Abstract
Molecular “nanowire” structures composed of the charge transfer complex of a bis-tetrathiafulvalene substituted macrocycle and tetrafluorotetracyanoquinodimethane were constructed on mica substrates by employing the Langmuir–Blodgett technique. The nanowires transferred from a dilute aqueous potassium chloride subphase had typical dimensions of 2.5 nm × 50 nm × 1 μm. The nanowires are oriented to specific directions, corresponding to the directions of the potassium-ion array on the mica surface having sixfold symmetry. Such correlation between the nanowires and the substrate surface was also observed when a dilute aqueous rubidium chloride subphase was used. On the other hand, the correlation completely disappeared when the subphase contained divalent cations, indicating that the molecular nanowires orient by recognizing the monocation array on the mica surface. The nanowires formed by the vertical dipping method coexist with the monolayers. Only nanowire structures are, however, observed when we apply the horizontal lifting method. Based on the crystal structure of a related complex, a possible structure of the nanowires is presented. The conductivity of the nanowires was estimated to be of the order of 10−3 S⋅cm−1. The nanowires formed specific (regular) structures such as T-shape junctions, suggesting their use in construction of future molecular nanoscale devices.
Molecular electronic devices have been attracting much attention in the field of nanoscale electronics (1–5). Since the Aviram and Ratner (6) proposal of a molecular rectifier, significant progress in the field of molecular electronics has been witnessed in the past two decades. Electrical switching between on and off states of a Langmuir–Blodgett (LB) film sandwiched between metal electrodes was observed in 1988 (7). Such switching is now possible for a single-layer LB film and has been extended to logic gates (8–10). A perceptron as a model for biomimetic computation was also constructed by using switching phenomena in molecular films of a phthalocyanine (11). Rectification at the molecular level was achieved on single-layer films of an amphoteric molecule in which donor and acceptor units were connected by conjugated bonds (12).
It is expected that future molecular based nano-computing systems may be composed of a vast number of nano devices. The devices (switches, memories, and wires) are formed through self-assembly processes, which are integrated into higher-order components, electronic circuits for example, through the bottom-up approach (13). In such systems, formation of a certain amount of defects is inevitable during the device fabrication processes. Recently, J. R. Heath et al. (14) pointed out that one possible way to build a powerful computer containing defective components is to keep sufficient communication bandwidth in the system, and consequently, massive numbers of nanowires are essential for constructing the system.
In the present silicon-based technology, wiring is achieved by lithography technique at the submicrometer level. Single-walled carbon nanotubes are currently promising candidates for wires, as well as switches and transistors, in future nano-electronics (15, 16). However, the bottom-up wiring technique at the nanometer scale has not yet been established. In this communication, we present a molecular “nanowire,” using the bottom-up self-assembly processes (17) of a molecular charge transfer complex (18–21). The “nanowires” reported here are semiconducting, and form a structured network in which the orientation of each nanowire is regulated through the specific interaction with the solid surface.
The amphiphilic bis-tetrathiafulvalene (TTF) substituted macrocycle 1 is composed of three components: (i) redox active TTF units, (ii) a macrocyclic polyether part, and (iii) two long alkyl chains. Compound 1 and the related macrocycle 2 were synthesized as illustrated in Scheme S1 from the cyanoethyl derivatives 3 and 5 (22–24). TTF is a well known π-donor that forms one-dimensional conducting columns through π-interaction in the partially oxidized state and exhibits high electrical conduction along the TTF stacking direction (18–21). The macrocyclic polyether part can recognize alkali metal cations in solution (25), and the long alkyl chains have a hydrophobic character. Because the polyether part is hydrophilic, the final macrocyclic ligand 1 becomes amphiphilic and the LB method for the fabrication of molecular films (26) can be used. This technique was applied to the tetrafluorotetracyanoquinodimethane complex of 1, affording formed oriented “nanowires,” rather than the expected ordinary molecular thin films, when complex (1)(F4-TCNQ)2 was transferred to a mica surface.
Scheme 1.
Materials and Methods
General Methods and Instrumentation.
The complex (1)(F4TCNQ)2 was prepared in situ by combining 1 and two equivalents of F4TCNQ in a CHCl3-CH3CN (9:1) solution. The concentration of the solution was adjusted to 1 mM with respect to 1. A conventional Langmuir trough (NIMA 5152D) was used for monolayer formation. Surface pressure–area (F–A) isotherms at 290.5 K were recorded at the barrier speed of 10 mm/min. Films were deposited at 10 mN/m by single up-stroke withdrawal of substrate at 10 mm/min (vertical dipping method) or by the horizontal lifting method at 0 mN/m (around 2 nm2/molecule of 1). Atomic force microscope (AFM) images were taken by a SPA400 (Seiko Instruments) operating in taping mode by using microcantilevers of a spring constant of 0.2 N/m. Cyclic voltammetry was performed on 0.1–0.5 mM solutions of compounds 1, 2, 4, 6, and 7 in anhydrous CH2Cl2 with n-Bu4PF6 (0.1 M) as supporting electrolyte, using platinum electrodes and Ag/AgCl as reference electrode at a scan rate of 0.1 V/s. Plasma desorption mass spectrometry (PDMS) was carried out on a Bio-Ion (Uppsala, Sweden) 2DK instrument. All reactions were carried out under inert atmosphere.
Synthesis.
Tetrathiafulvalene 4.
To a stirred solution of compound 3 (24) (1.40 g, 3.00 mmol) in dry degassed N,N-dimethylformamide (DMF) (40 ml) was added a solution of cesium hydroxide monohydrate (0.554 g, 3.30 mmol) in dry degassed methanol (10 ml) dropwise over 30 min. The color changed from orange to dark red and stirring was continued for another 30 min. The solution was slowly added to a stirred solution of 1,2-bis(2-iodoethoxy)ethane (8.88 g, 24.0 mmol) in dry degassed DMF (50 ml) during 5 h, by means of a perfuser pump. The orange reaction mixture was stirred for another hour before it was concentrated in vacuo. The residue was dissolved in dichloromethane (200 ml), washed with water (2 × 100 ml), dried (MgSO4), and concentrated under reduced pressure. Purification by column chromatography [silica gel, eluent: dichloromethane-petroleum ether (bp 60–70°C) 1:1 vol/vol until all excess 1,2-bis(2-iodoethoxy)ethane was off the column, then dichloromethane-ethyl acetate 98:2 vol/vol] afforded 4 as a red oil that solidified on standing in vacuo; (1.73 g, 88%), mp 57–58°C. Elemental analysis for C17H22INO2S8 (655.8): calculated C 31.14, H 3.38, N 2.14; found C 31.30, H 3.32, N 2.01. 1H-NMR (CDCl3): δ = 3.78–3.70 (m, 4H), 3.66 (s, 4H), 3.27 (t, 2H, J = 6.8 Hz), 3.09–3.02 (m, 4H), 2.72 (t, 2H, J = 7.2 Hz), 2.44 (s, 6H). MS(PDMS): m/z = 655.5 (M+, calculated: 655.8). CV: E1 = 0.60 V, E2 = 0.92 V.
Bis-tetrathiafulvalene 6.
To a stirred solution of compound 5 (24) (1.08 g, 1.50 mmol) in dry degassed DMF (70 ml) was added a solution of cesium hydroxide monohydrate (0.277 g, 1.65 mmol) in dry degassed methanol (10 ml), dropwise over 30 min. The color changed from orange to dark red and the stirring was continued for another hour. Compound 4 (1.08 g, 1.65 mmol) dissolved in dry degassed DMF (15 ml) was added to the reaction mixture in one portion and stirring was continued for 12 h, whereupon the color changed back to orange. The solution was concentrated in vacuo, redissolved in dichloromethane (200 ml), washed with water (2 × 100 ml), dried (MgSO4), and concentrated under reduced pressure. Purification by column chromatography (silica gel, dichloromethane-ethyl acetate 99:1 vol/vol) followed by recrystallization from toluene/petroleum ether (bp 60–70°C) afforded 6 as an orange powder; (1.53 g, 85%), mp 75–76°C. Anal. C46H68N2O2S16 (1194.1): calculated C 46.27, H 5.74, N 2.35; found C 46.32, H 5.76, N 2.28. 1H-NMR (CDCl3): δ = 3.71 (t, 4H, J = 6.4 Hz), 3.64 (s, 4H), 3.09–3.02 (m, 8H), 2.82 (t, 4H, J = 7.2 Hz), 2.72 (t, 4H, J = 7.3 Hz), 2.44 (s, 6H), 1.63 (quintet, 4H, J = 7.4 Hz), 1.41 (quintet, 4H, J = 6.5 Hz), 1.27 (m, 24H), 0.88 (t, 6H, J = 6.7 Hz). MS(PDMS): m/z = 1,193.0 (M+, calculated: 1,194.1). CV: E1 = 0.59 V, E2 = 0.93 V.
Bis-TTF 7.
Prepared like 6 from compound 5 (24) (1.44 g,
2.00 mmol) and cesium hydroxide monohydrate (0.369 g, 2.20 mmol) except
that this time alkylation was carried out using
1,2-bis(2-iodoethoxy)ethane (0.370 g, 1.00 mmol). Purification by
column chromatography (silica gel, dichloromethane) followed by
recrystallization from toluene/petroleum ether (bp 60–70°C)
afforded 7 as an orange powder; (1.09 g, 76%), mp
88–89°C. Elemental analysis for
C64H104N2O2S16
(1,446.6): calculated C 53.14, H 7.25, N 1.94; found C 53.09, H 7.29, N
1.78. 1H-NMR (CDCl3):
δ = 3.71 (t, 4H, J = 6.4 Hz), 3.64 (s, 4H),
3.09–3.02 (m, 8H), 2.82 (t, 8H, J = 7.3 Hz), 2.72 (t,
4H, J = 7.3 Hz), 1.63 (quintet, 8H, J =
7.4 Hz), 1.41 (quintet, 8H, J = 6.6 Hz), 1.27 (m, 48H),
0.88 (t, 12H, J = 6.8 Hz). MS(PDMS):
m/z = 1,444.7 (M+,
calculated: 1,446.6). CV: E
= 0.59 V,
E
= 0.92 V.
Bis-tetrathiafulvalene ligand 1.
To a stirred solution of compound 6 (1.194 g, 1.00 mmol) in dry degassed DMF (45 ml) was added a solution of cesium hydroxide monohydrate (0.369 g, 2.20 mmol) in dry degassed methanol (5 ml) in one portion. The solution was stirred for 1 h and changed color from orange to dark red. Meanwhile, a second solution consisting of 1,2-bis(2-iodoethoxy)ethane (0.370 g, 1.00 mmol) in dry degassed DMF (50 ml) was prepared. The two solutions were added simultaneously under high dilution conditions during 17 h to dry degassed DMF (100 ml), stirring vigorously by means of a perfuser pump. The resulting orange suspension was stirred for another 3 h before it was poured onto water (500 ml). The orange precipitate was filtered, washed with methanol (2 × 10 ml), and dried for 24 h in a dessicator over phosphorous pentoxide to give 1 as an analytically pure pale brown powder; (1.141 g, 95%), mp 108–109°C. Elemental analysis for C46H72O4S16 (1202.1): calculated C 45.96, H 6.04; found C 45.77, H 5.89. 1H-NMR (CDCl3): δ = 3.71 (t, 8H, J = 6.3 Hz), 3.65 (s, 8H), 3.03 (t, 8H, J = 6.2 Hz), 2.82 (t, 4H, J = 7.4 Hz), 2.43 (s, 6H), 1.63 (quintet, 4H, J = 7.4 Hz), 1.40 (quintet, 4H, J = 6.5 Hz), 1.27 (m, 24H), 0.88 (t, 6H, J = 6.7 Hz). MS(PDMS): m/z = 1,201.4 (M+, calculated: 1,202.1). CV: E1 = 0.52 V, E2 = 0.90 V.
Bis-tetrathiafulvalene ligand 2.
Prepared like 1 from compound 7 (0.723 g, 0.500
mmol), cesium hydroxide monohydrate (0.185 g, 1.10 mmol) and
1,2-bis(2-iodoethoxy)ethane (0.185 g, 0.500 mmol). Compound
2 was obtained as a red-brown analytically pure powder;
(0.619 g, 85%), mp 110–111°C. Elemental analysis for
C64H108O4S16
(1,454.6): calculated C 52.85, H 7.48; found C 52.67, H 7.37.
1H-NMR (CDCl3): δ =
3.71 (t, 8H, J = 6.3 Hz), 3.65 (s, 8H), 3.03 (t, 8H,
J = 6.2 Hz), 2.82 (t, 8H, J = 7.3 Hz),
1.63 (quintet, 8H, J = 7.5 Hz), 1.40 (quintet, 8H,
J = 6.6 Hz), 1.27 (m, 48H), 0.88 (t, 12H,
J = 6.8 Hz). MS(PDMS): m/z
= 1,455.2 (M+, calculated: 1,454.6). CV:
E
= 0.53 V,
E
= 0.91 V.
Results and Discussion
A solution of the charge transfer complex (1)(F4-TCNQ)2 was spread onto pure H2O or a 0.01 M metal chloride [MCl (M = Li, Na, K, Rb, Cs) or BaCl2] aqueous subphase. Fig. 1 shows the surface pressure–area (F–A) isotherms of (1)(F4-TCNQ)2 on pure water and a KCl-containing subphase. Introduction of other metal cations into the subphase gave similar results as that of using an aqueous solution of KCl in the subphase. The significant change in the F–A behavior on introducing metal chloride in the subphase suggests the recognition of metal cations by the macrocyclic moiety of donor 1 at the air–water interface. The films were subsequently transferred onto freshly cleaved mica substrates, and the surface morphologies were observed by tapping-mode AFM.
Figure 1.
Surface pressure–area isotherm of (1)(F4-TCNQ)2 at 290 K on pure water (dashed line) and on a potassium chloride solution (solid line).
Figs. 2 shows the AFM images (10 × 10 μm2 area) of transferred films of (1)(F4-TCNQ)2 from pure H2O and those from 0.01 M alkali metal chloride solutions. Although the films showed a variety of the surface morphologies according to the metal cation introduced into the subphase, they had a uniform height of ≈2.5 nm in all cases. Bundles of nanowires are observed in the monolayer transferred from pure H2O (Fig. 2a). The nanowire bundles curl both clockwise- and anticlockwise-forming domains and some ends of the nanowires are extended. On the other hand, the nanowires deposited from the surface of a KCl solution clearly arranged along specific directions, forming angles of 60 and 120° to each other with a typical length extending over 1 μm (Fig. 2d). The crystal axes of the mica surface observed by AFM were checked separately by a Laue photograph before film deposition and are indicated by arrows. By comparing with the mica unit cell, the nanowires were seen to grow in [100], [210], and [−210] directions. Similar results were obtained when Rb+ cations were introduced in the subphase (Fig. 2e).‖
Figure 2.
AFM (10 × 10 μm2) images of molecular nanowires of (1)(F4-TCNQ)2 on mica substrates transferred by the vertical dipping method from (a) pure water and (b) 0.01 M LiCl, (c) 0.01 M NaCl, (d) 0.01 M KCl, (e) 0.01 M RbCl, (f) 0.01 M CsCl, and (g) 0.01 M BaCl2 aqueous solution. Arrows indicate the unit cell directions of the mica surface.
When a CsCl solution was used as a subphase, the average length of the nanowires became much shorter than those deposited from a KCl solution (Fig. 2f). However, the direction of the nanowire growth was maintained. Such nanowire orientation was to some extent also observed when the smaller size monocations Li+ and Na+ (Fig. 2 b and c) were used. On the other hand, the correlation between nanowire growth and the mica crystal lattice was completely destroyed when a Ba2+-containing subphase was used (Fig. 2g), even though Ba2+ has a cation size similar to K+.
Mica (muscovite) has a stoichiometry of KAl3Si3O10(OH)2, with K+ ions located at the interlayer of the Al3Si3O10(OH)2 hexagonal lattice (27, 28). Upon cleavage, the potassium sites appear at the surface, which have C6 symmetry. The results described above strongly suggests that the nanowires recognize the cation arrays on the mica surface during the film deposition process. The interlayer K+ cations are divided onto two mica surfaces upon cleavage, forming many vacant K+ sites at the surface. These sites are filled with K+ cations when a KCl solution is used as subphase, and consequently the nanowires orient hexagonally recognizing the K+ cation arrays on the mica surface. The same mechanism applies to Rb+, which can fill the cation sites in place of K+, having similar ion radius. Because a Cs+ cation is much larger than a K+ cation, the recognition process is partly reduced, although the cavity size of the macrocyclic part of 1 fits well to cesium cation, as suggested from a CPK (Corey, Pauling, Koltun) model. Smaller cations such as Li+ and Na+ are also insufficient to allow formation of an oriented nanowire network structure. The divalent cation Ba2+ cannot occupy the cation sites of the mica substrate because of strong Coulombic repulsion, and therefore completely prevents the surface recognition and subsequent formation of nanowires. A similar oriented structure of a π-conjugated polymer on mica surface has been observed (29, 30).
The nanowires formed by the vertical dipping method coexist with the monolayer films of (1)(F4-TCNQ)2, as is observed as brighter region in the background of the AFM images in Fig. 2. The transfer ratio from the aqueous surface to mica was slightly larger than unity, suggesting that the nanowire formation occurs during the transfer process. The nanowire may form during transformation of the monolayer, while at same time the remaining monolayer to some extent fills the space between nanowires on the mica surface. On the other hand, when we apply the horizontal lifting method for the layer on the KCl aqueous subphase, only nanowires were deposited on the mica surface (Fig. 3). It should be noted that the image size of Fig. 3 is 30 × 30 μm2 and consequently the nanowires extend over several micrometers.
Figure 3.

AFM (30 × 30 μm2) image of the nanowires on mica, formed by the horizontal lifting method deposited from 0.01 M KCl aqueous solution.
The nanowires formed using a K+- or Rb+-containing subphase had typical dimensions of 2.5 nm (height) × 50 nm (width) × 1 μm (length), the thickness of which is comparable to those of carbon nanotubes (≈2 nm; refs. 15 and 16), silicon nanowires (≈15 nm; ref. 31), and thiophene nanowires (≈60 nm; ref. 32). Considering the molecular size of 1, the nanowire should be composed of one to three molecules in the height direction and 20 to 30 molecules in the width direction. Given that the nanowire has a local structure similar to ordinary molecular conducting TTF-based materials, the longitudinal direction of the nanowire should correspond to the π–π stacking direction of TTF units and/or F4-TCNQ molecules. A few thousand molecules are, therefore, necessary to form a 1-μm-long nanowire, taking into account that the standard π–π stacking distance is ≈0.35 nm.
Fig. 4 shows a schematic view of the nanowire. The crystal structure indicated in the figure is that of the F4-TCNQ complex of the tetraalkyl substituted bis-TTF macrocycle (2), (2)(F4-TCNQ)2, viewed along the b axis. Note that the alkyl chains are omitted for clarity. The electronic states of the TTF units and F4-TCNQ molecules in the nanowire, as well as the (2)(F4-TCNQ)2 crystal, are in a fully charged transferred state, as seen by comparing the IR spectra with that of a reported complex (33) and (1a)(F4-TCNQ)2 (34).
Figure 4.
Schematic view of a molecular nanowire of (1)(F4-TCNQ)2. Crystal structures indicated are those for (2)(F4-TCNQ)2.
In the crystal, the TTF unit forms an intramolecular dimer structure and stacks along the a axis alternately with F4-TCNQ dimers. The long axes of TTF units and F4-TCNQ molecules are parallel to the a+c direction. Assuming that the complex (1)(F4-TCNQ)2 has a similar crystal structure as (2)(F4-TCNQ)2, the π-stacking direction (the a-axis direction) may correspond to that of the nanowire long axis, because the intermolecular interaction is strongest in this direction.**
The electrical conductivity of a 20-layer film measured using evaporated gold electrodes with the gap distance of 0.5 mm was around 10−3 S⋅cm−1, a value that is reasonable for a fully charge transferred mixed stack molecular complex. Although the accumulated film should be composed of piles of nanowires and monolayers of (1)(F4-TCNQ)2, each nanowire may have the same order of electrical conductivity, because the contact resistance between the nanowires may be negligible at this level of conductivity. The films showed a semiconducting behavior in the measurement of temperature-dependent conductivity.
Fig. 5 shows selected nanowire structures observed at higher magnification. The morphology of the nanowires depends not only on the surface pressure and temperature on deposition, but also on the K+ concentration in the subphase. A characteristic T-shape junction, a parallel nanowire array, and a nanowire tree structure are seen in the figures. These structures suggest applications in molecular electronics such as a three-terminal device, a crossbar switch, and a network device, although further optimization of the molecular design will be needed, especially a higher conductivity.
Figure 5.

Selected images of molecular nanowires structures. (a) T-shape junction; (b) parallel nanowire array; (c) nanowire tree.
Conclusion
In conclusion, we have demonstrated formation of oriented nanowires on mica surface. The bottom-up self-assembly processes for fabricating electronic circuits from individual molecules is one of the key technologies for realizing molecular electronics. Therefore, development of nanowire networks and improvement of their electronic properties, as well as formation of highly conducting or metallic nanowires, are especially important. The latter may be achieved through careful design of the charge transfer complex (electronic dimensionality, band-width, carrier concentration, etc.). The guiding principles for improved electronic properties of TTF-radical cation complexes and salts are well documented in the field of molecular conductors (18–21).
Acknowledgments
We are grateful to Drs. Reiko Azumi, Hiroaki Tachibana, and Mutsuyoshi Matsumoto, National Institute of Advanced Industrial Science and Technology, for fruitful discussions. This work was partly supported by a Grant-in-Aid for Science Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan, and by the Proposal-Based New Industry Creative Type Technology R&D Promotion Program from the New Energy and Industrial Technology Development Organization (NEDO) in Japan. A grant from the Danish National Science Research Council (SNF) to J.B. is gratefully acknowledged.
Abbreviations
- AFM
atomic force microscope
- TTF
tetrathiafulvalene
- DMF
dimethylformamide
- PDMS
plasma desorption mass spectrometry
Footnotes
This paper was submitted directly (Track II) to the PNAS office.
The Rb+ cation has almost the same ion radius as K+.
It should be noted that (2)(F4-TCNQ)2 did not show any nanowire structure when using the Langmuir–Blodgett (LB) method, and also that complex (1)(F4-TCNQ)2 did not form good-quality single crystals.
References
- 1.Carter F L, editor. Molecular Electronic Devices. New York: Dekker; 1987. [Google Scholar]
- 2.Aswell G J, editor. Molecular Electronics. New York: Wiley; 1992. [Google Scholar]
- 3.Petty M C, Bryce M R, Bloor D. Introduction to Molecular Electronic Devices. New York: Oxford Univ. Press; 1995. [Google Scholar]
- 4.Aviram A, Ratner M, editors. Molecular Electronics: Science and Technology. New York: New York Academy of Science; 1998. [Google Scholar]
- 5.Joachim C, Gimzewski J K, Aviram A. Nature (London) 2000;408:541–548. doi: 10.1038/35046000. [DOI] [PubMed] [Google Scholar]
- 6.Aviram A, Ratner M R. Chem Phys Lett. 1974;29:277–288. [Google Scholar]
- 7.Sakai K, Matsuda H, Kawada H, Eguchi K, Nakagiri T. Appl Phys Lett. 1988;53:1274–1276. [Google Scholar]
- 8.Collier C P, Mattersteig G, Wong E W, Luo Y, Beverly K, Sampaio J F, Raymo M, Stoddart J F, Heath J R. Science. 2000;289:1172–1175. doi: 10.1126/science.289.5482.1172. [DOI] [PubMed] [Google Scholar]
- 9.Wong E W, Collier C P, Bhloradský M, Raymo F M, Stoddart J F, Heath J R. J Am Chem Soc. 2000;122:5831–5840. [Google Scholar]
- 10.Heath J R. Pure Appl Chem. 2000;72:11–20. [Google Scholar]
- 11.Nichogi K, Taomoto A, Yoshida K. Jpn J Appl Phys. 1992;31:3464–3468. [Google Scholar]
- 12.Metzger R M, Chen B, Hopfner U, Lakshmikantham M V, Vuillaume D, Kawai T, Wu X, Tachibana H, Hughes T V, Sakurai H, et al. J Am Chem Soc. 1997;119:10455–10466. [Google Scholar]
- 13.Philp D, Stoddart J F. Angew Chem Int Ed Engl. 1996;35:1154–1196. [Google Scholar]
- 14.Heath J R, Kuekes P J, Snider G S, Williams R S. Science. 1998;280:1716–1721. [Google Scholar]
- 15.Dekker C. Phys Today. 1999;52:22–28. [Google Scholar]
- 16.Rueckes T, Kim K, Joselevich E, Tseng G Y, Cheung C L, Lieber C M. Science. 2000;289:94–97. doi: 10.1126/science.289.5476.94. [DOI] [PubMed] [Google Scholar]
- 17.Lehn J-M. Supramolecular Chemistry. Weinheim: VCH; 1995. [Google Scholar]
- 18.Farges J-P, editor. Organic Conductors. New York: Dekker; 1994. [Google Scholar]
- 19.Nalwa H S, editor. Handbook of Organic Conductive Molecules and Polymers. Stuttgart, Germany: Wiley; 1997. [Google Scholar]
- 20.Williams J M, Ferraro J R, Thorn R J, Carlson K D, Geiser U, Wang H H, Kini A M, Whangbo M-H. Organic Superconductors. Englewood Cliffs, NJ: Prentice–Hall; 1992. [Google Scholar]
- 21.Ishiguro T, Yamaji K, Saito G. Organic Superconductors. 2nd Ed. New York: Springer; 1998. [Google Scholar]
- 22.Nielsen M B, Lomholt C, Becher J. Chem Soc Rev. 2000;29:153–164. [Google Scholar]
- 23. Nielsen, M. B. & Becher, J. (1997) Liebigs Ann. Recueil 2177–2187.
- 24. Simonsen, K. B., Svenstrup, N., Lau, J., Simonsen, O., Mørk, P., Kristensen, C. J. & Becher, J. (1996) Synthesis, 407–418.
- 25.Gokel G W. Crown Ethers & Cryptands. Cambridge, U.K.: R. Soc. Chem.; 1994. [Google Scholar]
- 26.Petty M C. Langmuir-Blodgett Films: An Introduction. Cambridge, U.K.: Cambridge Univ. Press; 1996. [Google Scholar]
- 27.Wyckoff R W G. Crystal Structures. Vol. 4. New York: Wiley; 1960. [Google Scholar]
- 28.Lima-de-Faria J. Structural Mineralogy: An Introduction. Dordrecht, The Netherlands: Kluwer; 1994. [Google Scholar]
- 29.Samorı̄ P, Francke V, Müllen K, Rabe J P. Thin Solid Films. 1998;336:13–15. [Google Scholar]
- 30.Samorı̄ P, Francke V, Müllen K, Rabe J P. Chem Eur J. 1999;5:2312–2317. [Google Scholar]
- 31.Hu J, Odom T W, Lieber C M. Acc Chem Res. 1999;32:435–445. [Google Scholar]
- 32.Bjørnholm T, Hassenkam T, Greve D R, McCullough R D, Jayaraman M, Savoy S M, Jones C E, McDevitt J T. Advanced Materials. 1999;11:1218–1221. [Google Scholar]
- 33. Akutagawa, T., Abe, Y., Hasegawa, T., Nakamura, T., Inabe, T., Christensen, C. A. & Becher, J. (2000) Chem. Lett. 132–133.
- 34.Meneghetti M, Pecile C. J Phys Chem. 1986;84:4149. [Google Scholar]




