Iminobispyrazole (IBP) photoswitches: two pyrazole rings can be better than one

Abstract

We recently demonstrated that suitably functionalised aryliminopyrazoles can exhibit useful photoswitching properties. This study investigates the photoswitching potential of iminobispyrazoles (IBPs). We find that the regiochemistry of the IBPs strongly dictates their photoswitching properties, most notably, the λ_max, the photostationary state, and the thermal half-life of the Z-isomer.

---

Photochromic imines^1–4 based on an aryliminopyrazole (AIP) motif have recently emerged as an interesting class of photoswitch, featuring a C=N bond that can photoisomerise while also exhibiting dynamic-covalent (DC) properties.^1,5 Specifically, AIPs functionalised with ortho-pyrrolidine units exhibit quantitative E-to-Z photoisomerism under visible light, with thermal half-lives (t_1/2) of the metastable Z-isomer exceeding 19 hours at room temperature.¹ In a subsequent study, we demonstrated how the DC properties of these photoswitches could be harnessed: by coupling a thermal equilibrium reaction with a reversible photochemical transformation, we generated a non-equilibrium steady state (NESS)⁶ of the transimination reaction⁷ under photoirradiation.⁵ This system exhibited behaviour consistent with an information ratchet,⁸ operating continuously under light irradiation.^9,10 Building on these foundations and inspired by the structural and property diversity observed in the azoheteroarene field,^11,12 we postulated that iminobispyrazole (IBPs), where both arene rings are pyrazoles, could also show useful photoswitching properties without the need for ortho-amination.

The groups of Li,¹³ Fuchter¹⁴ and Han have demonstrated the utility of azobispyrazoles for their photoswitching properties and potential in energy-storage applications.¹⁴ Inspired by their work on azobispyrazoles, we investigate the photoswitching properties of analogous imine-based systems. We hypothesise that the potential diversity in properties of these chemically similar structures could render IBPs compelling candidates in DC systems.¹⁵ Lehn and co-workers have already demonstrated network behaviours of DC systems,^16,17 which may be well suited to these IBPs.

In this communication, we present the photoswitching properties of a select library, a [3 × 3] matrix, of IBPs. The entries in this library are positional isomers, specifically, regioisomers. We analyse their photoswitching properties, noting that several derivatives exhibit significantly improved t_1/2 values (over 120 times longer) compared to their non-functionalised AIP analogues. Trends of the E-isomer's λ_max and the thermal stability of the Z-isomer are rationalised according to the IBP structure. This communication provides an overview of the photoswitching properties of IBPs and highlights how the regiochemistry of the pyrazole ring impacts these properties.

The nine IBPs shown in Fig. 1 were prepared from commercially available precursors using our previously reported procedure.¹ Specific details of the synthesis and characterisation data are provided in the ESI.† The IBPs were quantitatively prepared, as confirmed by ¹H NMR, subsequent purification on neutralised silica gel to remove excess amine revealed that the IBPs are less stable to column chromatography than their AIP counterparts.¹ The IBPs were isolated as the thermodynamically stable E-isomers and stored in the dark prior to photoswitching studies. A single crystal of E-4,4 suitable for X-ray diffraction was obtained by slow solvent evaporation. The X-ray crystal structure confirmed successful formation of the IBP and that the E-isomer is the thermodynamically stable structure (Fig. 2a). The structure is relatively planar, with only a slight twist away from planarity of the N-pyrazole ring due to steric clash with the adjacent imine bond, similar to the behaviour of the previously reported AIPs.¹

Fig. 1 (a) Schematic of the imine condensation of commercially available amine and aldehyde precursors in CDCl₃, yielding iminobispyrazole (IBP) photoswitches. Monitoring the condensation by ¹H NMR indicates quantitative formation of the imine. (b) Overview of the IBP photoswitches explored in this study. Colour coding refers to the site of substitution, names are given based on the site of substitution for amine:aldehyde precursors.

Fig. 2 (a) Proposed reasoning for the variation in λ_max for the different regioisomers. (b) The X-ray crystal structure of E-4,4 showing a relatively planar conformation, with the exception of the N-pyrazole ring twisting away from planarity (β′) due to steric clash between the C–H of the N-pyrazole and the C–H of the imine. Colour code in panel (a): C, grey; N, blue; H, white. CCDC 2370061.†

The photoswitching properties of the IBPs were investigated spectroscopically in acetonitrile. Compared to E-AIPs, the E-IBPs exhibited larger extinction coefficients at λ_max, ranging from 11 000 to 15 000 M⁻¹ cm⁻¹ (Table 1). The λ_max for the E-IBPs varied depending on the regiochemistry of the photoswitch. Notably, the IBPs formed from the 5,-precursor exhibited the largest red-shift in λ_max, following the order: 5,5 > 5,4 > 5,3, 4,5 > 3,5 > 4,3, > 4,4 > 3,4 > 3,3. We rationalised this trend by considering the extent of conjugation across the relatively planar E-isomer (Fig. 2a and b and Section S5 of the ESI†). For the 5-substituted pyrazoles, the position of the N-methyl motif in the pyrazole ring relative to the imine bond allows for greater conjugation of the lone-pair across the molecule compared to other regioisomers.¹⁸ Following nomenclature from azo-based photoswitches,^13,18,19 we refer to the degree of conjugation in the 5-substituted derivatives as “complete” conjugation, and in the 3- and 4-regioisomers as “partial” conjugation (Fig. 2b). Simply put, greater π-conjugation leads to a greater bathochromic shift of the π–π* transition. The three compounds with the longest absorption wavelengths are E-5,5 (324 nm), E-5,4 (320 nm) and E-5,3 (315 nm). Common to these compounds is that the 5-subsitution is located on the former amine component. These findings indicate that the use of 5-aminopyrazole is more critical for achieving the bathochromic shift than the corresponding aldehyde moiety.

Table 1 Overview of the photoswitching properties of the IBP photoswitches; the processed data can be found in the ESI. The λ_max and the corresponding extinction coefficient (ε) is measured for the E-isomer

	λ max (nm)	ε (M^−1 cm^−1)	t 1/2	% Z at PSS
				340 nm	365 nm	385 nm
a Note that the short t1/2 values preclude the measurement of the % Z at the PSS under these conditions. b The PSS could not be determined at 20 °C.
3,3	296	15 000	6.5 min	41	11	—
3,4	301	13 100	25.8 min	58	15	—
3,5	307	14 900	1.4 min	54	28	—
4,3	305	15 500	6.5 min	40	7	—
4,4	304	14 600	26.3 min	48	7	—
4,5	315	12 500	2.2 min	46	21	—
5,3	315	11 300	50.4 s	—^b	—^b	—
5,4	316	11 500	3.4 min	76	54	18
5,5	324	13 900	17.0 s	—^b	—^b	—

---

Next, we investigated the photoisomerism of these IBPs in acetonitrile solution at 20 °C. The amount of Z-isomer generated under photoirradiation, and at the photostationary state (PSS) for compounds with sufficiently stable Z-isomers (Table 1),¹ was determined using a diode array setup (Fig. S1, ESI†). Photoisomerism was monitored by tracking changes in the UV/vis absorption spectra when irradiated with 340 and 365 nm LEDs; longer wavelengths, particularly 385 nm and 405 nm, afforded negligible photoswitching, except for 5,4. The proportion of Z-isomer generated and the predicted spectra of the pure Z-isomer were calculated using the UV/vis spectroscopic approach reported by Fischer.²⁰ The largest amount of Z-isomer generated at a PSS was for 5,4, achieving up to 76% of the Z-isomer at the 340 nm PSS. In general, the percentage of the Z-isomer formed at the 340 nm PSS for the IBPs is comparable to their ortho-aminated AIP counterparts,¹ while the 365 nm PSS values of the IBPs are significantly lower. We attribute this to the poor spectral overlap of the LEDs’ emission with the UV/vis absorption spectrum of the E-isomers (Table S1, ESI†). Regarding the quantum yields (QYs) of photoisomerism for the IBPs, they are similar in magnitude to those of the previously reported AIPs (Section S4.4 of the ESI†).¹ Comparing the spectra of the E-isomers with the predicted spectra of the Z-isomers, and the similarity in the QY values to AIPs, we anticipate that adjusting the photoirradiation wavelengths may be able to further enhance the fraction of Z-isomer generated.

The UV/vis absorption spectra of the Z-isomers display two relatively intense absorption bands: one near the λ_max of the E-isomer (the lowest energy transition of the E-isomer corresponds to a π → π* transition, Fig. S32, ESI†), which is significantly lower in intensity, and one hypsochromically shifted to ca. 250 nm (Fig. 3a). A weaker absorption band is also observed, red-shifted with respect to the longest wavelength of absorption of the E-isomer, attributed to an n → π* transition (Fig. S32, ESI†). Thus, these IBPs exhibit positive photochromism, whereby λ_max(E) < λ_max(Z). While quantitative E-to-Z photoisomerism is not achieved for the IBPs under the LED wavelengths investigated here, it should be noted that the amount of Z-isomer generated at the PSS is greater than that of non-functionalised AIPs¹ and previously reported arylimine-based photoswitches.^2,21,22 Furthermore, the IBPs showed good resistance to fatigue; 4,4 could be switched back and forth between the 340 nm and 365 nm PSS reversibly with no observable degradation after 10 cycles (Fig. S30, ESI†).

Fig. 3 (a) UV/vis absorption spectra of 4,4 (40 μM, 20 °C, MeCN) as the E-isomer, at the 340 nm and 365 nm PSS, and the predicted spectrum of the Z-isomer obtained as described in Section S4.3 of the ESI.† Note the low-energy absorption band for the Z-isomer, which results in the positive photochromic properties. (b) The rate of thermal Z-to-E isomerism at 20 °C. (inset) Eyring plot of the rate of thermal reversion at different temperatures, with thermodynamic values reported.

The thermal stability of the metastable Z-isomers exhibited a broad range of t_1/2 values dependent on the regiochemistry. IBP 4,4 showed the longest t_1/2 at 26.3 minutes, followed by 3,4 with a half-life of 25.7 minutes. These t_1/2 values are significantly longer than those of previously reported arylimine switches,^2,21,22 only surpassed by the AIP containing two ortho-pyrrolidine units.¹ This highlights the potential to develop imine-based photoswitches with longer t_1/2 values compared to previous arylimine-based photoswitches,^2,21,22 without the need for ortho-amination. Interestingly, the IBPs with the most bathochromically shifted UV/vis absorption bands, i.e., the 5-substituted derivatives, generally show the lowest thermal stability of their Z-isomers (with t_1/2 values of less than 1.3 min at 20 °C). Similar observations for azo-based switches have been previously rationalised based on the polarisation of the photochromic double bond, where greater electron donation from a heteroaryl ring can decrease the bond order.¹⁸ A summary of the key photoswitching properties for this [3 × 3] matrix of IBPs is presented in Table 1.

Computational investigations were conducted to determine the optimised geometries of the E, Z and transition state (TS) structures using the ωB97X-D4/def2-TZVPP level of theory and a CPCM solvation model for acetonitrile.¹ The predicted conformation of E-4,4 aligns well with the single crystal X-ray structure (Fig. 2a and Fig. S37, ESI†). All E-IBPs exhibit a small twist away from planarity due to the steric clash of the N-pyrazole ring's C–H bonds with the imine bond's C–H. For the TSs, the IBPs adopt either a planar or perpendicular TS structure² and exhibit a linearisation of the imine motif (C=N–C), consistent with an inversion isomerisation pathway.^1,2 In terms of preference, the 5,-IBPs exhibit a perpendicular TS, the 4,-IBPs adopt planar TS conformations, while the 3,-analogues adopt a perpendicular TS geometry.

For the Z-isomers, most IBPs adopted a near T-shape conformation. Exceptions included Z-5,5 and Z-4,5, attributed to a preferred twisted structure that enables greater conjugation across the molecule. Interestingly, Z-4,3 adopts a planar structure, potentially stabilised by an intramolecular hydrogen bond (2.2 Å distance, Fig. S31, ESI†). This observation agrees with the azobispyrazoles reported by Li and co-workers.¹³ It is interesting to note that Z-3,4 does not adopt a planar geometry, indicating the diversity arising from the regiochemistry of these systems.

For derivatives that adopt a near T-shape conformation, the C–H bond of the aldehyde fragment is directed toward the π-system of the N-pyrazole ring. The proximity of the C–H bond to the centroid of the N-pyrazole ring, 2.6 Å for Z-4,4, may enable stabilising CH–π interactions, with some indication shown in the non-covalent interaction plot (Fig. S32, ESI†).^1,13,18 In the case of Z-5,5, the twisted geometry appears to significantly destabilise the Z-isomer compared to the E-isomer, indicated by a ΔG_Z–E of 30 kJ mol⁻¹. Unlike AIPs, where the t_1/2 appeared to be dominated by a stabilisation of the Z-isomer, the IBPs show a more balanced effect between TS destabilisation and Z-isomer stabilisation; ongoing work attempts to elucidate this behaviour.

In summary, we have prepared a [3 × 3] library of nine iminobispyrazoles and characterised their photoswitching properties. While the overall amount of the metastable isomer generated under photoirradiation is less than that achieved for ortho-aminated AIPs, the fraction of Z-isomer formed is improved compared to the previously reported arylimines. The highest fraction of Z-isomer generated under the conditions explored here was 76% for 5,4. Moreover, while relatively long t_1/2 values were only achieved upon ortho-functionalisation of the AIPs, we show that the IBPs, specifically 4,4 and 3,4, exhibit enhanced stabilities of the Z-isomer. Efforts are underway to explore the use of this accessible and diverse photoswitch library in the context of systems chemistry by exploiting the dynamic-covalent character of the imine bond.

J. W. and C. L. performed the synthesis, characterisation and photoswitching experiments. L. K. conducted the computational investigations. J. W. and J. L. G. prepared the manuscript, with input from all authors. J. L. G. conceived the project and supervised the research.

This work was funded by the Fonds der Chemischen Industrie (FCI, Liebig Fellowship for J. L. G., PhD Fellowship for J. W.). Olga Anhalt solved the crystal structure of E-4,4. We thank Prof. Roland Mitric for computational infrastructure. We thank Prof. Frank Würthner for support and for providing infrastructure.

Data availability

The experimental procedures, analytical data and computational details are available within the manuscript and its ESI.† CCDC number 2370061 contains the supplementary crystallographic data for IBP E-4,4. Data are available upon request from the authors.

Conflicts of interest

There are no conflicts to declare.

Notes and references

1. J. Wu, L. Kreimendahl, S. Tao, O. Anhalt and J. L. Greenfield, Chem. Sci., 2024, 15, 3872–3878.
2. Y. Luo, M. Utecht, J. Dokić, S. Korchak, H.-M. M. Vieth, R. Haag and P. Saalfrank, Chem. Phys. Chem., 2011, 12, 2311–2321.
3. D. Veselý, J. Jančík, M. Weiter, D. Blasi, N. Ivanova, J. Krajčovič and A. Georgiev, J. Photochem. Photobiol., A, 2022, 430, 113994.
4. A. Georgiev, D. Yordanov, D. Dimov, I. Zhivkov, D. Nazarova and M. Weiter, J. Photochem. Photobiol., A, 2020, 393, 112443.
5. J. Wu and J. L. Greenfield, J. Am. Chem. Soc., 2024, 146, 20720–20727.
6. I. Aprahamian and S. M. Goldup, J. Am. Chem. Soc., 2023, 145, 14169–14183.
7. M. Ciaccia and S. Di Stefano, Org. Biomol. Chem., 2015, 13, 646–654.
8. V. Serreli, C.-F. Lee, E. R. Kay and D. A. Leigh, Nature, 2007, 445, 523–527.
9. S. Borsley, D. A. Leigh and B. M. W. Roberts, Angew. Chem., Int. Ed., 2024, 63, e202400495.
10. T. Sangchai, S. Al Shehimy, E. Penocchio and G. Ragazzon, Angew. Chem., Int. Ed., 2023, 62, e202309501.
11. S. Crespi, N. A. Simeth and B. König, Nat. Rev. Chem., 2019, 3, 133–146.
12. R.-R. Griffiths, J. L. Greenfield, A. R. Thawani, A. R. Jamasb, H. B. Moss, A. Bourached, P. Jones, W. McCorkindale, A. A. Aldrick, M. J. Fuchter and A. A. Lee, Chem. Sci., 2022, 13, 13541–13551.
13. Y. He, Z. Shangguan, Z. Zhang, M. Xie, C. Yu and T. Li, Angew. Chem., Int. Ed., 2021, 60, 16539–16546.
14. A. Gonzalez, M. Odaybat, M. Le, J. L. Greenfield, A. J. P. White, X. Li, M. J. Fuchter and G. G. D. Han, J. Am. Chem. Soc., 2022, 144, 19430–19436.
15. L. Köttner and H. Dube, Angew. Chem., Int. Ed., 2024, e202409214.
16. Z. Yang, F. Esteve, C. Antheaume and J.-M. Lehn, J. Am. Chem. Soc., 2024, 146, 15438–15445.
17. Z. Yang, F. Esteve, C. Antheaume and J.-M. Lehn, Chem. Sci., 2023, 14, 6631–6642.
18. J. Calbo, C. E. Weston, A. J. P. P. White, H. S. Rzepa, J. Contreras-García and M. J. Fuchter, J. Am. Chem. Soc., 2017, 139, 1261–1274.
19. J. L. Greenfield, A. R. Thawani, M. Odaybat, R. S. L. Gibson, T. B. Jackson and M. J. Fuchter, Molecular Photoswitches, Wiley, 2022, pp. 83–112.
20. E. Fischer, J. Phys. Chem., 1967, 71, 3704–3706.
21. T. Matsuhira, H. Yamamoto, A. Onoda, T. A. Okamura and N. Ueyama, Org. Biomol. Chem., 2006, 4, 1338–1342.
22. P. J. Coelho, M. C. R. Castro and M. M. M. Raposo, J. Photochem. Photobiol., A, 2013, 259, 59–65.

---

Footnote

† Electronic supplementary information (ESI) available: A materials and methods description and additional data on the characterisation of precursor and product (ESI-MS, NMR, UV/vis), a description of the photoswitching experiments (PSS, t_1/2, QY) and the full data and description of the computational investigations. CCDC 2370061. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d4cc03517h

---