Skip to content
Publicly Available Published by De Gruyter June 25, 2015

Discovery and development of photochromic diarylethenes

  • Masahiro Irie EMAIL logo

Abstract

A new family of thermally irreversible photochromic molecules, named “diarylethene,” was serendipitously discovered a quarter of a century ago. The diarylethenes undergo thermally irreversible, fatigue resistant, ultrafast and highly sensitive photochromic reactions. Many of diarylethenes can undergo photochromic reactions even in the single-crystalline phase. The outstanding photochromic performance attracts considerable attention from both scientific and technological points of view, and stimulates research from fundamental physico-chemical aspects to various applications. Here, it is reported how such thermally irreversible photochromic diarylethenes were discovered and developed.

Introduction

Light-induced reversible transformations of a chemical species between two isomers with different absorption spectra are referred to as photochromism, and molecules capable of these reactions are called photochromic molecules [1, 2]. The two isomers have different electronic as well as geometrical structures. In most cases the photogenerated isomers are thermally unstable and return to the initial isomers at ambient temperature. The thermally reversible reactivity is inconvenient for the use of the molecules in optical memory media, photoswitching devices or molecular machines. Although thermal irreversibility is an essential and indispensable property for photonics applications, molecules that undergo thermally irreversible photochromic reactions are limited within a few families of molecules. Therefore, it is strongly desired to develop thermally irreversible photochromic molecules and obtain a guiding principle how to design such thermally irreversible photochromic molecules. In this paper it is reported how such thermally irreversible photochromic diarylethenes were discovered and developed.

The instant property changes of photochromic molecules by photoirradiation without any additional processes lead to their use in various photonics devices [3–7]. The electronic structure changes can be applied to optical memory media and photoswitching devices, while the geometrical structure changes can be applied to light-driven actuators and others. For such applications photochromic molecules should fulfill the following requirements: (1) thermal irreversibility, (2) fatigue resistant character, (3) high sensitivity, (4) rapid isomerization rates, and (5) reactivity in solid state.

The thermal irreversible and fatigue resistant properties are two indispensable requirements. High sensitivity, rapid response and solid state reactivity are also necessary properties for the applications. A new class of photochromic molecules, which fulfills the five requirements simultaneously, is a “diarylethene” family [6, 7]. Diarylethenes are derivatives of stilbene. When the phenyl rings of stilbene are replaced with five-membered heterocyclic rings with low aromatic stabilization energy, such as thiophene or furan rings, both isomers become thermally stable and coloration/decoloration cycles can be repeated many times, more than 10 000 cycles. The response times of coloration/decoloration reactions of diarylethenes are less than 10 ps and many of diarylethenes can undergo the photoisomerization reactions even in the single-crystalline phase. The diarylethene family was serendipitously discovered as described in the following section.

Discovery of diarylethenes

Photochromic molecules not only change their colors but also induce various changes in physical and chemical properties, such as geometrical structure, dipole moment, refractive index, fluorescent property and oxidation/reduction potential. When photochromic molecules are incorporated into polymer backbones or side groups, photoirradiation brings about changes in properties of the photoresponsive polymers [8].

Azobenzene is a well-known photochromic molecule, which undergoes isomerization from the trans to the cis form upon irradiation with UV light and reverts back to the trans form upon heating or irradiation with visible light. During the isomerization the distance between 4 and 4′ carbons (the long axis of the molecule) decreases from 9.0 to 5.5 Å. When the azobenzene chromophores were incorporated into a polymer backbone, the solution viscosity was found to reversibly change upon alternate irradiation with UV and visible light [9, 10]. Before UV light irradiation the polymer has a rod-like extended conformation. The isomerization from the trans to the cis form upon UV light irradiation kinks the polymer chain, as shown in Fig. 1, resulting in a compact conformation and a decrease in the viscosity. The compact conformation returns to the initial extended conformation either thermally or by visible light irradiation, causing the viscosity to increase. Not only the viscosity but also other properties, such as conformation, pH, solubility and sol-gel transition temperature, were controlled upon photoirradiation. We carried out comprehensive studies on the polymers having azobenzene chromophores in the backbones or the side groups [9–14].

Fig. 1: 
          Schematic illustration of the photoinduced conformational change of a polymer having azobenzene units in the backbone.
Fig. 1:

Schematic illustration of the photoinduced conformational change of a polymer having azobenzene units in the backbone.

Just like as azobenzene, stilbene can also undergo the trans-cis photoisomerization reactions and the photogenerated cis isomer has a longer lifetime than the cis isomer of azobenzene. After the studies of azobenzene polymers, we decided to extend the research to polymers having stilbene units. A polymer having stilbene units in the backbone can be prepared by 1,4-addition radical polymerization of 2,3-diphenylbutadiene, which is synthesized from acetophenone, as shown in Fig. 2 [15]. Upon irradiation with 313 nm light the poly(2,3-diphenylbutadiene) underwent photocyclization reactions to produce yellow colored dihydrophenanthrene units in deaerated dichloromethane. The trans-cis photoisomerization was strongly suppressed because of the rigidity of the polymer chain. When the relaxation time of the backbone chain exceeds the lifetime of the excited state, the chain prevents the rotation to produce the cis form. The yellow dihydrophenanthrene units returned to initial colorless 2,3-diphenylbutene units in less than 10 min at room temperature. On the other hand, in the presence of air the dihydrophenanthrene units converted to phenanthrene units by hydrogen elimination and the reversibility was lost. To prevent the hydrogen elimination and provide the reversibility even under aerated conditions 2,3-dimesitylbutadiene was designed, as shown in Fig. 3 (I) (a). The synthesis of 2,3-dimesitylbutadiene was attempted by photoreduction of 2,4,6-trimethylacetophenone, as used for the synthesis of 2,3-diphenylbutadiene. But, the synthesis of pinacol failed because of the bulky size of the mesityl group [Fig. 3 (II) (a)]. To reduce the steric hindrance, the mesitylene was replaced with 2,5-dimethylthiophene, as shown in Fig. 3 (I) (b). Then, 2,3-di(2,5-dimethyl-3-thienyl)butadiene was successfully synthesized from 2,5-dimethyl-3-acetylthiophene. The butadiene was polymerized to poly(2,3-di(2,5-dimethyl-3-thienyl)butadiene) by 1,4-addition radical polymerization, as shown in Fig. 3 (II) (b).

Fig. 2: 
          Synthesis and photochemical and thermal reactions of poly(2,3-diphenylbutadiene).
Fig. 2:

Synthesis and photochemical and thermal reactions of poly(2,3-diphenylbutadiene).

Fig. 3: 
          (I) Synthesis of polymers having (a) 2,3-dimesitylbutene units and (b) 2,3-di(2,5-dimethyl-3-thienyl)butene units in the backbone. (II) (a) A synthetic route to prepare 2,3-dimesitylbutdiene. (b) Synthetic routes and photochemical reactions of poly(2,3-di(2,5-dimethyl-3-thienyl)butadiene) and poly(2,3-di(2,5-dimethyl-3-furyl)butadiene).
Fig. 3:

(I) Synthesis of polymers having (a) 2,3-dimesitylbutene units and (b) 2,3-di(2,5-dimethyl-3-thienyl)butene units in the backbone. (II) (a) A synthetic route to prepare 2,3-dimesitylbutdiene. (b) Synthetic routes and photochemical reactions of poly(2,3-di(2,5-dimethyl-3-thienyl)butadiene) and poly(2,3-di(2,5-dimethyl-3-furyl)butadiene).

The polymer having 2,3-dithienylbutene units was dissolved in benzene and the solution was irradiated with 313 nm light. The colorless solution turned yellow along with the formation of a cyclized closed-ring form, as shown in Fig. 3 (II) (b), and subsequently the yellow color disappeared upon irradiation with visible light. In contrast to poly(2,3-diphenylbutadiene) the yellow color due to the closed-ring isomer units was found to remain stable overnight in the dark. The yellow closed-ring units were stable even at 100 °C and returned to the initial colorless open-ring units upon irradiation with visible light. The dithienylethene unit in the polymer was unexpectedly found to undergo a thermally irreversible photochromic reaction. The amazing result led us to study the photochemistry of the monomer unit, 2,3-di(2,5-dimethyl-3-thienyl)butene and its derivatives in detail [16]. This is the course of the serendipitous discovery of photochromic diarylethenes.

Theoretical study

As described above, the 2,3-diphenylbutene unit undergoes the thermally reversible photochromic reaction in the deaerated solution, while the 2,3-di(2,5-dimethyl-3-thienyl)butene unit exhibits the thermally irreversible reactivity. The photogenerated closed-ring state is thermally stable. “Stable” means that the state has a very long lifetime and practically does not revert back at ambient temperature. The difference between the two chromophores is the aryl groups. To reveal the different reactivity of the two units semiempirical MNDO calculation was carried out for the diarylethene derivatives 14 [17].

According to the Woodward–Hoffmann rules [18] based on π-orbital symmetries for 1,3,5-hexatriene, which is the simplest molecular framework of diarylethenes, the cyclization reaction is allowed in a conrotatory mode in a photoexcited state. Figure 4 shows the state correlation diagrams for the electrocyclic reactions from 1a to 1b and 2a to 2b in the conrotatory mode. The two aryl groups were assumed to be in the anti-parallel orientation. Full lines of the figure show that the interconnecting states belong to the same symmetry groups. The diagrams indicate that the conrotatory cyclizations of 1a and 2a in the ground state are prohibited, because each S0 state of the open-ring form correlates with a highly excited state of the closed-ring form. On the other hand, no large energy barrier exists in the cyclization processes in the S1 state for 1a and S2 state for 2a. This indicates that the cyclization reactions in the excited states can proceed.

Fig. 4: 
          State correlation diagrams in a conrotatory mode for the reactions from 1a to 1b and from 2a to 2b.
Fig. 4:

State correlation diagrams in a conrotatory mode for the reactions from 1a to 1b and from 2a to 2b.

What should be discussed here is the stability of the closed-ring isomers. Figure 4 suggests that in both 1b and 2b the cycloreversion reactions in the ground state have to overcome energy barriers, and the barriers correlate with the ground state energy differences between the open- and closed-ring isomers. The calculated energy differences are shown in Table 1. When the energy difference is large, as in the case of molecule 1, the energy barrier becomes small and the cycloreversion reaction can take place readily. On the other hand, the energy barrier becomes large when the energy difference is small, as shown in molecule 2. In this case the cycloreversion reaction hardly takes place. The correlation between the ground state energy difference and the energy barrier is well explained by the Horiuti–Polanyi rule as shown in Fig. 5. The energy difference in the ground states between the open- and closed-ring isomers controls the stability of the photogenerated closed-ring isomers.

Fig. 5: 
          The correlation between the ground state energy difference and the energy barrier. When the energy difference is small, the barrier becomes large, while the barrier becomes small when the energy difference is large.
Fig. 5:

The correlation between the ground state energy difference and the energy barrier. When the energy difference is small, the barrier becomes large, while the barrier becomes small when the energy difference is large.

Table 1

Relative ground-state energy differences between the open- and closed-ring isomers.

Compound Conrotatory (kcal/mol)
1,2-diphenylethene (1) 27.3
1,2-di(3-pyrrolyl)ethene (3) 15.5
1,2-di(3-furyl)ethene (2) 9.2
1,2-di(3-thienyl)ethene (4) −3.3

The next question is what makes the difference in the ground state energy levels of the two isomers. First, we compared the strain energy of the photogenerated six-membered rings. The optimized geometries of the closed-ring isomers, 1b and 2b, however, showed almost identical six-membered ring structures and the ring strain could not explain the energy difference.

Next, we examined the aromaticity change from the open- to the closed-ring isomers. During the cyclization reaction phenyl and heterocyclic rings change the structures as shown in Fig. 6. The aromaticity of the rings is lost during the cyclization reactions. The energy differences between the right- and left-side groups were calculated and shown in Table 2. The aromatic stabilization energy of the aryl groups correlates well with the ground state energy difference. The highest energy difference was calculated for the phenyl group and the lowest one for the thienyl group. The aromaticity is the key molecular property which controls the thermal stability of the closed-ring isomers.

Fig. 6: 
          The structure changes of phenyl and five-membered heterocyclic rings along with the cyclization reactions.
Fig. 6:

The structure changes of phenyl and five-membered heterocyclic rings along with the cyclization reactions.

Table 2

Aromatic stabilization energy differences.

Group Energy (kcal/mol)
Phenyl 27.7
Pyrrol 13.8
Furyl 9.1
Thienyl 4.7

The theoretical prediction was confirmed by the synthesis of diarylethenes with various types of aryl groups, as shown in Fig. 7 [6, 7, 17, 19, 20]. When the aryl groups are thiophene, benzothiophene, thiazol, or oxazole, which have low aromatic stabilization energy, the closed-ring isomers are thermally stable (more than 12 h at 80 °C). On the other hand, photogenerated closed-ring isomers of diarylethenes with indole rings, which have intermediate aromatic stabilization energy, undergo thermally reversible photochromic reactions (the half lifetimes of the closed-ring isomers: ∼10 h at 80 °C). The closed-ring isomer of diarylethene 13b with phenyl rings returned back to the open-ring isomer in less than a few minutes even at 20 °C.

Fig. 7: 
          Thermal stability of diarylethene derivatives.
Fig. 7:

Thermal stability of diarylethene derivatives.

Inferring from the above theoretical and experimental results the guiding design principle for the synthesis of thermally irreversible diarylethenes is obtained as follows. The thermally irreversible photochromic diarylethenes can be prepared by employing aryl groups with low aromatic stabilization energy, as shown below.

Development of diarylethenes

Since the discovery of thermally irreversible photochromic diarylethenes in the middle of eighties we synthesized various types of diarylethenes to improve their photochromic performance, as shown in Fig. 8. In the figure the structures of the derivatives and the years when the derivatives were developed are shown.

Fig. 8: 
          Development of photochromic diarylethenes.
Fig. 8:

Development of photochromic diarylethenes.

Upon irradiation with UV light 2,3-di(2,5-dimethyl-3-thienyl)butene undergoes a trans-cis isomerization reaction in addition to a cyclization reaction. A cyclic bridge, such as maleic anhydride or maleimide, was introduced to prevent the trans-cis photoisomerization from the competing photocyclization reaction. Although the diarylethene derivatives with maleic anhydride or maleimide undergo photocyclization reactions in less polar solvents, the reactivity was strongly suppressed in polar solvents, such as methanol or acetonitrile. To provide photochromic reactivity even in polar solvents, the ethene bridges were replaced with perfluorocycloalkenes with four-, five- and six-membered rings [21]. 1,2-Bis(2-methyl-1-benzothiophen-3-yl)perfluorocycloalkenes undergo photochromic cyclization/cycloreversion reactions not only in less polar methylcyclohexane but also in polar methanol or acetonitrile. Among the three derivatives, five-membered 1,2-bis(2-methyl-1-benzothiophen-3-yl)perfluorocyclopentene showed the highest fatigue resistant property. Since then we mainly prepared the perfluorocyclopentene derivatives.

During the course of study of diarylethene derivatives having oligothiophene aryl groups, as shown in Fig. 9 [22], we serendipitously found that 1,2-bis(2,4-dimethyl-5-iodo-3-thienyl)perfluorocyclopentene 15a underwent a thermally irreversible photochromic reaction even in the crystalline phase [23].

Fig. 9: 
          A synthetic route to a diarylethene having oligothiophene aryl groups.
Fig. 9:

A synthetic route to a diarylethene having oligothiophene aryl groups.

Compounds that exhibit photochromic reactions in the crystalline phase are very rare. In general, microcrystalline powders of photochromic compounds do not show any color change upon photoirradiation. To date, several compounds, such as paracyclophanes, N-salicylideneanilines, aziridines, triarylimidazol dimers, diphenylmaleonitriles, 2-(2,4-dinitrobenzyl)pyridines and triazines, have been reported to undergo photochromism in the crystalline phase [24]. Most of the photogenerated colored states are, however, thermally unstable at room temperature. Under the above circumstances we did not expect crystalline photochromism of diarylethenes. When it was found that the powder of diarylethene 15a changes the color from colorless to red upon irradiation with UV light, which is daily used for thin-layer chromatography, we, at first, thought that iodine released by the action of UV light was the origin of the red color. To confirm the effect of iodine, we eliminated iodo-substituents from the molecule and examined the color change of diarylethene 14a. Even in the absence of iodo-substituents the diarylethene 14a exhibited the color change from colorless to red upon UV irradiation. Thus, we discovered crystalline photochromism of diarylethene derivatives. The single-crystalline photochromism of diarylethenes was undoubtedly confirmed by the measurement of dichroism [25, 26] and in situ X-ray crystallographic analysis [27, 28]. Figure 10 shows a list of typical photochromic diarylethenes, which exhibit single-crystalline photochromism [29]. The single-crystals can be used for light-driven molecular crystal actuators [30–33].

Fig. 10: 
          Diarylethene derivatives showing single-crystalline photochromism.
Fig. 10:

Diarylethene derivatives showing single-crystalline photochromism.

Recently we found sulfone derivatives of 1,2-bis(2-ethyl-6-phenyl-1-benzothiophen-3-yl)perfluorocyclopentene exhibited very strong fluorescence (fluorescence quantum yield ∼0.9) in the closed-ring isomer [34, 35]. The turn-on fluorescent diarylethenes can be potentially applied to super-resolution fluorescent microscopy, such as PALM (or STORM) and RESOLFT. We are now developing water soluble diarylethens [36], as shown in Fig. 8.

The outstanding performance of diarylethene derivatives offers great potential for advancing future photonics and also certain promise for applications to biological science and technologies.


Article note

Paper based on the acceptance lecture for the 2014 Porter Medal, presented at the XXVth IUPAC Symposium on Photochemistry, Bordeaux, France, July 13–18, 2014.



Corresponding author: Masahiro Irie, Research Center for Smart Molecules, Rikkyo University, Nishi-Ikebukuro 3-34-1, Toshima-ku, Tokyo 171-8501, Japan, e-mail:

Acknowledgments

The author is grateful for the support from the Japan Scientific and Technology Agency (CREST) and from the Ministry of Education, Culture, Sports, Science and Technology (Grant-in-Aid for Scientific Research in Priority Area “New Frontiers in Photochromism (471)” and Strategic Research Foundation at Private Universities). The author also wishes to express sincere thanks to colleagues and students for their extraordinary efforts to accomplish this research.

References

[1] H. Dürr, H. Bouas-Laurent. Pure Appl. Chem. 73, 639 (2001).Search in Google Scholar

[2] H. Dürr, H. Bouas-Laurent (Eds.). Photochromism: Molecules and Systems, Elsevier, Amsterdam (2003).Search in Google Scholar

[3] M. Irie (Ed.). Photo-reactive Materials for Ultrahigh Density Optical Memory, Elsevier, Amsterdam (1994).Search in Google Scholar

[4] J. C. Crano, R. J. Guglielmetti (Eds.). Organic Photochromic and Thermochromic Compounds, Plenum Press, New York (1998).Search in Google Scholar

[5] M. Irie, Y. Yokoyama, T. Seki (Eds.). New Frontiers in Photochromism, Springer, Tokyo (2013).10.1007/978-4-431-54291-9Search in Google Scholar

[6] M. Irie. Chem. Rev. 100, 1685 (2000).10.1021/cr980069dSearch in Google Scholar PubMed

[7] M. Irie, T. Fukaminato, K. Matsuda, S. Kobatake. Chem. Rev. 114, 12174 (2014).Search in Google Scholar

[8] M. Irie. Adv. Polym. Sci. 94, 27 (1990).Search in Google Scholar

[9] M. Irie, K. Hayashi. J. Macromol. Sci. Chem. A13, 511 (1979).10.1080/00222337908066609Search in Google Scholar

[10] M. Irie, K. Hirano, S. Hashimoto, K. Hayashi. Macromolecules14, 262 (1981).10.1021/ma50003a007Search in Google Scholar

[11] M. Irie, W. Schnabel. Macromolecules14, 1246 (1981).10.1021/ma50006a021Search in Google Scholar

[12] M. Irie, H. Tanaka. Macromolecules16, 210 (1983).10.1021/ma00236a011Search in Google Scholar

[13] M. Irie, W. Schnabel. Macromolecules18, 394 (1985).10.1021/ma00145a016Search in Google Scholar

[14] M. Irie, R. Iga. Macromolecules19, 2480 (1986).10.1021/ma00164a004Search in Google Scholar

[15] M. Irie, W. Schnabel. Eur. Polymer J. 18, 15 (1982).Search in Google Scholar

[16] M. Irie, M. Mohri. J. Org. Chem. 53, 803 (1988).Search in Google Scholar

[17] S. Nakamura, M. Irie. J. Org. Chem. 53, 6136 (1988).Search in Google Scholar

[18] R. B. Woodward, R. Hoffmann. The conservation of Orbital Symmetry, Verlag Chemie, Weinheim (1970).10.1016/B978-1-4832-3290-4.50006-4Search in Google Scholar

[19] K. Uchida, T. Matsuoka, K. Sayo, M. Iwamoto, S. Hayashi, M. Irie. Chem. Lett. 28, 835 (1999).Search in Google Scholar

[20] S. Takami, S. Kobatake, T. Kawai, M. Irie. Chem. Lett. 32, 892 (2003).Search in Google Scholar

[21] M. Hanazawa, R. Sumiya, Y. Horikawa, M. Irie. J. Chem. Soc. Chem. Commun. 206 (1992).10.1039/c39920000206Search in Google Scholar

[22] M. Irie, T. Eriguchi, T. Takada, K. Uchida. Tetrahedron53, 12263 (1997).10.1016/S0040-4020(97)00558-9Search in Google Scholar

[23] M. Irie, K. Uchida, T. Eriguchi, H. Tsuzuki, Chem. Lett. 24, 899 (1995).Search in Google Scholar

[24] M. Morimoto, M. Irie. Chem. Commun. 3895 (2005).10.1039/b505256dSearch in Google Scholar PubMed

[25] S. Kobatake, T. Yamada, K. Uchida, N. Kato, M. Irie. J. Am. Chem. Soc. 121, 2380 (1999).Search in Google Scholar

[26] S. Kobatake, M. Yamada, T. Yamada, M. Irie. J. Am. Chem. Soc.121, 8450 (1999).Search in Google Scholar

[27] T. Yamada, S. Kobatake, K. Muto, M. Irie. J. Am. Chem. Soc. 122, 1589 (2000).Search in Google Scholar

[28] T. Yamada, S. Kobatake, M. Irie. Bull. Chem. Soc. Jpn. 73, 2179 (2000).Search in Google Scholar

[29] S. Kobatake, M. Irie. Bull. Chem. Soc. Jpn. 77, 195 (2004).Search in Google Scholar

[30] S. Kobatake, S. Takami, H. Muto, T. Ishikawa, M. Irie. Nature446, 778 (2007).10.1038/nature05669Search in Google Scholar PubMed

[31] L. Kuroki, S. Takami, K. Yoza, M. Morimoto, M. Irie. Photochem. Photobiol. Sci. 9, 221 (2010).Search in Google Scholar

[32] M. Morimoto, M. Irie. J. Am. Chem. Soc. 132, 14172 (2010).Search in Google Scholar

[33] F. Terao, M. Morimoto, M. Irie. Ang. Chem. Int. Ed. 51, 901 (2012).Search in Google Scholar

[34] K. Uno, H. Niikura, M. Morimoto, Y. Ishibashi, H. Miyasaka, M. Irie. J. Am. Chem. Soc. 133, 13558 (2011).Search in Google Scholar

[35] Y. Takagi, T. Kunishi, T. Katayama, Y. Ishibashi, H. Miyasaka, M. Morimoto, M. Irie. Photochem. Photobiol. Sci. 11, 1661 (2012).Search in Google Scholar

[36] Y. Shoji, A. Yagi, M. Horiuchi, M. Morimoto, M. Irie. Isr. J. Chem. 53, 303 (2013).Search in Google Scholar

Published Online: 2015-06-25
Published in Print: 2015-07-01

©2015 IUPAC & De Gruyter

Downloaded on 1.5.2024 from https://www.degruyter.com/document/doi/10.1515/pac-2015-0208/html
Scroll to top button