Visible-light triggered photoswitching systems based on fluorescent azulenyl-substituted dithienylcyclopentenes

Eugenia Andreea Dragua, Adrian E. Ionb, Sergiu Shovac, Daniela Balad, Constantin Mihailciucd, Mariana Voicescue, Sorana Ionescud and Simona Nica*a
aInstitute of Organic Chemistry “C. D. Nenitzescu” of The Romanian Academy, Splaiul Independentei, 202B, 060023, Bucharest, Romania. E-mail: snica@cco.ro; Fax: +40 21 3121 601; Tel: +40 21 3167 900
bInorganic Chemistry Laboratory, Faculty of Chemistry, University of Bucharest, Str. Dumbrava Rosie, no. 23, 020464, Bucharest, Romania
c“Petru Poni” Institute of Macromolecular Chemistry of The Romanian Academy, Grigore-Ghica Voda Alley, no. 41A, 700847, Iasi, Romania
dPhysical Chemistry Department, Faculty of Chemistry, University of Bucharest, Regina Elisabeta, no. 4-12, 030018, Bucharest, Romania
e“Ilie Murgulescu” Institute of Physical Chemistry of The Romanian Academy, Splaiul Independentei 202, 060021, Bucharest, Romania

Received 22nd June 2015 , Accepted 17th July 2015

First published on 17th July 2015


Abstract

Novel azulenyl functionalized dithienylethenes were synthesized, and their photochromic reactivity, fluorescent and electrochemical properties were investigated. By tailoring the conventional dithienylcyclopentene chromophore with azulene groups, the photocyclization occurred with 405 nm wavelength light irradiation. These compounds exhibited relatively strong fluorescence at 524 nm when excited at 360 nm, which decreased along with the photocyclization. Furthermore, their electrochemical oxidation induced cyclization.


Molecular switches that can reversibly change their properties upon excitation with light or other external stimuli have recently received considerable attention due to their potential applications for molecular electronic devices such as memories and switches.1 Among the various types of organic photoswitches, diarylethenes bearing two thiophene or benzothiophene rings are the most promising candidates owing to their excellent thermal stability and remarkable fatigue resistance.2 Attempts to modulate the photochromic properties of diarylethenes have led to structural variations in both the heterocyclic rings and the unsaturated bridge between.3 So far, the most commonly used diarylethenes are the dithienylperfluorocyclopentenes which exhibit good photochromic behaviour and high thermal stability during photocyclization.4 However, the expensive and rather volatile octafluorocyclopentene starting material is a major disadvantage. Furthermore, Feringa et al. reported that the photochromic behavior of perhydro- and perfluorocyclopentene switches is remarkably similar.3a Therefore, diheteroarylethenes with a cyclopentene bridge possess attractive photochromic properties and can be considered as promising alternative to diarylperfluorocyclopentenes. Although the field of molecular switches is rapidly growing,5 several key challenges remain to be addressed, such as the development of photochromic systems that can be triggered by irradiation using visible and/or near-IR light. In this context, substitution of the dithienylethene photochromic system with various chromophores may represent a new way of tuning the optical properties of molecular switches.6 The unique electronic and optical characteristics of azulene, prompted us to develop novel azulene-substituted dithienylcyclopentene switches. The design of these systems combines the photoswitching ability of the photochromic dithienylcyclopentene unit with azulene property to absorb in the visible region in order to access visible light-triggered cyclization. We report herein, the synthesis and photochromic behaviour of novel azulenyl functionalized dithienylethenes 4 (Scheme 1). Furthermore, because of azulene “anomalous” S2 → S0 fluorescence emission7 and interesting electrochemical properties,8 we also investigated the feasibility of fluorescence and electrochemical switching of these compounds.
image file: c5ra11974j-s1.tif
Scheme 1 Synthesis of azulenyl functionalized dithienylcyclopentenes 4.

As shown in Scheme 1, the target compounds 4a and 4b were synthesized in 24% and 43% yields, respectively, by Suzuki–Miyaura coupling reaction. The boronate ester 2, which was prepared through the lithiation and borylation of dichloride 1, was coupled with the appropriate azulenyl bromides using 3 mol% Pd(PPh3)4 as catalyst and aqueous Na2CO3 as base.

This synthetic protocol could not, however, be used for the preparation of dithienylethene 4c because the 1-bromoazulene precursor cannot be obtained separately by azulene bromination, but rather as an inseparable mixture with 1,3-dibromoazulene. Nevertheless, dithienylethene 4c was prepared in 20% yield by decarbonylation of 4b with pyrrole in acetic acid media (Scheme 1).9

The photoswitching behavior of compounds 4 was monitored by UV-Vis spectroscopy and the results are illustrated in Fig. 1. Compounds 4 showed absorption changes typical of photochromic diarylethene derivatives. Irradiation of dichloromethane solutions of the ring-open isomers 4(o) with 405 nm wavelength light triggered the photocyclization reaction and generated the ring-closed isomers 4(c) (Fig. 1). The visual demonstration of this photoreaction was the change in colour of the solutions from yellow-greenish to purple-bluish due to the formation of the extended π-conjugated backbone in the ring-closed isomers. The absorption maxima and molar extinction coefficients of both the ring-open and ring-closed isomers are summarized in Table S1 (ESI).


image file: c5ra11974j-f1.tif
Fig. 1 Absorption spectral changes of dichloromethane solutions of (a) 4a (1.55 × 10−5 M), (b) 4b (1.16 × 10−5 M) and (c) 4c (2.74 × 10−5 M) at different irradiation time.

In the open state, the absorption spectra of compounds 4 showed intense bands in the UV domain at wavelengths below 310 nm. Tailoring the thiophene rings with azulene moieties resulted in an additional band in the range between 380 and 420 nm, which is significantly red-shifted as compared to those of related dithienylcyclopentene derivatives.2,3 Photocyclization to the ring-closed isomers was demonstrated by the appearance of a new absorption band in the visible region located above 520 nm, accompanied by a significant decrease in the absorbances at shorter wavelengths. The photostationary state was reached within several minutes with the photocyclization quantum yields varying from 0.45 in the case of 4a to 0.39 for 4b and 0.40 for 4c, respectively (Table S1, ESI). Upon exposure to visible light for around two hours, using a normal bulb (tungsten, 100 W, 1340 lumen, 230 V), the initial UV-Vis spectra were recovered for all three compounds, showing the reversibility of the photocyclization reaction. In the case of compound 4a, 1H NMR spectra before and after irradiation provide further evidence for the proposed photocyclization reaction. The thiophene protons resonate at 6.94 ppm in the starting 4a(o), while upon irradiation a decrease of this signal accompanied by new resonance at 5.97 ppm was observed, in accordance with the formation of 4a(c) isomer. The ring-closed isomers of compounds 4a and 4b are stable at room temperature in dichloromethane solution for more than 10 h when kept in the dark, whereas in the case of compound 4c(c), a considerable decrease of the visible absorption maxima was observed within 30 minutes. Preliminary investigations showed that compounds 4a(c) and 4b(c) are also thermally stable at 50 °C in chloroform solutions for short periods of time, the half-life of these compounds being 8.5 min and 7.4 min, respectively. These results show that adequate substitution of the azulenyl moieties can be used to tune the thermal stability of this type of photoswitching compounds. The photochemical ring-opening/-closing experiments were performed with compound 4a which performed well five cycles, but further photoswitching for ten cycles lead to a decrease of the monitored absorbances to around 30% (Fig. S7, ESI).

In the crystalline form, 4a consists of two symmetrically related thiophene rings in antiparallel conformation (Fig. 2). The distance between the active carbons is 3.453(7) Å, significantly shorter than 4.2 Å. As reported earlier,10 this condition along with the antiparallel configuration of the methyl groups are crucial for the photochromic properties of such materials. The molecule as a whole is essentially non-planar, the dihedral angle between two symmetrically related thiophene and bromo-azulene rings are equal to 51.88(8)° and 30.08(8)°, respectively. An interesting feature of the crystalline structure is the one-dimensional supramolecular aggregates formed through π–π stacking interaction of 3.486 Å between centro-symmetrically related bromo-azulene rings (Fig. S8, ESI).


image file: c5ra11974j-f2.tif
Fig. 2 The molecular structure of compound 4a along with the atomic labelling scheme. Only the major component of disordered cyclopentene ring is shown. Symmetry code (i) 1 − x, y, 3/2 − z.

Next, we evaluated whether fluorescence switching of compounds 4 could be achieved by photocyclization. The ring-open isomers 4(o) exhibit strong fluorescence at 524 nm, due to characteristic S2 → S0 fluorescence emission of azulene when excited at 360 nm. The fluorescence quantum yields of 4a–c were determined as 0.21, 0.20 and 0.17, respectively, using quinine bisulfate (0.1 N solution in sulfuric acid as a reference)11 (Table S1, ESI). Thus, it is clear that the substituent of the azulenyl moiety has a significant effect on the emission intensity, but does not affect the position of the emission peak. Upon irradiation with 405 nm wavelength light, different degrees of fluorescence quenching were observed. The emission intensity of 4b and 4c in the photostationary state was quenched to ca. 38% and 20% respectively, whereas in the case of bromo-derivative 4a, nearly total quenching of fluorescence emission took place (Fig. 3). These results suggest that the photochromic fluorescence switching of azulenyl-functionalized dithienylcyclopentenes of type 4 can be finely-controlled by varying the substituents attached to the azulene rings. Recently, the reversible fluorescence switching of diarylethenes has attracted considerable attention, as these compounds might have potential applications in the design of erasable ultra-high density optical memory media.12 The ring-open isomer is the “ON” state and the ring-closed isomer is the “OFF” state. If the ratio of the two isomers is not influenced by the excitation wavelength, the fluorescence readout method is non-destructive. In the case of compound 4a, simple irradiation with 405 nm wavelength caused photocyclization which, in turn, quenched the fluorescence. Because neither cyclization nor cycloreversion are induced by excitation with 360 nm light, 4a can be potentially used for non-destructive recording.


image file: c5ra11974j-f3.tif
Fig. 3 Fluorescence emission spectra (λexc = 360 nm) of compounds 4a (a), 4b (b), and 4c (c) in CH2Cl2 (1.65 × 10−6 M) in the open form (solid line) and at the photostationary state under irradiation with 405 nm wavelength light (dashed line).

Finally, the possibility of electrochemical switching between the open and closed states of compounds 4 was examined by cyclic voltammetry. All compounds showed similar cyclic voltammogram profiles with oxidative ring-closure of the dithienylethene group above 1.7 V. For each compound, in the second cycle, an additional anodic peak appeared corresponding to the ring-closed isomer. For the ring-open isomer 4a(o), the cyclic voltammograms (Fig. 4) shows three anodic (0.76, 1.21 and 1.77 V) and two cathodic peaks (0.09 and 0.83 V) (Table S2, ESI) in the first cycle. In the second one, additional anodic peak appeared at ca. 0.20 V corresponding to the ring-closed isomer 4a(c). By comparison with the redox potentials of 1,3-dibromoazulene (Table S2, ESI), the first two anodic peaks were attributed to the sequential oxidation of the two azulenyl moieties bonded to the thiophene rings in 4a(o), whilst the peak at 1.77 V was tentatively assigned to the two-electron oxidative ring-closure of the dithienylethene group. This latter assignment is supported by previous reports13 and, furthermore, by the fact that when the oxidation potential was restricted to +1.0 and +1.4 V, respectively, the anodic peak at +0.2 V became negligible. The cyclic voltammetry results also suggest that the ring-closure occurs after the two-electron oxidation of the open form, that is, from 4a(o)2+ to 4a(c)2+. The electrochemical ring-closure of 4a(o) was further confirmed by spectroelectrochemistry (Fig. S10, ESI). Oxidation at 1.9 V (i.e., at a potential more anodic than the ring-closing peak at 1.77 V) led to an increase in the intensity of the absorption band at 633 nm, which is indicative of the formation of the ring-closed isomer. In the case of compounds 4b and 4c (Fig. S9, ESI), the oxidative ring-closure of the dithienylethene group occurred around 1.8 V. In the second cycle, additional anodic peaks were observed at 0.30 and 0.18 V (Table S2, ESI, Entries 6 and 8) corresponding to the ring-closed isomers 4b(c) and 4c(c), respectively.


image file: c5ra11974j-f4.tif
Fig. 4 Cyclic voltammetry of 4a in dichloromethane (10−3 M) at 100 mV s−1 in the 0.5–2.0 V potential range.

The electrochemical data can be qualitatively explained by the quantum chemical calculations. The relative stability of 4a(o) and 4a(c) changes in different oxidation states (Table S4, ESI). The neutral 4a(o) is 10.48 kcal mol−1 lower in energy than 4a(c). 4a(o) needs an energy of 5.10 eV to ionise and its monocation becomes less stable than 4a(c)+ by 5.69 kcal mol−1. The barrier to ring-closing is nonetheless still high (20.21 kcal mol−1) and the process highly improbable. With an additional 5.71 eV the second ionization takes place and not only does the dication 4a(o)2+ become 22.66 kcal mol−1 higher in energy than 4a(c)2+, but the barrier drops to 2.60 kcal mol−1. The ring-closing can take place, 4a(c)2+ forms and the ring-opening process has a low probability, due to the high energy barrier of 25.26 kcal mol−1, so that the equilibrium is shifted towards the dicationic 4a(c)2+.The oxidation energy of 4a(c) is 0.70 eV lower than that of 4a(o), meaning that it will oxidize earlier, as seen from the signal at 0.2 V in the voltammogram. The calculated difference between the first and second ionization steps for 4a(o) is 0.61 eV vs. 0.45 V in the voltammogram (Epa3Epa2), whereas the difference between the third and second ionization steps is 0.39 eV vs. the experimental value of 0.56 V (Epa4Epa3). The correlation holds also for compound 4c, where the calculated differences between the oxidation potential of 4c(c) and 4c(o) is 0.70 eV (Table S4, ESI). The relative energies and activation barriers for compounds 4a and 4c present less than 4 kcal mol−1 difference when compared with phenyl substituted dithienylcyclopentene.14 Therefore, substitution with azulene does not greatly change the relative stability of the species or the barriers.

Conclusions

In summary, we have synthesized novel azulene-substituted dithienylethenes 4 for which the photocyclization reaction proceed by irradiation with visible light. These compounds function also as fluorescent switches, exhibiting strong fluorescence in the ring-open state. According to the azulenyl substituent, different degree of fluorescent quenching occurs upon photocyclization. Furthermore, electrochemical investigations, correlated with quantum chemical calculations, have demonstrated that oxidative ring-closure of 4a occurs via dicationic species 4a(c)2+ resulting from the cyclization of 4a(o)2+. Further investigations regarding the influence of the azulenyl substitution on the photochromic properties of this novel system are underway.

Acknowledgements

The financial support of the National Authority for Scientific Research, Romania, grant Capacities 84/2007 is gratefully acknowledged for the HPC infrastructure.

Notes and references

  1. (a) A. C. Whalley, M. L. Steigerwald, X. Guo and C. Nuckolls, J. Am. Chem. Soc., 2007, 129, 12590 CrossRef CAS PubMed; (b) S. Kobatake, S. Takamis, H. Muto, T. Ishikawa and M. Irie, Nature, 2007, 446, 778 CrossRef CAS PubMed; (c) V. I. Minkin, Chem. Rev., 2003, 103, 2751 Search PubMed; (d) H. Tian and S. Yang, Chem. Soc. Rev., 2004, 33, 85 RSC; (e) M. Irie, Chem. Rev., 2000, 100, 1685 CrossRef CAS PubMed.
  2. (a) K. Matsuda and M. Irie, J. Photochem. Photobiol., C, 2004, 5, 169 CrossRef CAS; (b) S. Nakamura, S. Yokojima, K. Uchida, T. Tsujioka, A. Goldberg, A. Murakami, K. Shinoda, M. Mikami, T. Kobayashi, S. Kobatake, K. Matsuda and M. Irie, J. Photochem. Photobiol., A, 2008, 200, 10 CrossRef CAS PubMed.
  3. (a) L. N. Lucas, J. J. D. de Jong, J. H. van Esch, R. M. Kellog and B. L. Feringa, Eur. J. Org. Chem., 2003, 155 CrossRef CAS; (b) J. Cusido, E. Deniz and F. M. Raymo, Eur. J. Org. Chem., 2009, 2031 CrossRef PubMed; (c) T. Fukaminato, J. Photochem. Photobiol., C, 2011, 12, 177 CrossRef CAS PubMed; (d) H. Liu and Y. Chen, New J. Chem., 2012, 36, 2223 RSC; (e) K. Suzuki, T. Ubukata and Y. Yokoyama, Chem. Commun., 2012, 48, 765 RSC.
  4. (a) H. Cahova and A. Jaschkle, Angew. Chem., Int. Ed., 2013, 52, 3186 CrossRef CAS PubMed; (b) K. Kitagawa, H. Nishi and S. Kobatake, Angew. Chem., Int. Ed., 2013, 52, 9320 CrossRef PubMed; (c) P. Raster, A. Schmidt, M. Rambow, N. Kuzmanovic, B. König and G. Hilt, Chem. Commun., 2014, 50, 1864 RSC; (d) J. I. Kitai, T. Kobayashi, W. Uchida, M. Hatakeyama, S. Yokojima, S. Nakamura and K. Uchida, J. Org. Chem., 2012, 77, 3270 CrossRef CAS PubMed.
  5. S. Helmy, F. A. Leibfarth, S. Oh, J. E. Poelma, C. J. Hawker and J. R. de Alaniz, J. Am. Chem. Soc., 2014, 136, 8169 CrossRef CAS PubMed.
  6. S. Nakamura and M. Irie, J. Org. Chem., 1988, 53, 6136 CrossRef CAS.
  7. (a) B. D. Wagner, D. Tittelbach-Helmrich and R. P. Steer, J. Phys. Chem., 1992, 96, 7904 CrossRef CAS; (b) E. Amir, R. J. Amir, L. M. Campos and C. J. Hawker, J. Am. Chem. Soc., 2011, 133, 100486 CrossRef PubMed.
  8. (a) X. H. Zhang, C. Li, W. B. Wang, X. X. Cheng, X. S. Wang and B. W. Zhang, J. Mater. Chem., 2007, 17, 642 RSC; (b) S. Ito and N. Morita, Eur. J. Org. Chem., 2009, 27, 4567 CrossRef PubMed; (c) E. A. Dragu, S. Nica, M. Raicopol, A. Baran, D. F. Anghel, B. Cojocaru, L. Tarko and A. C. Razus, Tetrahedron Lett., 2012, 53, 2611 CrossRef CAS PubMed.
  9. S. Ito, N. Morita and T. Asao, Bull. Chem. Soc. Jpn., 1999, 72, 2543 CrossRef CAS.
  10. (a) S. Kobatake, Y. Matsumoto and M. Irie, Angew. Chem., Int. Ed., 2005, 44, 2148 CrossRef CAS PubMed; (b) S. Yamamoto, K. Matsuda and M. Irie, Angew. Chem., Int. Ed., 2003, 42, 1636 CrossRef CAS PubMed.
  11. W. H. Melhuish, J. Phys. Chem., 1961, 65, 229 CrossRef CAS.
  12. (a) Y. Kim, H. Jung, Y. H. Choe, C. Lee, S. H. Ko, S. Kouns, Y. Choi, B. H. Chung, B. C. Park, T. L. Huh, I. Shin and E. Kim, Angew. Chem., Int. Ed., 2012, 51, 2878 CrossRef CAS PubMed; (b) H. Jung, S. You, C. Lee, S. You and Y. Kim, Chem. Commun., 2013, 49, 7528 RSC.
  13. (a) W. R. Browne, J. J. D. de Jong, T. Kudernac, M. Walko, L. N. Lucas, K. Uchida, J. H. van Esch and B. L. Feringa, Chem.–Eur. J., 2005, 11, 6414 CrossRef CAS PubMed; (b) W. R. Browne, J. J. D. de Jong, T. Kudernac, M. Walko, L. N. Lucas, K. Uchida, J. H. van Esch and B. L. Feringa, Chem.–Eur. J., 2005, 11, 6430 CrossRef CAS PubMed.
  14. A. Staykov, J. Areephong, W. R. Browne, B. L. Feringa and K. Yoshizawa, ACS Nano, 2011, 5, 1165 CrossRef CAS PubMed.

Footnotes

Electronic supplementary information (ESI) available: Synthesis, characterization, spectroelectrochemistry and quantum calculations are available. CCDC 983049. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c5ra11974j
Crystal data for compound 4a: C35H26Br2S2, Mr = 670.50 g mol−1, size 0.45 × 0.10 × 0.01 mm3, monoclinic, space group C2/c, a = 24.2712(7) Å, b = 8.3649(3) Å, c = 14.4048(6) Å, β = 93.305(3)°, V = 2919.68(19) Å3, Z = 4, ρcalcd = 1.525 g cm−3, μ(MoKα) = 2.943 mm−1, F(000) = 1352, 8962 reflections in h(−29/29), k(−10/10), l(−17/17), measured in the range 5.16 ≤ Θ ≤ 51.98, T = 200 K, completeness Θmax = 99.91%, 2871 independent reflections, Rint = 0.0413, 178 parameters, 0 restraints, R1obs = 0.0438, wR2obs = 0.0936, R1all = 0.0655, wR2all = 0.1019, GoF = 1.036, largest difference peak and hole: 0.55/−0.34 e A−3.

This journal is © The Royal Society of Chemistry 2015
Click here to see how this site uses Cookies. View our privacy policy here.