Azophotoswitches containing thiazole, isothiazole, thiadiazole, and isothiadiazole

Abstract

We report a novel class of azophotoswitches incorporating various five-membered heteroaryl units such as thiazole, isothiazole, thiadiazole, and isothiadiazole. These azophotoswitches were developed through an initial screening of 24 compounds using DFT calculations to identify those with the wavelength of maximum absorption (λ_max) at a long wavelength. Subsequently, eight selected azophotoswitches were synthesized. Compounds containing both thiazole and isothiazole moieties showed relatively long λ_max compared to the other synthesized compounds. These azophotoswitches exhibited reversible isomerization under visible light irradiation at 430 nm, 450 nm, 470 nm (trans to cis) and 525 nm (cis to trans). Analysis of the X-ray crystal structures of the cis isomer of phenylazo[1,3,4-thiadiazole] exhibited a unique orthogonal geometry.

---

Introduction

Photoswitches are molecules that exist in two isomers, which can be interconverted reversibly by light irradiation. Many kinds of photoswitches have been developed and used for applications in materials and biological sciences.^1,2 In the case of a well-studied azobenzene photoswitch, the wavelength of maximum absorption (λ_max) is at ∼320 nm and undergoes trans to cis photoisomerization under 365 nm UV light irradiation. The cis to trans reverse isomerization occurs thermally or photochemically. The λ_max of azobenzene can be changed to a longer wavelength by introducing substituents at its phenyl ring, however, with a compensation of the thermal stability (half-life) of the cis isomer.^3–5

Recently, azophotoswitches containing “heteroaryl” units are increasingly gaining attention due to their unique light absorption and photoisomerization properties.⁶ In particular, photoswitches having five-membered “heteroaryl” units showed quite different photophysical and structural properties.⁷ Previously we and others reported photoswitches that contain five-membered “heteroaryl” units such as pyrrole, pyrazole, imidazole, isoxazole, triazole, thiazole and thiophene.^8–18 Each of these photoswitches has shown photophysical properties mostly affected by the type of “heteroaryl” unit present. For instance, a phenylazopyrrole derivative requires UV light for trans–cis isomerization, but the half-life of its cis isomer was exceptionally long.^19,20 In our recent case, phenylazothiazole showed a λ_max of 364 nm, 44 nm red-shifted compared to the λ_max of azobenzene and can isomerize under visible light (405 nm) irradiation with a moderate decrease in the half-life of its cis isomer (2.8 h in acetonitrile at 25 °C). Based on this finding, in the current study, we focused on the design of novel photoswitches that contain unexplored 5-membered heteroaryl units like oxazole, isothiazole, thiadiazole and isothiadiazole. Many pharmacologically active compounds contain one or more units of five-membered heteroaryl groups^21–25 and their photoisomerizable forms can be useful for application in photopharmacology. In this study, we used density functional theory (DFT) calculations to predict the λ_max of photoswitchable molecules, and our findings show a close match between the calculated data and experimental results. We found that the incorporation of both thiazole and isothiazole groups in the photoswitches results in a shift of λ_max to a longer wavelength than those of the other compounds studied. Furthermore, X-ray crystal analysis shows that the cis isomer geometry of phenylazo[1,3,4-thiadiazole] adopts a unique orthogonal configuration, supported by the theoretical calculation.

Results and discussion

Synthesis

We designed 24 photoswitchable molecules that have S, N or O heteroatoms in the five-membered aromatic ring and conducted DFT calculations to gain insight into their light absorption (λ_max) properties (Table S1†). Based on the calculated λ_max values, we selected eight compounds for synthesis, including those photoswitches with the longest and third longest λ_max (Table S1, entries k and x†). We used a general one-step procedure for the synthesis of compounds 1–8 (Scheme 1). Compounds 1–3 were synthesized via the Mills reaction using the corresponding 2-amino heteroarene derivatives and nitrosobenzene (Scheme S1†). Compounds 4–7 were synthesized by homo-oxidation of the corresponding 2-amino heteroarene derivatives (Scheme S2†). Although azobisthiazole showed the second longest λ_max by calculation (440 nm, Table S1, entry j†), the oxidation of 2-amino thiazole did not give the target compound, probably due to the undesired oxidation at the S atom of thiazole. However, a methyl substituent at position 5 of the amino thiazole gave azobis(methylthiazole) (compound 6) in 18% yield. This compound can also be synthesized by a multiple-step reaction as described in a recent thesis report.²⁶ The asymmetric heteroaryl azo compound 8 was synthesized by oxidizing 2-amino-3-methylthiazole and 5-amino-3-methylisothiazole in 7.3% yield along with the corresponding symmetric heteroaryl azo compounds (Scheme S2†). The NMR, mass and X-ray analyses confirmed the structures of the target compounds 1–8 (Fig. S1–S24† and Fig. 3).

Scheme 1 Synthesis of photoswitches from amino-heteroaryl derivatives via the Mills reaction with nitrosobenzene (1–3) or oxidation (4–8).

Photophysical studies

We then studied the absorption maximum of the 1–8 photoswitches (Fig. 1 and 2a). As estimated from the DFT-calculated λ_max values, the experimental λ_max of compounds 1–8 varied depending on the type of aryl moiety. The absorption spectra of these photoswitches exhibited a strong absorption band (λ_max = 332–429 nm) assignable to π–π* electron transition bands. Fig. 2b demonstrates the correlation between the calculated and experimental λ_max values. Compounds 1–8 exhibited a clear positive correlation between the calculated and experimental values, with the smallest calculated λ_max value corresponding to the smallest experimental λ_max value, and similarly for the largest values. This method showed a consistent trend across the compounds, demonstrating its reliability in theoretical predictions.

Fig. 1 (a–h) UV-visible absorption spectra of 1–8 in acetonitrile at 25 °C before irradiation (BI, black lines) and in the cis-rich PSS under 365, 405 or 430 nm irradiation (365_PSS, 405_PSS, 430_PSS; red lines) and in the trans-rich PSS under 430, 470, 525 or 568 nm irradiation (430_pss, 470_PSS, 525_PSS, 568_PSS; blue lines). In (h), intermediate PSSs at 450 and 470 nm are added (450_PSS and 470_PSS; green and yellow lines, respectively).

Fig. 2 (a) Selected photophysical parameters of photoswitches 1–8. (b) Relationship between the experimental and calculated λ_max values. A grey shadow indicates a positive correlation between them.

Upon light irradiation at suitable wavelengths (365, 405, 430, 450 or 470 nm), a change in the absorption bands was observed, indicating that the photoswitches isomerized from the trans isomer to the cis isomer (black curves: before irradiation and red curves: after irradiation) (Fig. 1 and S25†). The reverse isomerization occurred either thermally or upon irradiating a longer wavelength light (430, 470, 525 or 568 nm; blue curves: after irradiation) (Fig. 1 and S26†). The composition of trans and cis isomers in the photostationary state (PSS) at various wavelengths was determined either from ¹H NMR (compounds 1, 2, 3 and 7) or absorption spectra (compounds 4, 5, 6 and 8) (Fig. 2a and S27–S30†). Compounds 7 and 8, having the isothiazole group, showed efficient photoisomerization under longer wavelength visible light irradiation, with a large composition shift between trans-rich and cis-rich PSSs. For example, compound 7 achieved 81% cis in the PSS with 405 nm light and 74% trans in the PSS with 470 nm light. Compound 8 isomerized efficiently from the trans to cis isomer under visible light irradiation at various wavelengths, achieving >73% cis isomer in the PSS with 430 nm light, >66% cis isomer in the PSS with 450 nm, and >47% cis isomer in the PSS with 470 nm. Under 525 nm light, compound 8 gave >83% trans isomer in the PSS (Fig. 2a).

Assuming that the compounds followed first-order thermal isomerization kinetics, the half-lives (t_1/2) of cis isomers were studied by monitoring the change in absorbance at λ_max after the cis-rich PSS (Fig. 2a and S31–S36†). The t_1/2 values vary from several seconds to tens of hours. Compound 1 showed the longest half-life (t_1/2 = 45.2 h) among the 8 compounds studied. Compounds 4 and 6 exhibited relatively short half-lives of their cis-isomers, so brief that their changes could not be accurately tracked by a conventional spectrophotometer. It was estimated that they have half-lives of approximately several seconds. Compounds 7 and 8 showed half-lives of 2.9 hours (λ_max = 338 nm) and 2.7 min (λ_max = 399 nm), respectively. The cis-life time of these photoswitches decreased with an increase in λ_max, a tendency similar to that seen in substituted azobenzenes like 4-aminoazobenzene.^27,28

Structural aspects of photoswitches in trans and cis isomers

To understand the molecular structures of the photoswitches, single X-ray crystallography was performed. Single crystals of the trans isomers (compounds 2–6 and 8) and cis isomers (compounds 2 and 3) were obtained using the vapor diffusion method in a chloroform–hexane solvent system, maintained at low temperature for 2 days. A planar geometry was observed in the trans isomers of compounds 2, 3, 4, 5, 6 and 8, similar to the azobenzene molecule,²⁹ where the phenyl, azo, and heteroaryl moieties all lie in the same plane (Fig. 3a–f). Unlike the typically observed twisted geometry of the cis isomer of azobenzene, the cis isomer of compound 3, which contains a 1,2,3-thiadiazole group, displayed an orthogonal geometry. In this isomer, the phenyl and thiadiazole rings are oriented perpendicularly with the sulfur atom of the thiadiazole ring facing the phenyl ring (Fig. 3b). This configuration is similar to that observed in our previously reported phenylazothiazole photoswitch.¹⁶ In the case of compound 2, with a 1,3,4-thiadiazole ring, the phenyl and thiadiazole rings are in face-to-face orientation (Fig. 3a). Additionally, time-dependent density-functional theory (TDDFT) calculations were carried out in acetonitrile medium to analyse the molecular structure (Fig. 3g–j). The trans isomers of compounds 2–5 showed planar geometries with phenyl, azo and thiadiazole moieties in the same plane. The cis isomers of compounds adopted twisted and orthogonal conformations as seen in the X-ray crystal structures of compounds 2 and 3, respectively. The theoretical calculations of molecular structures match the X-ray crystallography results, confirming their reliability in predicting the molecular geometry of the designed molecules.

Fig. 3 (a–f) Single-crystal X-ray structures of both trans and cis isomers of 2 and 3 and the trans isomers of 4–6 and 8. (g–j) Geometry optimized calculated conformations of both trans and cis isomers of 2–5 (gray = C; light gray = H; purple = N; gold = S).

Conclusions

In conclusion, we have synthesized novel azophotoswitches with various five-membered heteroaryl groups such as thiazole, isothiazole, thiadiazole and isothiadiazole. These compounds were designed with heteroaryl units on either one or both sides of the azo bond. DFT calculations were used to predict the λ_max values of 24 compounds, resulting in the selection of eight compounds for study, including those with the longest λ_max. Compound 8 containing thiazole and isothiazole moieties showed longer λ_max than any of the synthesized compounds. The λ_max of the photoswitch shifted to a longer wavelength with a decrease in cis-life time, as seen in substituted azobenzenes. It efficiently isomerized under visible light at 430 nm, 450 nm, 470 nm (trans–cis) and 525 nm (cis–trans). X-ray crystallography showed that the trans isomers displayed a planar geometry, while the cis isomer of compound 3 adopted a unique orthogonal conformation. The calculations employed in this study reliably predicted both the absorbance maximum and molecular geometry. This work will expand the range of visible-light active photoswitch design, facilitating advancements in photopharmacology.

Experimental

General procedure for the synthesis of compounds 1–3

To a warm (40 °C) NaOH solution (40% aq., 1 mL), 2-amino heteroaryl compounds (2.3 mmol) in 1,4-dioxane (2.5 mL) were added and gently heated to 60 °C. Then, nitroso-benzene (2.53 mmol) was added slowly, and the mixture was stirred for 2 hours. The reaction mixture was then cooled and quenched with water and extracted with ethyl acetate (3 × 25 mL). The organic layer was washed with brine, dried over MgSO₄, and concentrated under reduced pressure. The target compound was then isolated by column chromatography.

General procedures for the synthesis of compounds 4–8

(1) To a cold sodium hypochlorite solution, a cool suspension of aminothiadiazole (4.9 mmol) in dichloromethane (15 mL) was added dropwise followed by stirring for 1 hour at 0 °C. The reaction mixture was quenched with water and extracted with dichloromethane (3 × 25 mL). The organic layer was washed with brine, dried over MgSO₄, and concentrated in vacuo. The target compound was then isolated by column chromatography.

(2) A heteroaryl amino compound (5 mmol) was dissolved in dichloromethane. Potassium permanganate (15 mmol) and ferrous sulfate heptahydrate (10 mmol) were ground together until obtaining a homogeneous powder. The resulting powder was added to the dichloromethane solution. The mixture was stirred for 24 hours at room temperature. Afterward, the reaction mixture was filtered through Celite, and the solvent was evaporated. Finally, the target compound was isolated using column chromatography.

Measurement of thermal half-life

A freshly prepared solution was irradiated at 365 nm, 405 or 430 nm until reaching its photostationary state and immediately kept for thermal back Z–E isomerization in the dark at 25 °C. In all cases, 6–8 spectra at fixed time intervals were recorded under dark conditions to minimize the effect of a light beam on thermal back isomerization. Then, a first order rate constant (k) for the thermal back Z–E isomerization reaction was obtained using the equation:
Abs(BI) = absorbance in the initial state at λ_max.Abs(PSS) = absorbance in the photostationary state at λ_max.Abs(time) = absorbance at λ_max at different time intervals for thermal back isomerization.

For a first order reaction, half-life (t_1/2) can be calculated using the equation:

t_1/2 = 0.693/k

Single crystal X-ray crystallography

General procedure for the crystallization of compounds 2, 3, 4, 5, 6 and 8. A small vial containing the compound dissolved in chloroform (10 mmol, 200 μL) is placed inside a larger container with hexane (70 mL). The system is maintained at −20 °C for 2 days, ensuring it is well sealed to prevent solvent evaporation. During this time, the concentration of hexane in the vial increases, causing the solubility of the compound in the solvent to gradually decrease, leading to the slow precipitation of crystals from the solution.

Crystallization of cis isomers of compounds 2 and 3. The compound solution in a small vial was irradiated with 365 nm light until it reached the photostationary state. Subsequently, the small vial was immediately transferred to a larger container containing hexane and allowed to crystallize, following the general procedure.

Cryoloop was used to mount the single crystals. Crystallographic data were collected using a Rigaku XtaLab Synergy diffractometer with a single microfocus Mo Kα X-ray radiation source (PhotonJet-S), equipped with a Hybrid Pixel (HyPix) array detector (HyPix-6000HE). Data collection, cell refinement, and data reduction were carried out with CrysalisPRO (Rigaku Oxford Diffraction, 2017). The initial structure was solved by SHELXT and expanded using Fourier techniques and refined on F² by the full-matrix least-squares method using the SHELXL2018/3 package compiled into the OLEX2 package. All parameters were refined using anisotropic temperature factors, except for hydrogen atoms, which were refined using the riding model with a fixed C–H bond distance.

DFT calculations

Theoretical calculations were performed using Gaussian 09 (Revision D.01). GaussView 6.0 was used to draw and visualize the molecular structures and feed the inputs. Density functional theory (DFT) and time dependent density functional theory (TDDFT) were employed to optimize the geometries and obtain the electronic transition in the ground state, respectively. The 6-31+G(d,p) basis set with Becke's three-parameter hybrid exchange and Lee–Yang–Parr's correlation functional (B3LYP) was used for geometry optimization. The solvent stabilization of different isomers was investigated by considering the integral equation formalism-polarizable continuum model (IEF-PCM) and choosing acetonitrile as the medium.

Author contributions

N. M. C. conducted most of the experiments and data analyses. N. T. conceptualized the project and P. K. H. helped in the synthesis and data analysis. S. S. performed DFT calculations. K. T. and T. N. measured X-ray crystal structures. H. M. and K. I. discussed the results, N. T. supervised, and all authors contributed to writing the manuscript.

Data availability

The data supporting this article have been included as part of the ESI.†

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

N. M. C. acknowledges the Hokkaido University EXEX Doctoral Fellowship Program. P. K. H. acknowledges the “FY2023 SOUSEI Support Program for Young Researchers” by Hokkaido University and the 9th Hokkaido University Interdepartmental Symposium Research Grant Silver Award. This work was partially supported by the “Crossover Alliance to Create the Future with People, Intelligence and Materials” from MEXT, Japan to P. K. H.

References

1. A. Mukherjee, M. D. Seyfried and B. J. Ravoo, Angew. Chem., Int. Ed., 2023, 62, e202304437.
2. P. Kobauri, F. J. Dekker, W. Szymanski and B. L. Feringa, Angew. Chem., Int. Ed., 2023, 62, e202300681.
3. O. Sadovski, A. A. Beharry, F. Zhang and G. A. Woolley, Angew. Chem., Int. Ed., 2009, 48, 1484–1486.
4. A. A. Beharry, O. Sadovski and G. A. Woolley, J. Am. Chem. Soc., 2011, 133, 19684–19687.
5. S. Samanta, A. A. Beharry, O. Sadovski, T. M. McCormick, A. Babalhavaeji, V. Tropepe and G. A. Woolley, J. Am. Chem. Soc., 2013, 135, 9777–9784.
6. S. Crespi, N. A. Simeth and B. König, Nat. Rev. Chem., 2019, 3, 133–146.
7. C. E. Weston, R. D. Richardson and M. J. Fuchter, Chem. Commun., 2016, 52, 4521–4524.
8. S. Devi, M. Saraswat, S. Grewal and S. Venkataramani, J. Org. Chem., 2018, 83, 4307–4322.
9. A. K. Gaur, H. Kumar, D. Gupta, I. P. Tom, D. N. Nampoothiry, S. K. Thakur, A. Mahadevan, S. Singh and S. Venkataramani, J. Org. Chem., 2022, 87, 6541–6551.
10. J. Otsuki, K. Suwa, K. K. Sarker and C. Sinha, J. Phys. Chem. A, 2007, 111, 1403–1409.
11. J. Calbo, C. E. Weston, A. J. P. White, H. S. Rzepa, J. Contreras-García and M. J. Fuchter, J. Am. Chem. Soc., 2017, 139, 1261–1274.
12. D. Fang, Z. Y. Zhang, Z. Shangguan, Y. He, C. Yu and T. Li, J. Am. Chem. Soc., 2021, 143(36), 14502–14510.
13. C. Slavov, C. Yang, A. H. Heindl, H. A. Wegner, A. Dreuw and J. Wachtveitl, Angew. Chem., Int. Ed., 2020, 59, 380–387.
14. A. D. W. Kennedy, I. Sandler, J. Andréasson, J. Ho and J. E. Beves, Chem. – Eur. J., 2020, 26, 1103–1110.
15. K. Matsuo, S. Thayyil, M. Kawaguchi, H. Nakagawa and N. Tamaoki, Chem. Commun., 2021, 57, 12500–12503.
16. R. Lin, P. K. Hashim, S. Sahu, A. S. Amrutha, N. M. Cheruthu, S. Thazhathethil, K. Takahashi, T. Nakamura, T. Kikukawa and N. Tamaoki, J. Am. Chem. Soc., 2023, 145, 9072–9080.
17. T. Dang, D. Dong, J. Zhang, Y. He, Z.-Y. Zhang and T. Li, Angew. Chem., Int. Ed., 2023, 62, e202301992.
18. T. Dang, Z.-Y. Zhang and T. Li, J. Am. Chem. Soc., 2024, 146(29), 19609–19620.
19. C. E. Weston, R. D. Richardson, P. R. Haycock, A. J. White and M. J. Fuchter, J. Am. Chem. Soc., 2014, 136, 11878–11881.
20. Y. He, Z. Shangguan, Z. Y. Zhang, M. Xie, C. Yu and T. Li, Angew. Chem., Int. Ed., 2021, 60, 16539–16546.
21. M. J. Naim, O. Alam, F. Nawaz, M. J. Alam and P. Alam, J. Pharm. BioAllied Sci., 2016, 8, 2–17.
22. S. S. Alghamdi, R. S. Suliman, K. Almutairi, K. Kahtani and D. Aljatli, Drug Des., Dev. Ther., 2021, 15, 3289–3312.
23. H. Z. Zhang, Z. L. Zhao and C. H. Zhou, Eur. J. Med. Chem., 2018, 144, 444–492.
24. F. Chen, H. Chai, M. B. Su, Y. M. Zhang, J. Li, X. Xie and F. J. Nan, ACS Med. Chem. Lett., 2014, 5(6), 628–633.
25. S. W. Zhang, C. J. Gong, M. B. Su, F. Chen, T. He, Y. M. Zhang, Q. Q. Shen, Y. Su, J. Ding, J. Li, Y. Chen and F. J. Nan, J. Med. Chem., 2020, 63, 804–815.
26. D. Adrion, Calculating photochemical reaction mechanisms with multiconfigurational calculations and dynamics, Doctoral dissertation, Northeastern University, Digital Repository Service at Northeastern University, 2023.
27. B. Uno, Y. Matsuhisa, K. Kano and T. Kubota, Chem. Pharm. Bull., 1984, 32, 1691.
28. G. S. Hartley, J. Chem. Soc., 1938, 633.
29. C. J. Brown, Acta Crystallogr., 1966, 21, 146–152.

---

Footnote

† Electronic supplementary information (ESI) available. CCDC 2371065–2371072. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d4ob01573h

---