnitude) just after photo-excitation.

In the (EDO)2X system, the TMI was tuned by the chemical modifications. Deuteration of EDO(d2-EDO) increased TMIby ca. 2.5 K, while the complexation with larger counter anions decreased TMIin the

order of X = ClO4 (>337 K) > PF6(278 K) > AsF6 (ca. 268 K) > SbF6 (ca. 240 K).[1g,h]

These findingsindicate the important role of Coulomb attractive energy in the MI transition of the system. The deformationof EDO molecules and local Coulomb interaction between EDO and anion molecules are anticipated totrigger the cooperative MI transition.

To realize a molecular phase-switching device controllable by light irradiation with 1 ps response time(i.e., THz region), it is essential to develop a material that shows highly sensitive and ultra-fast PIPT

phenomena near RT with high repeatability and durability. Such an ultra-fast transition has been observed in apurely electronic origin. Although the ultra-fast transition (within a few hundred fs) accompanied by themolecular conformational change has been observed for systems such as retinal in rhodopsin, [3] this is auni-molecular nano-system. As for the meso-size scale switch, an electron-lattice coupled coherent system isnecessary for the ultra-fast transition.B-3: C60

and Monomer-Dimer Phase Transition

As mentioned in Section A-4-4, C60radical anions have a strong tendency for dimerization, and when

they are allowed to approach each other they normally form diamagnetic single- or double-bonded (C60)2

dimers (Fig. 15). We have been working on the monomer-dimer transition of fullerenes.[4, A-4-14]Single crystalX-ray diffraction of Cp*2CrC60(PhCl2)2revealed that monovalent C60formed a single-bonded dimer (Fig.15a) at 100 K, whereas it was a freely rotating monomer at RT. [A-4-14b]The reversible transformation betweendiamagnetic dimer and paramagnetic monomer phases was confirmed by an abrupt change of staticsusceptibility at 200230 K (Fig. 22). The transition was accompanied by the changes in unit cell parameters,the decrease of the magnetic moment from 4.20 B(4.27 Bfor S=3/2 + 1/2) to 3.88 B(3.87 Bfor S=3/2)and the appearance of EPR signal from Cp*2Cr

+, simultaneously. The two latter effects were the result of thequenching of magnetism by the formation of diamagnetic (C

60

-)2 dimers. Similar transformation was

observed for C70anions in multicomponent complex (Cs+)2(CTV)(C70

)2(DMF)7(PhH)0.75.[4a]C70moleculesformed a dimer structure even at RT through a covalent single-bond (Fig. 15b), and showed no sizablemagnetic moment owing to the diamagnetic nature of (C70)2

2dimers. On heating, the magnetic susceptibilityabruptly increased at ca.360 K, possibly indicating the cleavage of the covalent bond and thus the formationof paramagnetic C70

monomers. A similar magnetic behavior was observed for an ionic C70salt with Bz2Cr,

Figure 21. Probe photon energy dependence of the time profile forR/Robservedat 180 K and 260 K. The pump photon energy (E//b) was 1.55 eV and the probephoton energy (E//b) was 1.72 and 1.38 eV for the upper and lower panels,respectively.The oscillations inR/Rrelate molecular deformation modes.

S

S S

S

O

OD

Dd2-EDO

Effec

tivemoment(

B)

Effec

tivemoment(

B)

Figure 22. Temperature dependence of magnetic moment forCp*2CrC60(PhCl2)2.

T / K

Scheme 9


27/40

27

27

Cp2Co, Cp*2M (M = Cr, Ni).[A-4-14c, 4b]Table 3summarizes the (fullerene)compounds in which dimers or

polymers are formed at low temperatures.[5] Most of them exhibit monomerdimer (or polymer) phasetransition.

It likely seems that the transition temperature is connected with the length of the bridging intradimerbond; C60(200230 K) < C70(360 K) < neutral C60(435 K) vs.C60(1.597 ) > C70(1.584 ) > neutralC60 (1.575 ). Also we expect the following order of dimer stability:-(C60

)2dimer < -(C60)2dimer of Type II, Fig. 29a). Molecular orbital calculations revealed that molecular

(hyper)polarizabilities can be modulated by tuning (Fig. 29b). I13CNQ-R exhibited a solvatochromic shift,and the ground state changed from neutral ( 0.5) in less-polar solvents to ionic ( 0.5) in polar solvents

(Fig. 29c). Two-photon absorption properties of I13CNQ-R showed a significant substituent effect andindicated that is a fundamental parameter for modulating non-linear optical properties (Fig. 29d).

C-3: Zwitter Ions (Betainic -Radical Molecules)

One of the potential candidates for single-component organic conductors is the betainic radical system

which possesses both cation and anion moieties as well as a radical electron as a conduction carrier. For the

betainic radical system, on-site Coulomb repulsion (U), which is an essential retarding factor for conduction

and leads to the insulating property, can be diminished, according to the LeBlancs proposal, U= (1 /r3),

where and r are the molecular polarizability and average charge distance, respectively.[47]Neilands et al.

Figure 29. a) Plot of chargevs.BLRfor In3CNQ-R molecules. Green circles and red triangles indicate molecules of Type I and IIconformations, respectively. The lines represent least squares fits for each plot. The numbers and symbols in the figure correspond tothose in Table 4 in Ref. 46i. The chemical structure indicates the separation of D (red) and A (blue) moieties and the -bridge isincluded in the A moiety because it has a negative charge. b) Plots of (hyper)polarizabilities ,, and obtained by the MOcalculation against dipolefor I13CNQ-R, I63CNQ-CF3, and In3CNQ-H (3: n = 3 and 10: n = 10)molecules. Green circles and redtriangles show Type I and II molecules, respectively. Solid lines are the theoretical curves calculated; = 3.15 1022(1)2+42.9,= 7.17 1043(1)3(12), and = 2.14 1054(1)4(15+ 52). Dashed lines are the least squares fittings for each

plot; = 7.32 1012(1)2+ 64.3,= 1.09 1053(1)3(12), and = 2.82 1054(1)4(15+ 52). c) Plots ofhCT(2nd) values of I13CNQ-R vs. Reichardts ET values in various solvents for (a) I13CNQ-F2, (b) I13CNQ-H, and (c)I13CNQ-(EtO)2. Solvents: a, MeOH (ET= 55.5); b, MeCN (46.0); c, acetone (42.2); d, PhCl (37.5); e, 1,4-dioxane (36.0); f, PhH(34.5); g, PhMe (33.9); and h, CS2 (32.6). d) Two-photon absorption spectra of (a) I13CNQ-CF3 in MeCN and (b)I13CNQ-(EtO)2 in CHCl3. Shaded zones cover the wavelength dependence of (2). (c) Plot of (2)max vs. solvof I13CNQ-Rderivatives. Vertical lines are the error bars, and the shaded zone covers the wavelength and dependences of (2)max. The dottedline in (c) represents(2)max= 4.03 10

3(1 )(1 2)2+ 60.3.

(a)

(d) Two-photon absorption

N-I transition by solvent polarity

per)polarizabilities vs

(c)

Hyper)polarizabilities vs

(b)


33/40

33

33

have demonstrated conducting betainic radicals based on 2,4-dioxo-(1H,3H)- and

2-amino-4-oxo-(3H)-pyrimido- fused TTF derivatives (1and 2, RT=103and 101Scm1, respectively).[12]

By taking this advantage, we have been studying for the single-component organic conductors based on

betainic radicals of pyrimido-fused TTF derivatives.[13]

One of the features of 1 and 2 is that only 23% Curie

spins per unit are observed in the magnetic susceptibilitymeasurement. This feature implies the strong antiferromagneticinteraction caused by the dimerization, which splits the half-filledmetallic band and forms an energy gap for conduction. Twodimerization processes can be drawn for betainic radicals of the

present system: (1) face-to-face stacks or side-by-side chalcogenatom interactions as often observed in the CT solids and (2)complementary hydrogen-bonds characteristic of nucleobasesystems. In addition, their very poor solubility to common organicsolvents, which is probably caused by intermolecularhydrogen-bonds, prevents us from the crystal structure analysisand deeper discussions on their transport properties.

To examine the effects of hydrogenbond interactions onbetainic radicals of pyrimido-fused TTF, we have synthesized newN-methyl substituted derivatives as shown in Fig. 30 (3H and4H).[13c] Their betainic radicals are expected to be free fromrobust hydrogen-bonds and can interact with neighboring molecules only byface-to-face -stacks or side-by-side SS contacts. In the crystal structures of their tetrabutylammoniumsalts, complementary hydrogen-bonds inherent in pyrimido-fused TTF derivatives were inhibited by the

methyl substitution, and the crystals were constructed by the segregated motifs of cations and anions. Betainicradicalsprepared by one-electron oxidation of tetrabutylammonium salts (3 and 4) exhibited relativelyhigh conductivities (ca. 104Scm1at RT) as single- component organic molecules. The optical measurementof betainic radicals showed considerably low-energy CT absorption between radical molecules compared tothose of conventional TTF systems, indicating the reduction of on-site Coulomb repulsion. Table 4summarizes the first CT band energy, room temperature conductivity and activation energy for conduction forthe betainic neutral radicals so far prepared in our group (1 4).

S

S

S

S

Me

MeN

N

O

O

S

S

S

S

Me

MeN

N

O

N

R

R

1 : R = H3 : R = Me

2 : R = H4 : R = Me

R

Scheme 12

Figure 30. Synthetic procedures for 3H and 4H, and the preparation method of neutral betainic radicals. (i) Tetramethyl- ortetraethylthiuramdisulfide, K2CO3, DMF, 90C; (ii) conc. H2SO4, 70C; (iii) Na2CO3; (iv) NaPF6 aqueous solution, RT; (v)

NaHSe, water or EtOH, RT; HCl aq or AcOHHCl; (vi) BTSA, benzene or toluene, reflux; and (vii) 2 equiv 8, excess PPh3, RT.


34/40

34

34

Table 4. The first CT transitions (hCT) in the UV-Vis-NIR spectra, room temperature conductivities

(RT) and activation energies (a) of 1

4

hCT (10

3cm-1)a) RT(S cm1)b) a(eV)

b)1

8.8 4.3 104 0.222

4.5 1.0 101 0.143

7.6 2.1 104 0.214

7.8 4.2 104 0.21

a: Measured for KBr pellets. b: Measured for compaction pellets.

Reference C1. J. Ferraris, D.O. Cowan, V. Walatka, J.H. Perlstein,J. Am. Chem. Soc.,95(1973)948.2. a) T. Ishiguro, K. Yamaji, G. Saito, Organic Superconductors, second ed.,Springer-Verlag, Berlin, 1998;

b) G. Saito, Y. Yoshida,Bull. Chem. Soc. Jpn., 80(2007) 1.3. T. Y

Documents

Development of Conductive Organic Molecular Assemblies