Organic Redox Systems: Synthesis, Properties, and Applications

PREFACE

One-electron oxidation or reduction of organic compounds generates generally highly reactive cation or anion radicals. However, the diversity in the design of organic and organo main group compounds enables the construction of various novel transition-metal-free molecules that show reversible redox reactions at room temperature. Such redox systems not only provide a new methodology for understanding and applying a unique chemical bonding and electronic state, organic conductors and magnets, and so on but also help develop organic materials for field-effect transistors, photovoltaics, secondary batteries, and so on. Consequently, organic redox systems are emerging as the target of intensive investigations in recent years.

The present book covers fundamental research related to novel π-electron systems showing reversible redox reactions. The previous books dealing with designed redox systems (e.g., Redox Systems under Nano-Space Control, Springer) mainly focused on transition metal complexes in which the transition metals play a vital role in the redox reaction. The books on stable radicals (e.g., Stable Radicals, Wiley) partially involved some organic redox systems, but most of the compounds summarized in the books were oxygen- or nitrogen-centered neutral radicals. The other old books of ion radicals (e.g., Radicals, Ion Radicals, and Triplets, Wiley) dealt with the reactive intermediate or species observable at very low temperatures. Thus, as far as I am aware, there is no similar book that provides specific information on the transition-metal-free redox systems based on new π-conjugated organic and organo main group compounds.

The first half of this book provides accounts mostly on the redox systems in which conventional elements of nitrogen, oxygen, and sulfur play an essential role in stabilizing the redox states. On the other hand, in the second half, hydrocarbons, heteroles (in which the radical center of the oxidized state tends to locate on the carbon atoms), and unconventional elements of aluminum and heavier group 14 and 15 elements are the leading part of the systems. Complete coverage of the entire organic redox systems from the huge list of organic substances is not the aim of this book, and readers may miss some specific compounds. Rather, it is my hope that this book offers sufficient current information, especially about how the structure and chemical element affect the redox properties of π-systems. In this regard, I express my sincere gratitude to all contributors of each chapter who have provided an informative and interesting overview of their expertise.

Tohru Nishinaga

Spring 2015

1

INTRODUCTION: BASIC CONCEPTS AND A BRIEF HISTORY OF ORGANIC REDOX SYSTEMS

Tohru Nishinaga

Department of Chemistry, Graduate School of Science and Engineering, Tokyo Metropolitan University, Hachioji, Tokyo, Japan

1.1 REDOX REACTION OF ORGANIC MOLECULES

Redox is a portmanteau word of reduction and oxidation. Originally, oxidation meant a chemical reaction in which oxygen combines with another substance, after Antoine Lavoisier, late in the eighteenth century, called a product of the reaction an oxide [1]. The term reduction had been used long before the introduction of the term oxidation in the smelting to produce iron from ore and coke [1]. In the contemporary definition recommended by IUPAC [2], oxidation is a reaction that satisfies criteria 1 the complete, net removal of one or more electrons from a molecular entity and 2 an increase in the oxidation number of any atom within any substrate and meets in many cases criterion 3 gain of oxygen and/or loss of hydrogen of an organic substrate. Conversely, reduction is the reverse process of oxidation.

For transition metals, a direct one-electron transfer related to the aforementioned criterion 1 is common due to their relatively lower ionization energy in comparison with main group elements [3] and low reactivity of the unpaired d-electrons. In contrast, the mechanisms of common organic redox reactions do not involve a direct one-electron transfer [4], and reactions based on the criterion 3 are typical. For example, oxidation of primary alcohol (RCH2OH) to aldehyde (RHC═O) with Cr(VI)O3 proceeds via chromic ester intermediate (RCH2O3Cr(VI)OH), and proton and HOCr(IV)O2− are eliminated from the intermediate [5] (Scheme 1.1a). In this reaction, the total number of electrons in the outer shell decreases from 14 at the C─O moiety to 12 at the C═O moiety, that is, two-electron oxidation, while the formal oxidation number of Cr changes from +6 to +4, that is, two-electron reduction. Similarly, reduction of carbonyl group to alcohol with NaBH4 in ethanol formally proceeds via nucleophilic attack of a pair of electrons in hydride to electron-deficient carbonyl carbon [5] (Scheme 1.1b). Thus, formally, a pair of two electrons moves together in typical organic redox reactions as known in other organic reactions such as substitutions.

SCHEME 1.1 (a) Oxidation of alcohol to aldehyde with Cr(VI) and (b) hydride reduction of aldehyde to alcohol.

On the other hand, one-electron oxidation or reduction of a neutral or ionic molecule (Scheme 1.2) gives generally highly reactive ion radicals or radicals, and follow-up reactions such as radical coupling and deprotonation are prone to take place [6]. Nevertheless, some organic molecules give persistent species after one-electron transfer at ambient temperature [7, 8]. Simple π-extension and substituents of resonance electron donating R2N─, RO─, RS─ or withdrawing N≡C─, C═O groups cause delocalization of spin and charge density, which reduces the reactivity of the reactive center. As the other thermodynamic stabilization, aromatization after electron transfer plays an important role for certain molecules. An appropriate steric protection is also an effective strategy for protecting a reactive radical center [9]. As a result of these effects, they can be reversibly regenerated by the reverse electron transfer. This book deals with organic π-electron systems and related organo main group compounds that show such reversible one-electron transfer.

SCHEME 1.2 One-electron oxidation and reduction of neutral and ionic molecules.

1.2 REDOX POTENTIAL IN NONAQUEOUS SOLVENTS

Redox potential is the important measure for redox systems, by which one can predict how easily one-electron oxidation or reduction takes place with other redox reagents. For the measurement of redox potential, cyclic voltammetry is usually the first choice, because not only the redox potential but also the stability of the species generated after electron transfer can be observed. Several types of reference electrodes are used to measure redox potentials. The standard hydrogen electrode (SHE) or normal hydrogen electrode (NHE), which is determined by redox potential of 2H+/H2 couple in an aqueous media, is defined as 0 V in standard electrode potential. However, since the setting of apparatus of SHE is complicated, other reference electrode such as saturated calomel electrode (SCE) and saturated Ag/AgCl or Ag/Ag+ electrode is commonly used for routine laboratory experiments. A saturated aqueous KCl solution is used for SCE and saturated Ag/AgCl electrodes, while polar solvent, for example, acetonitrile can be used for Ag/Ag+ electrode.

As for the absolute electrode potential, the value −4.44 ± 0.02 V vs NHE (25°C in H2O) is recommended by IUPAC [10]. The standard and absolute electrode potentials of NHE, SCE (=0.244 V vs NHE 25°C in H2O) [11], and saturated Ag/AgCl (=0.199 V vs NHE 25°C in H2O) [12] are shown in Figure 1.1. Since the potential of Ag/Ag+ electrode in a nonaqueous solvent varies with the conditions (solvent polarity, electrolyte, surface of Ag, etc.), the conversion of the Ag/Ag+ scale to SCE or Ag/AgCl scale is not straightforward.

FIGURE 1.1 Conversion of relative electrode potentials into electronic energies for aqueous systems. Note that this graph cannot be used to convert SCE or Ag/AgCl scale into Fc/Fc+ scale in the electrochemical measurements performed in nonaqueous media.

Most organic redox compounds do not dissolve in water, and hence their electrochemical measurements have to be taken in a nonaqueous solvent such as dichloromethane, DMF, and acetonitrile. In the case of the use of Ag/Ag+ reference electrode in a polar organic solvent, a careful preparation of the reference electrode is required for the reproducible measurements. If an SCE or saturated Ag/AgCl electrode is used as reference, liquid junction potential [13] generated between the aqueous media in reference electrode and the organic solvent used in the measurement cell cannot be negligible. Liquid junction potential causes a shift in the observed value from the inherent redox potential. The liquid junction potentials between saturated aqueous KCl solution and various aprotic polar organic solvents were shown to be 100–200 mV [14]. Occasionally, liquid junction potential exceeds 200 mV [13].

For this reason, IUPAC recommends the use of ferrocene/ferrocenium couple as internal reference for electrochemical measurements in a nonaqueous medium and also to report the potential in the scale against the redox potential of ferrocene (the abbreviation for the potential as V vs Fc/Fc+) [15]. The observed potential of Fc/Fc+ couple in various solvents and supporting electrolytes using an SCE reference electrode were reported [16]. The selected data are shown in Table 1.1. The observed values both in tetra-n-butylammonium hexafluorophosphate (TBAPF6) and perchlorate (TBAClO4) electrolytes tend to increase with decreasing solvent polarity. The liquid junction potential between aqueous media in the SCE reference electrode and the organic solvents is involved in the observed difference in the potentials for the Fc/Fc+ couple. Therefore, care must be taken when comparing the reported data in SCE or Ag/AgCl scale measured in different solvents. It is important to understand that such a comparison involves an unknown potential shift caused by liquid junction potential. Nevertheless, because of reproducibility of an SCE reference electrode even in nonaqueous media, the potentials in Table 1.1 can be used for the conversion from SCE scale to Fc/Fc+ scale, when the measurement conditions (solvent and supporting electrolyte) are identical [16].

TABLE 1.1 Formal Potentials (V) for the Ferrocene/Ferrocenium Couple vs SCE [16]

The HOMO and LUMO levels of organic redox compounds are often estimated from the oxidation (Eox V vs Fc/Fc+) or reduction (Ered V vs Fc/Fc+) potential obtained from electrochemical measurements and the energy level of Fc/Fc+ couple to vacuum (EFc/Fc+ V(abs)) by Equations (1.1) and (1.2) as follows:

(1.1)

(1.2)

For the EFc/Fc+ value, 4.8 eV is frequently used. The value was originally reported in 1995 [17] based on rather crude approximation that the absolute electrode potential for NHE was −4.6 V (the data from an older book) and that redox potential of Fc/Fc+ couple was 0.2 V vs NHE in acetonitrile (which is not consistent with the later value shown in Table 1.1). Then, the problems of the rough estimation were raised in 2011 [18].

If 0.40 V vs SCE for the redox potential of Fc/Fc+ couple in acetonitrile and TBAPF6 electrolyte [16] is used under neglecting solvent and electrolyte effects including liquid junction potential (i.e., 0.64 V vs NHE using the conversion scale shown in Figure 1.1), the energy level of Fc/Fc+ couple is estimated to be −5.1 eV [18] based on the absolute electrode potential −4.44 V vs NHE. Since the liquid junction potential caused by different electrolytes in the same solvent media is usually smaller than that between aqueous and aprotic nonaqueous media [13], 0.400 ± 0.005 V vs NHE for the redox potential of Fc/Fc+ couple in aqueous media [19] may give better estimation. In this case, the energy level of Fc/Fc+ couple is estimated to be 4.84 ± 0.025 eV (= (4.44 ± 0.02) + (0.400 ± 0.005)) below the vacuum level (Fig. 1.1; see also Section 11.3). Although difference in solvation of analyte between aqueous and nonaqueous media would cause some shift of the observed potential referred to Fc/Fc+ couple, such a shift may be smaller than the shift caused by liquid junction potential. For example, the difference in redox potential of bis(biphenyl)chromium referred to Fc/Fc+ couple between in highly polar propylene carbonate (dielectric constant ε = 64 (H2O ε = 80)) and in less polar dichloromethane (ε = 9) was only 34 mV [15]. In any case, it is encouraged to disclose the method by which HOMO/LUMO energies are estimated [18].

1.3 A BRIEF HISTORY OF ORGANIC REDOX COMPOUNDS

Redox reaction catalyzed by enzyme (oxidoreductase) is one of the important metabolic processes. Apart from cytochrome and ferredoxin bearing iron–porphyrin core and iron–sulfur clusters, transition-metal-free coenzymes, for example, nicotinamide adenine dinucleotide (NAD+) and ubiquinone (n = 10, CoQ10) (Scheme 1.3), play a central role in redox metabolisms. In this context, living organisms have utilized organic redox reaction from time immemorial. Among these coenzymes, NAD+ shows one-step two-electron reduction as observed in many organic redox reactions, while ubiquinone shows stepwise one-electron process involving the radical intermediate (ubisemiquinone) [20].

SCHEME 1.3 Redox reactions of transition-metal-free coenzymes.

The synthetic organic redox compounds at the early stage were brought about together with pursuing synthetic dyes and medicines. In 1826, Otto Unverdorben isolated aniline for the first time by destructive distillation of natural dye, indigo (Fig. 1.2) [21]. This report was prepared 2 years before Friedrich Wöhler’s pioneering discovery in organic chemistry that the organic compound of urea CO(NH2)2 can be synthesized from the inorganic compound of ammonium cyanate NH4CNO [22]. Then, aniline was obtained by various ways such as distillation from coal tar [23] and reduction of nitrobenzene [24]. However, these products had not been recognized as the same compound, until in 1843 August Wilhelm von Hofmann, who named the term aromatic for benzene derivatives, revealed the fact [25]. After the first synthetic dye mauveine, which was accidentally prepared during the attempt synthesis of quinine from aniline by Hofmann’s assistant William Henry Perkin in 1856 [26, 27], various synthetic dyes were prepared from aniline and related derivatives.

FIGURE 1.2 Structures of aniline, indigo, and mauveine.

Thus, radical cation of N,N,N′,N′-tetramethyl-p-phenylenediamine (TMPD (Fig. 1.3), E1/2ox1 = −0.27 V vs Fc/Fc+, E1/2ox2 = 0.33 V in CH3CN) [28] known as Wurster’s blue was isolated in 1879 [29]. As a related compound derived from aniline, methylene blue, which is a well-known redox indicator [30] and the first synthetic medicine [31], was prepared 3 years earlier than Wurster’s salts in German chemical company BASF (Badische Anilin und Soda-Fabrik (English: Baden Aniline and Soda Factory)). However, methylene blue shows one-step two-electron reduction [30] and does not generate an open-shell species in aqueous media.

FIGURE 1.3 Structures of TMPD and methylene blue. The redox potentials are vs Fc/Fc+.

Other anilines that give stable radical cations are triarylamines. Synthesis of neutral triphenylamine (TPA; Fig. 1.4) was reported in 1873 [32]. Although radical cation of TPA immediately gives tetraphenylbenzidine (TPB) via radical coupling followed by deprotonation [33], the isolation of radical cation salt of its p-methyl derivative (tritolylamine (TTA), E1/2ox1 = 0.33 V vs Fc/Fc+ in CH2Cl2) [34] was reported in 1907 [35]. Because of the stability at the oxidized state, various derivatives of triarylamines radical cations are used for oxidizing reagents [16]. Furthermore, their amorphous nature due to the conformational flexibility of triarylamine moiety is advantageous when preparing pinhole-less thin film [36] that is an important prerequisite for the electronic devices. Thus, triarylamine derivatives such as 4,4′-cyclohexylidenebis[N,N-bis(4-methylphenyl)aniline] (TAPC) and N,N′-bis-(1-naphthalenyl)-N,N′-bis-phenyl-(1,1′-biphenyl)-4,4′-diamine (α-NPD) are commonly used as charge transport materials for organic photoconductors (OPCs) [37] and organic electroluminescence (OEL) [38, 39].

FIGURE 1.4 Structures of triarylamines. The redox potential is vs Fc/Fc+.

Many aromatic hydrocarbons react with alkali metals to give relatively stable radical anion in ethereal solutions. For example, Na-naphthalene [40] can be handled at room temperature and are often used as reducing reagents. Stable radical cation of some aromatic hydrocarbons such as perylene (Fig. 1.5; E1/2ox1 = 0.54 V vs Fc/Fc+, E1/2red1 = −2.12 V in CH3CN) [41] can be generated in concentrated sulfuric acid [42], although radical cation of smaller aromatic hydrocarbons are not stable enough. In 1954, Hideo Akamatsu and Hiroo Inokuchi demonstrated that charge transfer complex of perylene with bromine showed high conductivity up to 1.3 S cm−1 [43]. Before the study, organic compounds were regarded as insulator. However, this discovery opened the research field of molecular conductors and stimulated the development of various organic acceptors and donors.

FIGURE 1.5 Structures of perylene, TCNE, and TCNQ. The redox potentials are vs Fc/Fc+.

Initially, various strong acceptors such as tetracyanoethylene (TCNE, E1/2red1 = −0.29 V vs Fc/Fc+, E1/2red2 = −1.29 V in DMF) [16, 44] and tetracyanoquinodimethane (TCNQ, E1/2red1 = −0.32 V vs Fc/Fc+, E1/2red2 = −0.90 V in DMF) [16, 44] were synthesized [45, 46] in DuPont from late 1950s to mid-1960s after the commercial success of Teflon (polytetrafluoroethylene, PTFE) from mid-1940s. For TCNQ, aromatization plays an important role in the reversible two-step one-electron reduction at lower potentials (Scheme 1.4) in a similar manner to ubiquinone (Scheme 1.3).

SCHEME 1.4 Redox reactions of TCNQ.

As for the donor related to molecular conductors, tetrathiafulvalene (TTF, E1/2ox1 = −0.09 V vs Fc/Fc+, E1/2ox2 = 0.28 V in benzonitrile) [47] is most important. Although the mixture of dimethyl and diphenyl derivatives of TTF were reported in 1965 [48], and even the parent TTF appeared in a dissertation of Würzburg University in 1968 [49], the broad interest arose after the report of the synthesis in 1970 [50] and the metallic conduction of TTF–TCNQ in 1973 [51]. Similar to TCNQ, good donor ability of TTF is partly owing to aromatization from dithiolylidene to dithiolium moiety upon the oxidation process (Scheme 1.5). In a TTF–TCNQ crystal, partial electron transfer occurs from TTF to TCNQ which forms segregated columnar stacks. Therefore, holes and electrons are separated and can traverse in a one-dimensional direction along the TCNQ and TTF columns, respectively. Since then, numerous investigations have been carried out to develop molecular conductors [52] and supramolecular systems [53] based on TTF and related molecules. As a result, the first superconductors based on organic radical salts were observed in (TMTSF)2PF6 (TMTSF tetramethyltetraselenafulvalene) at 0.9 K and 6.9 kbar [54].

SCHEME 1.5 Redox reactions of TTF. The redox potentials are vs Fc/Fc+.

In 1976, Hideki Shirakawa revealed in Alan Graham MacDiarmid laboratory that positively doped polyacetylene with halogen vapor shows high conductivity [55]. The discovery of conducting polymer, awarded with Nobel Prize in Chemistry 2000, had an impact in the field of organic electronic materials. For example, in 1981, a prototype cell using polyacetylene for negative electrode with a LiCoO2-positive electrode was fabricated, which leaded to the birth of the current lithium-ion battery by using carbonaceous material into which lithium ions can be intercalated [56]. In the negative electrode, reduction and oxidation of electrode materials occurs during charging and discharging processes, respectively. The application of some organic and organo main group compounds to the electrodes in secondary batteries is presented in this book (see Chapters 4 and 6). As another application of conducting polymers, the first organic field-effect transistor based on polythiophene was reported in 1986 [57]. Then, oligothiophene and other conjugated polymers and oligomers have been shown to have semiconducting properties. Enormous organic semiconductors for field effect transistors have been reported and reviewed [58–62]. Some examples are also shown in this book (see Chapters 10, 11, 13, and 14).

The discovery of C60 and fullerenes in 1985 [63], which was awarded with Nobel Prize in chemistry 1996, and carbon nanotubes in 1991 [64], further pushed the development of carbon-based electronic devices. C60 shows reversible multistep one-electron oxidations as well as reductions (E1/2ox1 = 1.26 V vs Fc/Fc+, E1/2ox2 = 1.71 V, E1/2red1 = −1.06 V, E1/2red2 = −1.46 V, E1/2red3 = −1.89 V, in CH2Cl2), when weakly nucleophilic anion AsF6– is used as electrolyte in CV measurement (Fig. 1.6) [65]. Concerning the conductive properties of these carbon materials, superconductivity was observed for K3C60 in 1991 [66]. All armchair nanotubes are metallic, while other chiral nanotubes are semiconducting [67]. Thus, to understand and utilize these unique electronic properties, various fragments of fullerenes and nanotubes have also been synthesized. Since the topic has been summarized in one of the recent books published by Wiley [68], most of it will not be covered here, but a model system of graphene sheet and other unique hydrocarbon systems are introduced in this book (see Chapters 6, 9, and 10).

FIGURE 1.6 (a) Cyclic voltammogram of 0.15 mM C60 in 0.05 M TBAAsF6 CH2Cl2 solution. Working electrode: Pt disk (125 µm diameter). Scan rate: 1 V/s. T = 25°C. Dashed curve: as solid curve after background current correction.

Reprinted with permission from Ref. 65. © 2003 American Chemical Society.

(b) Drawing of C60. The redox potentials are vs Fc/Fc+.

In summary, organic redox systems have been developed in relation to the application of organic materials to functional dyes and electronic devices. The emerging interest in organic materials will further advance the development of organic redox systems and deepen the knowledge of their various properties such as electronic state and bonding interaction between stable radicals, magnetism, and conductivity.

REFERENCES

1. C. W. Mansfield, Oxidation Reduction Potentials of Organic Systems, Williams & Willkins, Baltimore, 1960.

2. P. Muller, Pure Appl. Chem. 1994, 66, 1077–1184.

3. P. S. Matsumoto, J. Chem. Educ. 2005, 82, 1660–1661.

4. M. B. Smith, March’s Advanced Organic Chemistry: Reactions, Mechanisms, and Structure: 7th Ed., John Wiley & Sons, Inc., Hoboken, 2013.

5. K. P. C. Vollhardt, N. Schore, Organic Chemistry Structure and Function: 6th Ed., W. H. Freeman, New York, 2010.

6. M. Schmittel, A. Burghart, Angew. Chem. Int. Ed. 1997, 36, 2550–2589.

7. R. G. Hicks, Ed., Stable Radicals, John Wiley & Sons, Ltd, Chichester, 2010.

8. T. Nishinaga, K. Komatsu, Org. Biomol. Chem. 2005, 3, 561–569.

9. K. Komatsu, T. Nishinaga, Synlett 2005, 187–202.

10. S. Trasatti, Pure Appl. Chem. 1986, 58, 955–966.

11. R. G. Bates, in Treatise on Analytical Chemistry: 2nd Ed., Eds., I. M. Kolthoff, P. J. Elving, John Wiley & Sons, Inc., New York, 1978, p. 793.

12. D. T. Sawyer, J. L. Roberts Jr., Experimental Electrochemistry for Chemists, John Wiley & Sons, Inc., New York, 1974.

13. K. Izutsu, Electrochemistry in Nonaqueous Solutions, Wiley-VCH, Weinheim, 2009.

14. J. W. Diggle, A. J. Parker, Aust. J. Chem. 1974, 27, 1617–1621.

15. G. Gritzner, J. Kůta, Pure Appl. Chem. 1984, 56, 461–466.

16. N. G. Connelly, W. E. Geiger, Chem. Rev. 1996, 96, 877–910.

17. J. Pommerehne, H. Vestweber, W. Guss, R. F. Mahrt, H. Bässler, M. Porsch, J. Daub, Adv. Mater. 1995, 7, 551–554.

18. C. M. Cardona, W. Li, A. E. Kaifer, D. Stockdale, G. C. Bazan, Adv. Mater. 2011, 23, 2367–2371.

19. A. M. Bond, E. A. McLennan, R. S. Stojanovic, F. G. Thomas, Anal. Chem. 1987, 59, 2853–2860.

20. J. F. Turrens, A. Alexandre, A. L. Lehninger, Arch. Biochem. Biophys. 1985, 237, 408–414.

21. O. Unverdorben, Ann. Phys. 1826, 84, 397–410.

22. F. Wöhler, Ann. Phys. 1828, 88, 253–256.

23. F. F. Runge, Ann. Phys. 1834, 107, 65–78.

24. N. Zinin, J. Prakt. Chem. 1842, 27, 140–153.

25. A. W. Hofmann, Justus Liebigs Ann. Chem. 1843, 47, 37–87.

26. O. Meth-Cohn, M. Smith, J. Chem. Soc. Perkin Trans. 1 1994, 5–7.

27. J. Seixas de Melo, S. Takato, M. Sousa, M. J. Melo, A. J. Parola, Chem. Commun. 2007, 2624–2626.

28. K. Hu, D. H. Evans, J. Electroanal. Chem. 1997, 423, 29–35.

29. C. Wurster, E. Schobig, Chem. Ber. 1879, 12, 1807–1813.

30. A. Hulanicki, S. Glab, Pure Appl. Chem. 1978, 50, 463–498.

31. R. H. Schirmer, B. Coulibaly, A. Stich, M. Scheiwein, H. Merkle, J. Eubel, K. Becker, H. Becher, O. Müller, T. Zich, et al., Redox Rep. 2003, 8, 272–275.

32. V. Merz, W. Weith, Chem. Ber. 1873, 6, 1511–1520.

33. E. T. Seo, R. F. Nelson, J. M. Fritsch, L. S. Marcoux, D. W. Leedy, R. N. Adams, J. Am. Chem. Soc. 1966, 88, 3498–3503.

34. S. Amthor, B. Noller, C. Lambert, Chem. Phys. 2005, 316, 141–152.

35. H. Wieland, Chem. Ber. 1907, 40, 4260–4281.

36. C. W. Tang, S. A. VanSlyke, Appl. Phys. Lett. 1987, 51, 913–915.

37. D. S. Weiss, M. Abkowitz, Chem. Rev. 2010, 110, 479–526.

38. U. Mitschke, P. Bauerle, J. Mater. Chem. 2000, 10, 1471–1507.

39. A. P. Kulkarni, C. J. Tonzola, A. Babel, S. A. Jenekhe, Chem. Mater. 2004, 16, 4556–4573.

40. N. D. Scott, J. F. Walker, V. L. Hansley, J. Am. Chem. Soc. 1936, 58, 2442–2444.

41. C. Koper, M. Sarobe, L. W. Jenneskens, Phys. Chem. Chem. Phys. 2004, 6, 319–327.

42. Y. Yokozawa, I. Miyashita, J. Chem. Phys. 1956, 25, 796–797.

43. H. Akamatsu, H. Inokuchi, Y. Matsunaga, Nature 1954, 173, 168–169.

44. R. Gross-Lannert, W. Kaim, B. Olbrich-Deussner, Inorg. Chem. 1990, 29, 5046–5053.

45. T. L. Cairns, R. A. Carboni, D. D. Coffman, V. A. Engelhardt, R. E. Heckert, E. L. Little, E. G. McGeer, B. C. McKusick, W. J. Middleton, R. M. Scribner, et al., J. Am. Chem. Soc. 1958, 80, 2775–2778.

46. D. S. Acker, W. R. Hertler, J. Am. Chem. Soc. 1962, 84, 3370–3374.

47. M. Hasegawa, K. Daigoku, K. Hashimoto, H. Nishikawa, M. Iyoda, Bull. Chem. Soc. Jpn. 2012, 85, 51–60.

48. H. Prinzbach, H. Berger, A. Lüttringhaus, Angew. Chem. Int. Ed. 1965, 4, 435.

49. G. Kiesslich, Dissertation, Würzburg University, 1968.

50. F. Wudl, G. M. Smith, E. J. Hufnagel, J. Chem. Soc. D 1970, 1453–1454.

51. J. Ferraris, D. O. Cowan, V. Walatka, J. H. Perlstein, J. Am. Chem. Soc. 1973, 95, 948–949.

52. P. Batail, Ed., Chem. Rev., 2004, 104, no. 11, special issue for Molecular Conductors.

53. D. Canevet, M. Salle, G. Zhang, D. Zhang, D. Zhu, Chem. Commun. 2009, 2245–2269.

54. D. Jérome, A. Mazaud, M. Ribault, K. Bechgaard, J. Phys. Lett. 1980, 41, 95–98.

55. H. Shirakawa, E. J. Louis, A. G. MacDiarmid, C. K. Chiang, A. J. Heeger, J. Chem. Soc. Chem. Commun. 1977, 578–580.

56. A. Yoshino, Angew. Chem. Int. Ed. 2012, 51, 5798–5800.

57. A. Tsumura, H. Koezuka, T. Ando, Appl. Phys. Lett. 1986, 49, 1210–1212.

58. A. R. Murphy, J. M. J. Fréchet, Chem. Rev. 2007, 107, 1066–1096.

59. C. Wang, H. Dong, W. Hu, Y. Liu, D. Zhu, Chem. Rev. 2011, 112, 2208–2267.

60. J. E. Anthony, Chem. Rev. 2006, 106, 5028–5048.

61. J. Zaumseil, H. Sirringhaus, Chem. Rev. 2007, 107, 1296–1323.

62. K. Takimiya, S. Shinamura, I. Osaka, E. Miyazaki, Adv. Mater. 2011, 23, 4347–4370.

63. H. W. Kroto, J. R. Heath, S. C. O’Brien, R. F. Curl, R. E. Smalley, Nature 1985, 318, 162–163.

64. S. Iijima, Nature 1991, 354, 56–58.

65. C. Bruno, I. Doubitski, M. Marcaccio, F. Paolucci, D. Paolucci, A. Zaopo, J. Am. Chem. Soc. 2003, 125, 15738–15739.

66. A. F. Hebard, M. J. Rosseinsky, R. C. Haddon, D. W. Murphy, S. H. Glarum, T. T. M. Palstra, A. P. Ramirez, A. R. Kortan, Nature 1991, 350, 600–601.

67. R. H. Baughman, A. A. Zakhidov, W. A. de Heer, Science 2002, 297, 787–792.

68. M. A. Petrukhina, L. T. Scott, Eds., Fragments of Fullerenes and Carbon Nanotubes: Designed Synthesis, Unusual Reactions, and Coordination Chemistry, John Wiley & Sons, Inc., Hoboken, 2012.

2

REDOX-MEDIATED REVERSIBLE σ-BOND FORMATION/CLEAVAGE

Takanori Suzuki¹, Hitomi Tamaoki¹, Jun-ichi Nishida², Hiroki Higuchi³, Tomohiro Iwai¹, Yusuke Ishigaki¹, Keisuke Hanada¹, Ryo Katoono¹, Hidetoshi Kawai⁴, Kenshu Fujiwara¹ and Takanori Fukushima⁵

¹ Department of Chemistry, Faculty of Science, Hokkaido University, Sapporo, Hokkaido, Japan

² Department of Materials Science and Chemistry, Graduate School of Engineering, University of Hyogo, Himeji, Hyogo, Japan

³ Institute for Materials Chemistry and Engineering, Kyushu University, Kasuga, Fukuoka, Japan

⁴ Department of Chemistry, Faculty of Science, Tokyo University of Science, Tokyo, Japan

⁵ Chemical Resources Laboratory, Tokyo Institute of Technology, Yokohama, Kanagawa, Japan

2.1 DYNAMIC REDOX ("DYREX") SYSTEMS

2.1.1 π-Electron Systems Exhibiting Drastic Structural Changes upon Electron Transfer

Redox reactions [1, 2] of an organic π-conjugated system induce a change in bond order. Each of the conjugated bonds undergoes characteristic change: the bond order of C¹═C² and C³═C⁴ bonds decreases whereas that of C²─C³ bond increases in 1,3-butadiene (Scheme 2.1). Such changes can be rationalized by considering decreased contribution of Ψ2 upon one-electron oxidation to the cation radical or increased contribution of Ψ3 upon one-electron reduction to the anion radical. In the π-systems suffering steric hindrance (e.g., tetra-substituted alkenes including overcrowded ethylenes [3] or sterically hindered quinodimethanes [4–6]), the changes in bond order induce characteristic structural changes because the effective mode to release the strain depends on the bond order of the certain bonds (Scheme 2.2). According to the X-ray analyses, the C═C bond of tetra(2-thienyl)ethene with four SMe groups (1) is planar whereas the corresponding dication 1²+ adopts a perpendicular conformation with an ultimate twisting angle of 90° [7] due to the decreased bond order of the ethene bond upon two-electron (2e)-oxidation. In the 9,10-anthraquinodimethane 2 [8] and 9,10-phenanthraquinodimethane 3 [9] with electron-donating aryl substituents on the exocyclic bonds (Ar = 4-MeOC6H4), drastic change in geometries were also exemplified since the bond order of the exocyclic C═C bonds was reduced upon 2e-oxidation to allow easy transformation of their molecular geometries from the bent/skewed shape in 2 and 3 into the twisted one in 2²+ and 3²+. Such geometrical changes were demonstrated by X-ray analyses on both of the neutral π-systems and their corresponding charged species isolated as stable salts. The isolation of anionic species was often hampered by facile protonation, so that, the X-ray analyses were often conducted on the cationic species as shown above.

SCHEME 2.1 A change in bond order upon one-electron oxidation/reduction in 1,3-butadiene system.

SCHEME 2.2 The π-type dyrex systems exhibiting dynamic structural changes upon electron transfer: (a) tetrathienylethene 1 with a planar double bond giving perpendicularly twisted dication 1²+; (b) 9,10-anthraquinodimethane 2 with a bent geometry giving twisted dication 2²+; (c) 9,10-phenanthraquinodimethane 3 with a skewed geometry giving twisted dication 3²+. All of the geometrically changes were verified by X-ray analyses.

In some cases, the structural changes after the first one-electron oxidation/reduction makes the second redox process proceed facilely, resulting in successive two-electron transfer nearly at the same potential. Moreover, there are cases where the electron-transfer of the ion radicals to the corresponding doubly charged diions proceeds more easily than the first oxidation/reduction process, thus minimizing the steady-state concentration of the intermediary ion radicals. By considering that the open-shell species are generally reactive to often cause side reactions [10], such a feature is favored to attain high reversibility of the redox interconversion.

Another characteristic is the separation of redox potentials of the neutral and the charged species since not only the geometrical structure but also the electronic structure are perturbed upon electron transfer. By adopting the twisted geometry in the doubly charged species, the on-site Coulombic repulsion could be reduced to stabilize the diions electronically. For example, the butterfly-shaped 9,10-anthraquinodimethane derivative 2 [8] undergoes 2e-oxidation at +0.44 V vs Fc/Fc+ [1a, 2] whereas the corresponding dication 2²+ with a twisted geometry exhibits 2e-reduction potential at −0.08 V [1a, 2]. Similarly, Eox of helically deformed 9,10-phenanthraquinodimethane derivative 3 [9a] is +0.40 V [1b, 2] (2e), which is far different from Ered of the twisted dication 3²+ (−0.10 V [1b, 2] (2e)). The observed separation of redox potentials for the reversibly interconvertible species endows the redox pairs with the electrochemical bistability, which is one of the required features in developing promising functionalized materials such as molecular switches or memories.

The "dynamic redox (dyrex) system is the name given to these systems that experience the dynamic changes in geometrical and electronic structures upon redox reactions as described above. Their characteristic features are facile 2e-transfer nearly at the same potentials and the electrochemical bistability. For the examples shown above, the redox-induced change in the bond order occurs in the π-conjugated systems, thus, they are classified as the π-type" dyrex systems.

2.1.2 Redox Switching of a σ-Bond upon Electron Transfer

The redox-induced change in the bond order occurs not only in the π-conjugated systems but in the system where the σ-bond formation/cleavage is accompanied by electron transfer (e.g., bond order change of 0 → 1 or 1 → 0). Plethora of research on the photoinduced electron transfer (PET) reactions afforded many diagnostic examples (Scheme 2.3). Thus, the one-electron oxidation of arylcyclopropanes induced the C─C σ-bond cleavage to give 1,3-propanediyl-type cation radicals [11], whereas dimerization of styrenes was also induced under the PET conditions by forming a new C─C σ-bond at their β-positions [12]. The latter reaction was especially favored when the two styrene units are connected by a spacer so that the bonding process being intramolecular process: for example, 2,5-diaryl-1,5-hexadienes to cyclohexane-1,4-diyl cation radicals [13].

SCHEME 2.3 (a) A PET reaction involving C─C bond cleavage in arylcyclopropane; (b) a PET reaction involving C─C bond formation in 2,5-diaryl-1,5-hexadiene.

In these PET reactions, formation/cleavage of the σ-bond was mostly evidenced by time-resolved spectroscopy since the generated ion radicals are highly reactive to undergo various follow-up reactions to give characteristic reaction products. However, by incorporating charge-stabilizing groups, the ion radicals/diions could be, if possibly, isolated as salts. So that, they can be involved in constructing the reversible dyrex systems. In terms of drastic structural and electronic changes, the dyrex systems involving redox-switching of a σ-bond is more attracting than the π-type dyrex systems shown in the previous section.

The historical prototypes of σ-type dyrex systems include 1,3-dimethylenecyclobutane 4 having two dihydropyridine chromophores on the exocyclic bonds, which underwent oxidative C─C bond formation to give bicyclo[1.1.0]butane-1,3-diyl dication 5²+ (Scheme 2.4a) [14]. The homoconjugation between the two chromophores would facilitate the transformation, which was also the case in their pyran analogues. Although reduction of tethered dipyridinium derivatives 6²+ were reported to give spiro-type compounds 7 by C─C bond formation (Scheme 2.4b) [15], dipyridinium 5²+ generated from 4 was transformed to the starting diene 4 upon 2e-reduction accompanied by cleavage of the newly formed C─C σ-bond.

SCHEME 2.4 (a) Redox interconversion between 1,3-dimethylenecyclobutane 4 and bicyclo[1.1.0] dication 5²+ upon electron transfer accompanied by reversible C─C bond formation/cleavage (σ-type dyrex system). (b) The C─C bond formation of dipyridinium 6²+ upon two-electron reduction to give spiro-type compound 7.

2.1.3 Two Types of Dyrex Systems Exhibiting Redox Switching of a σ-Bond

The dyrex behavior involving redox-switching of a σ-bond was also observed in cyclooctane-inserted dibenzo-TTF derivative 8a (Scheme 2.5a) [16]. As shown by the lower Eox value (−0.01 V [1a, 2] (2e) vs Fc/Fc+), 8a is a strong donor than the fully conjugated analogue (dibenzo-TTF) (Eox = +0.18 V) [1a, 2], showing the effective through-space interaction between the two 1,3-dithiol-2-ylidene chromophores at the 1,5-positions of cyclooctane. Due to the effective transannular interaction, 8a underwent facile C─C bond formation to give cis-bicyclo[3.3.0]octane-1,5-diyl dication 9a²+ (Ered = −0.14 V) [1a, 2] upon oxidation, the structure of which was unambiguously demonstrated by an X-ray analysis. The same dication 9a²+ was obtained accompanied by C─C bond cleavage upon oxidation of 10a (Eox = +0.47 V) [1a, 2], which was obtained by the photochemical valence isomerization of 8a. Upon reduction, dication 9a²+ exclusively underwent C─C bond cleavage to regenerate TTF-analogue 8a. Thus, the reversible dyrex behavior was observed between TTF-analogue 8a and bicyclic dication 9a²+. This pair can be classified as the exo-type dyrex system: the charges in the diion are located at the exocyclic carbons of the newly formed ring system after the C─C σ bond formation.

SCHEME 2.5 (a) Reversible redox interconversion between 8a and 9a²+ categorized as an exo-type dyrex system; (b) reversible redox interconversion between 10b and 9b²+ categorized as an endo-type dyrex system.

Interestingly, by changing the ring size of cycloalkane unit, the reversible dyrex behavior was observed between the tricyclic isomer and the bicyclic dication: for example, tricyclo[4.4.2.0¹,⁶]-type donor 10b and cis-bicyclo[4.4.0]-1,6-diyl type dication 9b²+ (Scheme 2.5b). Thus, the dication 9b²+ (Ered = −0.36 V) [1a, 2] was generated upon oxidation of both of tricyclic-donor 10b (Eox = +0.39 V) [1a, 2] or TTF-analogue 8b (Eox = +0.13 V) [1a, 2] with a cyclodecane ring, however, its reduction selectively generated tricyclic-donor 10b. In the reversible dyrex pair of 10b and 9b²+, the charges in the diion are located at the endocyclic carbons of the starting ring system which undergoes the C─C σ bond cleavage. This pair represents the behavior of the endo-type dyrex systems.

The two-types of dyrex systems (exo- and endo-systems) can be schematically presented as in Scheme 2.6. By incorporating a ring structure in the molecules, all of the bond formation/cleavage reactions can proceed as unimolecular processes although it is not the essential prerequisite for the dyrex behavior [17]. It is evident that the redox reaction into the corresponding diion is accompanied by the σ bond formation in the exo dyrex systems with leaving the charges on the exocyclic groups. In contrast, in the endo dyrex systems, the σ-bond cleavage is the key to transform into the ring-opened diions. When the strong chromophores, such as triarylmethylium dyes [18], are incorporated, the dyrex pair would exhibit vivid change in color upon interconversion [19], thus endowing the systems with the electrochromic function [20]. Due to the dynamic geometrical changes upon redox reactions, the chromism based on the dyrex systems could be accompanied by more advanced features as shown in the following sections [21, 22].

SCHEME 2.6 Two types of σ-type dyrex systems: (a) an exo-type system consisting of a dication with the positive charges located at the exocyclic carbons of the newly formed ring system; (b) an endo-type system consisting of a dication with the positive charges located at the endocyclic carbons of the starting ring system undergoing bond cleavage.

2.2 ADVANCED ELECTROCHROMIC RESPONSE OF ENDO-TYPE DYREX SYSTEMS EXHIBITING REDOX SWITCHING OF A σ-BOND

2.2.1 Tetraaryldihydrophenanthrenes as Prototypes of Endo-Dyrex Systems

9,9,10,10-Tetraphenyl-9,10-dihydrophenanthrene (Ph4DHP) has been known as a stable compound [23] in spite of the fact that it inherits severe steric repulsion around the C9─C10 bond characteristic to hexaphenylethane [24]. So that, the C9─C10 bond of Ar4DHP derivatives would be expanded, and thus the bond length would be much greater than the standard value [25–27]. The elongated bond has a smaller bond-dissociation energy [27, 28] and could be cleaved easily upon redox reactions [29], which had made us to study in detail on a series of Ar4DHPs as a promising class of compounds to establish the endo-type dyrex systems.

By attaching the electron-donating aryl groups, such as 4-Me2NC6H4 (a) or 4-MeOC6H4 (b), at the C9 and C10 positions, Ar4DHPs (11) underwent facile 2e-oxidation accompanied by C9─C10 bond cleavage [30]. At the same time, the resulting biphenyl-2,2′-diyl dications (12²+) were found to be stable enough for isolation. The electrochemical bistability of these pairs were confirmed by the large difference in redox potentials (+0.21 and −0.95 V for 11a/12a²+; +0.91 and −0.32 V for 11b/12b²+, respectively, vs Fc/Fc+) [1a, 2] (Fig. 2.1). Upon reduction of dications 12a,b²+, DHPs 11a,b were effectively regenerated accompanied by formation of a C─C bond. The interconversion of the endo-type dyrex pairs of 11/12²+ consists of several steps: the two-step electron transfer and the σ-bond formation/cleavage, for which the reaction sequence was proven as shown in Scheme 2.7 [31]. This scheme can account for the negligible steady-state concentration of intermediary open-shell species during the interconversion of 11/12²+.

FIGURE 2.1 Cyclic voltammogram (E/V vs SCE, scan rate 500 mV s−1) of 11a in CH2Cl2 containing 0.1 M Bu4NBF4 (Pt electrode). The reduction peak at −0.45 V was absent when the voltammogram was first scanned to the cathod.

SCHEME 2.7 Reaction sequence for the two-step electron-transfer and the C─C bond formation/cleavage in the representative endo-type dyrex system of 9,9,10,10-tetraaryl-9,10-dihydrophenanthrene 11/ biphenyl-2,2′-diylbis(diarylmethylium) 12²+.

2.2.2 Tricolor Electrochromism with Hysteretic Color Change in Non-c2-Symmetric Endo-Dyrex Pair

Another characteristic feature of the dyrex pairs of 11/12²+ is drastic difference in their UV-Vis spectra [30]. Ar4DHPs 11 exhibit absorptions only in the UV region whereas 12²+ have strong bands in the visible region as in triarylmethylium dyes, whose absorption maxima can be modified by changing the substituents on the aryl groups [12a²+ (Ar = Ar′ = 4-Me2NC6H4): λmax 661 nm (log ε 4.92), 604 (5.05); 12b²+ (Ar = Ar′ = 4-MeOC6H4): 539sh (4.72), 514 (4.78) in MeCN]. Thus, 11a/12a²+ and 11b/12b²+ pairs could demonstrate electrochromicity with vivid change in color between colorless-deep blue and colorless-deep red, respectively. When the spectral changes were followed by UV-Vis spectroscopy, several isosbestic points were observed because of negligible steady-state concentration of open-shell intermediates.

When the dication was attached with two different dye units as in 12c²+ (Ar = 4-Me2NC6H4; Ar′ = 4-MeOC6H4) of non-C2-symmetry [31], it exhibits violet color due to the presence of both blue [λmax 632 nm (log ε 4.93)] and red [519 (4.72)] chromophores. More striking feature is its redox behavior since the two chromophores in 12c²+ undergoes one-electron reduction at the different potentials. Because the C─C bond formation occurs only after two-electron reduction of 12²+ to 12²·, the first reduction process of 12c²+ (Ered = −0.98 V [1a, 2] vs Fc/Fc+) is reversible. Thus, 12c+· can be a long-lived species to be involved in the two-stage electrochromic behavior of tricolor changes: 12c²+ (violet) → 12c+· (blue) → 11c (colorless). On the other hand, the steady-state concentration of the intermediate 12c+· is negligible during the oxidation of 11c, since 12c+· (Eox = −0.43 V) [1a, 2] is more easily oxidized than 11c (Eox = +0.30 V) [1a, 2]. Thus, only bicolor change [11c (colorless) → 12c²+ (violet)] was observed upon oxidation (Fig. 2.2).

FIGURE 2.2 A continuous change in the UV-Vis spectrum of (a) 11c (3 × 10−5 M in CH2Cl2 containing 0.05 M Bu4NBF4) upon constant-current electrochemical oxidation (10 μA) at 5-min intervals, and 12c²+ (6 × 10−6 M in CH2Cl2 containing 0.05 M Bu4NBF4) upon constant-current electrochemical reduction (40 μA): (b) stage 1, at 2-min intervals; (c) stage 2, at 8-min intervals.

Tricolor electrochromicity based on discrete molecules is rare although there have been several successful examples based on the polymeric materials [32]. Even among the successful examples, a hysteretic pattern for the color change, where there is a difference between oxidation [color 1 → color 2] and reduction [color 2 → color 3 → color 1], is quite unique for the present dyrex systems.

2.2.3 Electrochromism with Chiroptical Output of Chiral Endo-Dyrex Pair

Each of 11 and 12²+ has a chiral element of helicity and axial chirality, respectively. So that, they exist as a pair of enantiomers (Scheme 2.8). They are configurationally unstable to exist as racemic mixtures due to rapid ring inversions in 11 or facile rotation around the biaryl axis in 12²+. However, when associated with a chiral host, such as cyclodextrin (CyD) [33], the diastereomeric pairs of the complex (e.g., (Rax)/(Sax)-12²+@CyD) could be configurationally biased to prefer one-handedness.

SCHEME 2.8 Configurationally unstable endo-type dyrex system of (M)-11/(Rax)-12²+ and (P)-11/(Sax)-12²+ undergoing facile inversion of helicity and axial chirality.

This is the case for complexation of 12a²+ with γ-CyD. The UV-Vis and circular dichroism (CD) spectra of 12a²+ (10−5 M−1) changed continuously upon gradual addition of γ-CyD (1–4 equiv.) in water (Fig. 2.3). The association constant was proven to be as large as 10⁶ M−1 [34]. Preference of (Rax)-12a²+ in 50% de at 25°C is the reason for giving optically active (M)-dihydrothiepin (13) upon reaction of 12a²+ salt with Na2S in the presence of γ-CyD (Scheme 2.9). Since 13 is configurationally more stable than 12a²+, diastereoselective complexation of 12a²+ with γ-CyD followed by chemical transformation into 13 can demonstrate supramolecular chirality transfer [35] on the dyrex component. Upon treatment with iodine, 13 (Eox = +0.19 V vs Fc/Fc+) [1b, 2] underwent oxidative desulfurization to regenerate 12a²+ and elemental sulfur [36]; thus, the above transformation can be conducted repeatedly.

FIGURE 2.3 Changes in the (a) UV-Vis and (b) CD spectra of 12a²+(BF4−)2 (1.0 × 10−5 M in H2O) upon addition of γ-CyD (1–4 equivs) at 25°C. Diastereomeric complexes of (Rax)-12a²+@γ-CyD and (Sax)-12a²+@γ-CyD are equilibrated over the minutes, and the CD spectra were measured 30 min after admixing.

SCHEME 2.9 Diasteroselective complexation of γ-CyD with (Rax)-12²+ and successive transformation into dihydrothiepin (M)-13, demonstrating successful transmission of supramolecular chirality to molecular chirality.

Besides the intermolecular chirality transmission in 12a²+/γ-CyD system, the sense of axial chirality of the dication was successfully biased to prefer one-handedness by attaching proper chiral auxiliary on the aryl groups [37]. Thus, a CH2Cl2 solution of dication (R,R,R,R)-12d²+ (Ar = Ar′ = (R)-sec-BuOC6H4; Ered = −0.39 V vs Fc/Fc+) [1a, 2] exhibits bisignated Cotton effects [λext 564 nm (Δε +31), 531 (−23)] in the circular dichroism (CD) spectrum, which is strong enough to be used as an output upon electrolysis of (R,R,R,R)-11d (Eox = +1.03 V) [1a, 2]. This is the successful example of electrochiroptical response systems [38–40], in which the electrochemical input is transduced not only into UV-Vis but also chiroptical outputs.

Since the corresponding monocation with two (R)-sec-BuO groups exhibits negligible CD signals [λext 506 nm (Δε −1.3) in CH2Cl2], the point chiralities in (R,R,R,R)-12d²+ must be intramolecularly transmitted [41] to the axial chirality through the π–π overlap of two cationic chromophores. Thus, very strong CD signals are produced in the dication by exciton coupling mechanism [42]. Comparisons of the X-ray geometries of two diastereomers show that the dication with Sax-configuration is less sterically hindered, and thus would be more populated in solution (Fig. 2.4). Based on the NMR analyses, the diastereomeric excess of (R,R,R,R,Sax)-12d²+ over (R,R,R,R,Rax)-12d²+ was determined to be only 5% in CH2Cl2, however, the de value is 50% in benzene [λext 569 nm (Δε +294), 531 (−244)]. Such solvent dependency of de [43, 44] can be accounted for by considering that stabilization by the π–π overlap is more important in the less polar solvent.

FIGURE 2.4 ORTEP drawings of (a) (R,R,R,R,Sax)-12d²+ and (b) (R,R,R,R,Rax)-12d²+ determined by the low-temperature X-ray analysis of (R,R,R,R)-12d²+(SbCl6−)2 salt. Both diastereomers coexist in the same crystal in a 1 : 1 ratio. The cationic chromphores are stacked in a face-to-face manner, and the steric replusion between the chiral auxiliaries is smaller in (Sax)-isomer. (c) Schematic drawing of the geometry for [4-(R)-sec-BuOC6H4]2C+ unit commonly observed for both diastereomers in (R,R,R,R)-12d²+(SbCl6−)2 crystal.

The electrochiroptical systems working under the intramolecular chirality transmission are attracting since the de value (i.e., chiroptical properties) can be modified by external stimuli such as solvent polarity, temperature, or pH. They are promising candidates in developing multi-input molecular response systems [45].

2.2.4 Multi-Output Response System Based on Electrochromic Endo-Dyrex Pair

Due to high sensitivity, fluorescence (FL) is an attractive output to be involved in the molecular response systems [46, 47]. Based on our previous studies on the dyrex systems exhibiting ON/OFF switching of fluorescence [48], 3,4-dihydro[5]helicene (dibenzo[c,g]phenanthrene) was selected as a fluorophore [49], and thus dyrex pair of 14/15²+ was designed by fusing benzene rings on the biphenyl unit in prototype 11/12²+. Due to their configurational stability, optically pure samples of (M)- or (P)-14a (Eox = +0.76 V vs Fc/Fc+) [1a, 2] with four 4-MeOC6H4 groups were isolated, which underwent reversible interconversion with dications (Rax)- or (Sax)-15a²+ (Ered = −0.30 V) [1a, 2], respectively, without loss of enantiopurity (Scheme 2.10).

SCHEME 2.10 Configurationally stable endo-type dyrex system of (M)-14/(Rax)-15²+ and (P)-14/(Sax)-15²+ undergoing no inversion of helicity and axial chirality.

As designed, 14a is fluorescent (λem 407 nm, ΦF = 0.18) but the corresponding dication is not. On the other hand, dication 15a²+ has a strong absorption [λmax 534 nm (log ε 4.83) in CH2Cl2] and Cotton effects [(Rax): λext 577 nm (Δε +147), 196 (−28.9) in CH2Cl2] in the visible region. Thus, the optically pure dyrex pair of 14a/15a²+ underwent electrochemical interconversion with vivid changes in three kinds of spectra (UV-Vis, CD, FL) [50]. The similar three-way-output response [51] was also observed in the derivatives of 14b-f/15b-f²+ having long alkoxy chains (nC8H17O or nC16H33O) [50] or two alkoxy chains on each of the aryl groups [52] with slight modification of the molecular properties (Fig. 2.5) [50]. In developing feature molecular devices, these multi-output response systems have an advantage to the simple electrochromic ones in terms of the error-free detection since the output signals can be verified through the multiple channels.

FIGURE 2.5 Continuous changes in the (a) UV-Vis, (b) CD, and (c) FL spectra of (M)-14b with four 4-nC8H17OC6H4 groups (2 × 10−5 M in CH2Cl2 containing 0.05 M Bu4NBF4) upon constant-current electrochemical oxidation (24 μA) to (Rax)-15b²+ at 20-min intervals.

2.3 ADVANCED ELECTROCHROMIC RESPONSE OF EXO-TYPE DYREX SYSTEMS EXHIBITING REDOX SWITCHING OF A σ-BOND

2.3.1 Bis(diarylethenyl)biphenyls as Prototypes of Exo-Dyrex Systems

In addition to the dimethylenecycloalkanes such as 4 (Section 2.1.2) and 8a (Section 2.1.3), α,ω-divinyl compounds can be promising candidates to perform as exo-type dyrex systems when the two vinylic chromophores are arranged in proximity to undergo facile ring closure by σ-bond formation upon electron transfer. The successful execution of this idea was demonstrated by biphenyl-2,2′-diyl-type compounds 16 attached with two electron donating 2,2-diarylethenyl groups (Scheme 2.11).

SCHEME 2.11 Representative exo-type dyrex system of 2,2′-bis(2,2-diarylethenyl)biphenyl 16 / trans-9,10-dihydrophenanthrene-9,10-diylbis(diarylmethylium) 17²+. The latter species is classified as a butane-1,4-diyl dication, which easily undergoes deprotonation into 1,3-butadiene (e.g., 3) unless special stabilizing effects are present.

The key issue in this scheme is kinetic stability of dications 17²+ with a butane-1,4-diyl dicationic structure [53]. The dications of this category have been known to undergo deprotonation easily to generate the corresponding 1,3-dienes, as well-exemplified by the synthetic schemes for vinylogous TTF derivatives from 2-methylene-1,3-dithioles (Scheme 2.12) [54]. In fact, 17b²+ with four 4-MeOC6H4 groups underwent deprotonation under the reaction conditions, resulting in formation of 9,10-phenanthraquinodimethane 3 (Section 2.1.1) [9]. On the other hand, the similar butane-1,4-diyl dication could be stabilized when attached with the stronger donating groups (e.g., 4-Me2NC6H4), allowing isolation of trans-17a²+ as a stable salt [17, 39]. Thus, interconversion between 16a (Eox = +0.03 V vs Fc/Fc+) [1a, 2] and 17a²+ (Ered = −0.79 V) [1a, 2] proceeded quantitatively with exhibiting electrochromism thanks to strong coloration of 17a²+ [λmax 588 nm (log ε 4.98) in MeCN] with two Michler’s Hydrol Blue chromophores.

SCHEME 2.12 Oxidative dimerization of 2-methylene-1,3-dithioles into TTF vinylogues via butane-1,4-diyl dication intermediates.

2.3.2 Electrochromism with Chiroptical Output of Chiral Exo-Dyrex Systems

As in the endo-dyrex pairs of 11 and 12²+, both components of exo-dyrex pairs of 16 (axial chirality) and 17²+ (helicity, two point chiralities) have the asymmetric elements. In the dyrex pairs having substituents on the biaryl moiety 18/19²+, optically pure samples of (Rax)- or (Sax)-18 (Eox = +0.11 V (a) [1a, 2]; +0.06 V (b) [1b, 2] vs Fc/Fc+, respectively) were isolated as configurationally stable molecules [39a, 55] which were transformed into dications 19²+ (Ered = −0.76 V (a) [1a, 2]; −0.78 V(b) [1b, 2], respectively) with (M,R,R)- or (P,S,S)-configuration, respectively (Scheme 2.13a). The 1,4-dicationic part in 19²+ was again stabilized by 4-Me2NC6H4 groups. The successful transmission of the axial chirality in 18 to the point chiralities in 19²+ could be realized due to conformational preference of the outward form in 18 over the corresponding inward form in terms of the orientation of bulky diarylethenyl groups to avoid steric repulsion (Scheme 2.13b). Due to the stereospecific transformation, not only UV-Vis but also CD spectral changes were observed with several isosbestic points during the electrolyses of optically pure samples, thus demonstrating their electrochiroptical response [37, 38, 40].

SCHEME 2.13 (a) Configurationally stable exo-type dyrex system of (Rax)-18/ (M,R,R)-19²+ and (Sax)-18/ (P,S,S)-19²+ undergoing no inversion of helicity and axial chirality. (b) Conformational preference for the outward form over the inward form in 18 resulting in the observed stereospecificity upon cyclization into 19²+.

2.3.3 Electrochromism of Exo-Dyrex Systems in Aqueous Media

Very low electrophilicity of the 1,4-dication moiety in 17a²+ and 19²+ stabilized by 4-Me2NC6H4 groups prompted further molecular design toward the water-soluble derivatives by attaching two TEG chains on each of the amino nitrogens, so that electrochromism can be realized in aqueous media (Scheme 2.14). Configurationally stable molecules (Rax)-20 [55] are especially interesting since they are potentially used as chiral dopants to change molecular alignment in lyotropic liquid crystals upon redox