Combined experimental and theoretical insights into the solvent-dependent photochromic properties of nitrospiropyrans

Abstract

Spiropyrans exhibit reversible photochromism that strongly depends on solvent polarity; yet a quantitative understanding of this relationship remains incomplete. Here we combine detailed spectroscopic and kinetic experiments with ab initio calculations to investigate 5’-methoxy-1',3',3'-trimethyl-6-nitrospiro[chromene-2,2'-indoline] (SP). Ultraviolet irradiation at 368 nm populates the S₃ (¹ππ*) state, initiating the ring-opening reaction to the merocyanine (MC) form, which is subsequently followed by thermal ring-closing. Theoretical calculations indicate that photoexcited SP undergoes nonradiative relaxation to the S₁ charge-transfer state, from which the system reaches the ground state via a ring-opened conical intersection. Experimental kinetics studies have revealed that increasing solvent polarity slows relaxation and raises the activation energy from 89.7 kJ mol⁻¹ in dioxane to 104.6 kJ mol⁻¹ in acetone. Calculations have reproduced this trend, showing that polar solvents preferentially stabilize MC over the transition state. The observed blue shift of the MC absorption band with increasing polarity confirms the presence of negative solvatochromism. This combined experimental-theoretical analysis provides a quantitative framework linking solvent polarity, photochromic efficiency, and the activation energy in a model nitrospiropyran system.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-29654-3.

Introduction

Spiropyrans are a widely studied class of organic photochromic molecules capable of reversible structural transformation between a colorless spiro (SP) form and a colored merocyanine (MC) form upon irradiation with ultraviolet light^1,2. Without external light stimuli, spiropyrans exist in the closed SP form. UV-A irradiation induces the cleavage of the C_spiro–O bond, resulting in the formation of the metastable MC form (cf. Figure 1). This reaction alters the electronic structure and optical properties of these compounds, enabling their applications in light-responsive materials, sensors, and optical data storage.

Fig. 1.

Photochromic reaction of the SP compound under investigation.

Among spiropyrans, 6-nitro derivatives are particularly interesting due to their enhanced coloration efficiency and tunable photochemical pathways³. Firstly, it increases the quantum yield of photocoloration by introducing a triplet pathway. Secondly, it stabilizes the zwitterionic merocyanine form of the molecule, which inhibits the thermal back reaction (MC → SP). Finally, a proper modification of the structure of 6-nitrospiropyrans enables the alteration of the type of photochromism from positive to negative⁴. The efficient and stable color change process favors 6-nitrospiropyrans as useful compounds for the fabrication of ophthalmic lenses, intelligent pigments, or UV radiation indicators^5–7. 6-nitrospiropyrans were also successfully utilized to create smart materials, such as multi-responsive amphiphilic copolymers, which can potentially be used as drug-release systems⁸, or temperature-, light- and stress-responsive hydrogels exhibiting relatively high mechanical properties⁹.

On the other hand, the introduction of the nitro group adversely affects the photochemical stability of the spiropyran, which is inherently low³. Such behavior hinders the practical implementation of 6-nitrospiropyrans in erasable, directly readable, and writable memory materials, which require highly fatigue-resistant and thermally stable photochromic compounds. This has prompted researchers to develop new synthetic routes or modify existing photochromic compounds to improve their performance¹⁰. The challenge of poor photostability has inspired many studies investigating how solvent effects, complexation, degradation, and the molecular structure influence the photochromic behavior and stability of 6-nitrospiropyrans^10–19. In addition to photostability, the spectral and kinetic properties of all spiropyrans are crucial for their further investigation in a practical application-oriented context. For example, when building spiropyran-based systems for dynamic holography, the most desirable feature is a swift opening-closing (coloring-decoloring) reaction under irradiation²⁰. Contrarily, in applications such as photochromic pigments and paints, more important than the reaction rate is the efficiency and stability of the color change.

While previous studies, notably by Görner¹⁸, established that solvent polarity affects both the thermal back reaction and solvatochromism, systematic correlations between experimental kinetics and theoretical activation parameters remain limited. In this paper, the spectral, kinetic, and structural properties of 5’-methoxy-1',3',3'-trimethyl-6-nitrospiro[chromene-2,2'-indoline] (SP), representative of the 6-nitrospiropyran family, are investigated. Firstly, the synthesis and thorough structural analysis, including the magnetic nuclear resonance and X-ray diffraction experiments, are described. Subsequently, the spectral characterization is presented. It consists of UV–Vis absorption spectra measured before and immediately after irradiation with UV light, as well as during the thermal relaxation process. In the case of a kinetic study of the back reaction, the reaction rate constants of the ring-closing process and its activation energy are determined in solvents of different polarities, and solvatochromic and solvatokinetic effects are discussed. Finally, the ab initio results are used to clarify the photochromic mechanism, with particular emphasis on the nonradiative deactivation pathway, which is analyzed here for the first time in nitrospiropyrans. This integrated approach also clarifies how the solvent environment modulates the relative stabilities of the MC and SP forms and quantitatively links experimental and theoretical activation energies.

Results and discussion

The investigated spiroindoline derivative (SP) belongs to the spiropyran family of compounds, which exhibit reversible photochromism. The chemical structure of the studied SP and the photochromic reaction it undergoes are presented schematically in Fig. 1. The main part of the compound structure, in its stable closed SP form, is the pyran moiety, which is fused to the nitrophenyl group, forming a benzopyran-like structure connected to the indoline fragment. There are three methyl groups attached to the nitrogen-containing ring of the indoline group (at positions 1, 3, and 3). There is also a methoxy group linked to the benzene portion of the indoline group. In solutions at room temperature, the closed-ring isomer of SP is the dominant form. Solutions of the compound in all solvents, including dioxane, DCM, and acetone, are colorless, as spiropyran in its closed form does not absorb light in the visible range. Irradiation of solutions with UV light of wavelength 368 nm causes the splitting of the bond between the spiro carbon atom and the oxygen atom in the pyran ring and subsequent isomerization to the metastable MC form. The latter solution, depending on the solvent used, becomes either blue, purple, or pink (as shown further). The strong electron-accepting nitro group in the structure induces the zwitterionic character of the merocyanine form. The reverse ring-closing reaction does not require light stimulation as it occurs spontaneously after switching off the UV light source due to thermal relaxation. The thermal back reaction is observed as a gradual bleaching of the colors of the previously irradiated solutions.

Characterization of the SP compound

NMR spectra of SP were obtained at frequencies of 150 MHz (¹³C NMR spectrum) and 600 MHz (¹H NMR spectrum), and they are provided as supplementary information (Supplementary Fig. S2 and Fig. S3). The obtained HRMS (ESI–TOF) spectrum confirmed the molecular composition of the SP compound (cf. Supplementary Fig. S4).

5'-methoxy-1',3',3'-trimethyl-6-nitrospiro[chromene-2,2'-indoline]

¹H NMR (δ/ppm, chloroform-d, 600 MHz): 8.03–7.98 (m, 2H), 6.91 (d, 1H, J = 10,3 Hz), 6.77 (d, 1H, J = 8.8 Hz), 6.74–6.71(m, 2H), 6.46 (d, 1H, J = 9.0 Hz), 5.85 (d, 1H, J = 10.3 Hz), 3.80 (s, 3H), 2.69 (s, 3H), 1.28 (s, 3H), 1.19 (s, 3H). ¹³C NMR (δ/ppm, chloroform-d, 150 MHz): 159.97, 154.17, 141.88, 140.87, 137.70, 128.17, 125.81, 122.64, 121.57, 118.68, 115.42, 111.43, 109.53, 107.18, 106.82, 55.89, 52.38, 29.18, 25.79, 19.84. HRMS (ESI–TOF): m/z [M + H]⁺ calcd for C₂₀H₂₁N₂O₄⁺ 353.1501; found 353.1499. ¹H NMR parameters are in agreement with the literature data²¹.

The SP compound crystallizes in the triclinic crystal system, in the centrosymmetric space group P-1 (Supplementary Table S1). The data analysis has revealed that one five-membered and one six-membered ring share a common carbon atom, which is a chiral center. The symmetry-independent part of the unit cell contains two molecules of R configuration on the C2 and C23 atoms, as shown in Fig. 2. The S enantiomers are generated with the inversion center and complete the volume of the unit cell.

Fig. 2.

(a) Molecular structure and atomic labeling scheme for SP. Displacement ellipsoids for non-H atoms were drawn at a 30% probability level. (b) Overlay of molecular structures of two symmetry-independent molecules.

Comparing both symmetry-independent molecules, their respective bond lengths are virtually identical (Supplementary Table S2). Figure 2b) shows the overlay of molecular structures, indicating their high similarity. The most crucial difference in the molecular structure is observed for the nitro group, which occupies positions on both sides of the aromatic ring. The nitro group plays a significant role in the formation of the crystal structure. The oxygen atoms are engaged in the formation of hydrogen bonds, where the methyl group of the adjacent molecule donates a hydrogen atom (Supplementary Table S3). Only the C30–H30B···O2 hydrogen bond is clearly observed on the d_norm function mapped on the Hirshfeld surface, as shown in Fig. 3²².

Fig. 3.

The d_norm parameter mapped on the Hirshfeld surface for two symmetry-independent molecules. Atom labels in C–H···O and NO₂···NO₂ interactions are indicated.

The other C–H···O interactions are barely visible, depicted as white areas on the map. Apart from hydrogen bonds, the nitro groups adopt one of the possible modes of interaction (either parallel or perpendicularly oriented). In the studied crystal, the adjacent NO₂ groups are not ideally parallel to each other but are rather slightly tilted. The O2···O7 distance is below the sum of the van der Waals radii of 2.931 Å. It can play a crucial role in the cohesion of the structure, especially in molecular systems lacking functional groups that can form strong hydrogen bonds^23–34. Experimental details from the crystallographic studies of the investigated compound are gathered in Supplementary Table S1.

Photochromic properties

The photochromic behavior of SP dissolved in three different organic solvents was investigated in detail using UV–Vis spectroscopy. The results are presented in Fig. 4. The initial spectra (before irradiation) of SP in dioxane (Fig. 4a1)), DCM (Fig. 4a2)) and acetone (Fig. 4a3)) were measured at the beginning of spectroscopic studies. The colorless closed spiro form of SP behaves similarly in all the chosen solvents. There are two characteristic absorption bands observed in the initial spectra in the UV region for the SP solutions in dioxane and DCM, as shown in Fig. 4a1) and Fig. 4a2), respectively. The first strong absorption band, with a maximum localized at 266 nm, and the second, of lower intensity, with a maximum located at approximately 317 nm (319 nm in DCM), are clearly observed. The theoretical calculations (discussed later) indicate that these bands correspond to ¹π → π* electronic transitions occurring in the 2H-1-benzopyran part of SP. In the case of the SP solution in acetone (cf. Figure 4a3)), the first of the discussed bands is not observed in the initial spectrum, which is due to the solvent used and its absorption edge (UV cutoff for acetone λ_cutoff = 330 nm; the absorption of acetone below λ_cutoff is too intense and makes it impossible to observe bands in the spectrum).

Fig. 4.

Photochromic behavior and kinetics of the thermal ring-closing reaction of the SP compound studied using UV–Vis spectroscopy. Photochromic reactions of SP dissolved in: (a1) dioxane, (a2) DCM, and (a3) acetone; in each graph (a1)–(a3) the absorption spectrum before irradiation (the dark brown curve), the spectrum right after switching off the source of light (the brown curve), and the spectrum changes due to the thermal back reaction (the intermediate curves) are presented; the vertical line indicates the irradiation wavelength. Determination of the reaction rate constant of the thermal ring-closing reaction (k) of SP solutions in: (b1) dioxane, (b2) DCM, and (b3) acetone; graphs (b1)–(b3) show the kinetic curves of thermal relaxation obtained by measuring the absorbance changes as a function of time at the selected wavelength (corresponding to the absorption maximum of the merocyanine form). The insets in (b1)–(b3) illustrate the determination of the ring-closing reaction rate constant based on Eq. (1): the black line is a plot of the function f(t), which corresponds to the experimental data and the kinetic curves shown in (b1)–(b3), whereas the red line is a linear fit of the f(t) dependence; the determined k values have been provided in the insets. The measurements were performed at ambient temperature (298 K).

The photochemical ring-opening reaction of SP was induced by irradiating the solution with UV light (λ_em = 368 nm) for 1 min. The measurements were performed for all SP solutions and conducted under the same experimental conditions. After the initialization of the photochromic process, the open-ring isomer of SP began to form, and its concentration increased during the irradiation time. The formation of the colored merocyanine form was clearly revealed in the spectrum by the appearance of a new, intense absorption band with a maximum localized in the visible range (the spectrum was measured immediately after switching the light off). The photochemical ring-opening reaction of SP and the subsequent formation of the new absorption band in the visible range, which corresponds to the MC form, occurred in the three studied SP solutions, and the results are presented in Fig. 4a1)–a3). In the case of the dioxane solution, a new MC band with a maximum localized at 593 nm was observed (cf. Figure 4 a1)). In contrast, in more polar DCM and acetone solvents, the absorption band maximum of the merocyanine form was blue-shifted to 582 nm ((Fig. 4a2)) and 569 nm (Fig. 4a3)), respectively. The positions of the absorption band maxima of the MC form in different solutions are given in Table 1. Considering the polarity of the solvent used and its influence on the position of the absorption maximum of the MC form, a clear dependency is observed, indicating that the solvatochromic effect is present. The results show that the more polar the solvent is, the more the MC band is blue-shifted. The most significant difference in the positions of the absorption band maxima between the least polar dioxane and the most polar acetone amounts to 24 nm (Table 1). In the intermediately polar DCM, the absorption band maximum of MC falls between these limits (Table 1). Since the band of the merocyanine form is blue-shifted as the solvent polarity increases, negative solvatochromism is observed, which is in accordance with literature reports ^4,13,14,16.

Table 1.

Results obtained from spectral and kinetic studies of the compound SP dissolved in different organic solutions: the position of the absorption band maximum of the merocyanine form (λ_{max, B}), the position of the isosbestic point (λ_iso), the efficiency of the UV light induced ring-opening reaction (), the kinetic rate constant of thermal back ring-closing reaction (k) and the time constant (τ) (k and τ measured at room temperature), the activation energy of thermal relaxation (E_a).

Solvent	λmax, B [nm]	λiso [nm]	[a.u.]	k [s-1]	t [s]	Ea [kJ/mol]
Dioxane	593	348	0.887	2.15 × 10–2	46.5	89.7
Dichloromethane (DCM)	582	353, 324	0.368	3.35 × 10–3	298.5	100.6
Acetone	569	–*	0.776	9.07 × 10–4	1102.5	104.6

*There was no isosbestic point observed during the spectroscopic measurements.

The efficiency of the photochemical ring-opening reaction of SP can be expressed by the change in absorbance at the maximum of the merocyanine form ( – the absolute value of the difference between the absorbance before irradiation and the absorbance right after switching the light source off). The values of are provided in Table 1. The reaction is efficient in all studied solvents; however, there is no observed the influence of the polarity of the solvent on the reaction efficiency. The reaction occurs most efficiently in the least polar dioxane, while it occurs with the lowest efficiency in the intermediately polar DCM (cf. Table 1).

The changes in the absorption spectrum of the SP solution resulting from UV irradiation and the occurrence of the ring-opening reaction leading to the formation of MC, as well as the influence of solvent polarity on the position of the absorption band maximum of the merocyanine form, were macroscopically observed by the naked eye as a change in the color of the solution, which has been presented by the photos gathered in Fig. 5. Colorless solutions in dioxane, methylene chloride, and acetone become, blue, purple and pink, respectively, after UV irradiation (cf. Figure 5). After switching the light off, the SP compound returns to its stable closed spiro form, which is reflected in the spectrum changes measured at different times, counted from the moment of turning-off the UV diode (cf. intermediate curves shown in Fig. Figure 4 a1)–a3)) and manifests as a change in the color of the solution (cf. Figure 5). The thermally-driven reverse reaction causes the system to return to its initial equilibrium state.

Fig. 5.

Photochromic reactions in SP solutions in dioxane, DCM and acetone, respectively, from left to right in each photo, as seen by the naked eye; left photo (A): solutions before irradiation – the colorless solution indicates the presence of the closed spiro form (dominance of the closed-ring isomers); middle photo (B): solutions right after switching off the light source – the colored solution indicates the presence of the merocyanine form (dominance of the open-ring isomers resulting from the occurrence of the ring-opening reaction); right photo (A’): solutions a few tens of seconds after switching off the light source – the disappearance of the color of the solution due to the thermal relaxation process B → A’ (the back ring-closing reaction).

Despite the clearly observed photochromic properties presented in Fig. 4a1)–a3), the isosbestic points (λ_iso), which confirm the occurrence of the light-induced and thermally-driven photochromic reactions, have been observed only in solutions of SP in dioxane at 348 nm and in methylene chloride at 324 nm and 353 nm. The positions of the isosbestic points have been provided in Table 1. It is worth pointing out that a quasi – isosbestic point is observed in every SP solution at around 450 nm, which is evidently seen in Fig. 4a1)–a3). At this specific wavelength, the curves do not intersect; however, they are very close to each other, and the absorption at this point is minimal.

Additionally, solutions of SP in ethanol and in hexane were investigated. However, the SP compound in these solvents behaves differently due to their respective extreme polarity and non-polarity, which requires further detailed investigation. These two SP solutions will be the subject of further studies and will be described in a separate paper.

The kinetics of the thermal ring-closing reaction of SP was measured immediately after switching off the UV light source by monitoring the absorbance changes as a function of time at the wavelength corresponding to the maximum of the absorption band of the merocyanine form. The results, i.e., the kinetic curves, obtained for all three solutions of SP are shown in Fig. 4b1)–b3). Based on these results, the kinetic reaction rate constant (k) of the ring-closing reaction was determined by linear regression according to the following equation:

1

where A₀ corresponds to the absorbance at the beginning of the back reaction (the maximum absorbance of the merocyanine form), A_t is the momentary absorbance (changing during the reaction), and A_∞ is the absorbance in the final equilibrium state. The determined rate constants of the ring-closing reaction and the respective plots are presented as insets in Fig. 4b1)–b3). The Eq. (1) has been derived based on the assumption of first-order kinetics for the ring-closing reaction and the applicability of the Lambert-Beer law. The determined values of the reaction rate constants are gathered in Table 1.

To understand the mechanism of the ring-closing reaction of SP in solutions differing in the polarity of the solvent, as well as to compare the experimental results with the theoretical ones, the activation energies (E_a) were also determined. The Arrhenius equation (in logarithmic form) has been employed for this purpose:

2

where k corresponds to the reaction rate constant in the measurement temperature (T) and R is the gas constant (8.314 J·K^-1mol^-1). Measurements of the kinetic curves (monitoring the absorbance changes at the maximum of the merocyanine band as a function of time after switching off the UV light source) were conducted at different temperatures. The occurrence of thermal back reaction was measured at six temperatures, varying from 293 to 343 K, for the SP solution placed in a thermostated cuvette mounted within the spectrophotometer holder. The obtained data have been used for the determination of the reaction rate constants (Eq. 1), which were further used to determine the activation energy (Eq. 2). Supplementary Figure S5 presents the results of the determination of the ring-closing rate constants in the following temperatures: 293 K, 303 K, 313 K, 323 K, 333 K and 343 K, along with the activation energies of the reaction obtained for the SP solutions in dioxane, DCM, and acetone. The values of the parameters that characterize the ring-closing reaction of SP, as determined from kinetic studies, as well as the parameters obtained from spectral studies, are presented in the Table. 1.

Results of kinetic studies indicate that not only do spectral parameters depend on the polarity of the solvent, but the reaction kinetics also depends on solvent polarity. The thermal ring-closing reaction at room temperature occurs fastest for the SP compound dissolved in the least polar dioxane (it occurs in a few dozen seconds). In the less polar DCM, the reaction time constant is on the order of hundreds of seconds, while in the most polar acetone, the reaction occurs the slowest and lasts a few minutes (cf. Table 1). Therefore, it can be concluded that the more polar the solvent used for the preparation of the solution, the slower the thermal relaxation is.

Comparing the activation energy, its value increases with the rise in solvent polarity. In acetone, the most polar solvent investigated, thermal relaxation requires the greatest activation energy (104.6 kJ/mol). The activation energy in the less polar DCM is lower (100.6 kJ/mol) and is the lowest for dioxane, which has the lowest polarity (cf. Table 1).

Theoretical studies of the mechanism of photochromic reaction

To complement the experimental data, we performed geometry optimization of the title compound and the possible merocyanine isomers. The ground state structures were optimized using the PBE0-D3BJ/def2-TZVP method, assuming the CPCM solvent model for acetone, dichloromethane and 1,4-dioxane. Since the character of the low-lying electronic states and structural data are very similar for all considered solvents, we will focus on the results obtained for the most polar one, namely acetone. The transition states between selected isomers were located using the Nudged Elastic Band (NEB) approach, assuming the same electronic structure calculation method. Figure 6 shows a schematic diagram of the located isomers. In general, the located isomers are similar to those reported in previous computational studies^25–28. We assume the customary nomenclature of MC isomers, indicating the cis–trans isomerization of the π-conjugated methane bridge²⁵. The relative free energies of selected isomers and transition states, calculated using the DLPNO-CCSD(T)/def2-TZVPP/CPCM method while assuming geometries and thermochemical corrections calculated at PBE0-D3BJ/def2-TZVP/CPCM, are also shown in Fig. 6.

Fig. 6.

A schematic representation of the studied isomers of the title compound. SP denotes the spiropyran isomer, and the labeling of merocyanine isomers indicates the cis–trans isomerization of the π-conjugated polyene bridge. The values in parentheses are the relative Gibbs free energies (in kJ/mol) of the local minimum energy structures, and the numbers above the arrows are the corresponding transition state free energies calculated assuming the CPCM model of acetone.

Comparison of the experimentally derived activation energies reported in Table 1 with the calculated data summarized in Table 2 indicates that the most stable merocyanine isomer, TTC, is likely the dominant form in solution, even though the CCT and CCC isomers yield a better overlap with the experimental spectra. The activation free energies for the equilibrium thermal relaxation of CCT, CCC, and TCC in acetone, which are 5.5, 26.6, and 28.7 kJ/mol, respectively, are well below the experimental activation energy of 104.6 kJ/mol. Only for the TTC isomer do we obtain the comparable value of 90.6 kJ/mol. The same applies to the results obtained assuming the CPCM model of 1,4-dioxane and DCM (cf. Supplementary Table S4 for details). Although the experimental activation energies are offset by roughly 14 kJ/mol (for DCM and acetone) to 20 kJ/mol (for 1,4-dioxane), the correlation of experimental and theoretical data is high (R² = 0.996). Assuming the Eyring equation, the RT term of 2.4 kJ/mol at ambient conditions should be added to the computed activation free energies to compare with the activation energies derived from the Arrhenius equation, which slightly improves the agreement of experimental and calculated data; however, that does not affect the conclusions. According to our calculations, polar solvents tend to significantly stabilize the merocyanine form (by roughly 20 kJ/mol when passing from 1,4-dioxane to more polar DCM and acetone) and only slightly (by a few kJ/mol) stabilize the corresponding transition state (cf. Supplementary Table S4). Hence, the activation free energy is rising with increasing solvent polarity. It is interesting to note that the calculated free energy barrier for TTC to TTT isomerization of 43.0 kJ/mol agrees well with the experimentally derived value of 43.6 ± 3 kJ/mol for a similar compound (6-nitro-8-bromo-BIPS)²⁹.

Table 2.

The activation free energies (ΔG^‡) in kJ/mol for the dark relaxation of MC to SP calculated for the selected merocyanine isomers using the DLPNO-CCSD(T)/def2-TZVPP/CPCM method, assuming geometries and thermochemical corrections calculated at the PBE0-D3BJ/def2-TZVP/CPCM level.

Isomer	1,4-dioxane	DCM	acetone
CCC	30.7	25.9	26.6
CCT	3.0	4.4	5.5
TCC	26.2	28.5	28.7
TTC	69.9	86.8	90.6
Exp	89.7	100.6	104.6

The vertical excitation energies computed at the TD-DFT level, shown in Supplementary Table S5, are relatively far from the maxima of the experimental absorption bands of the irradiated solutions (since the former neglect the Franck–Condon progression)^30,31. However, the S₁ absorption bands simulated using the Adiabatic Hessian After a Step (AHAS) model, implemented in the Excited State Dynamics module of ORCA 6.0.1, show much better agreement. This demonstrates the importance of the vibrational progression in assigning the absorption spectra of the MC form, providing an interesting observation that complements the conclusions of Hernández and Curchod³². Figure 7 shows the selected AHAS absorption bands simulated for the S₁ state, assuming the CPCM model of acetylene solvent, while the corresponding maxima of absorption bands are reported in Table 3.

Fig. 7.

Plot of the experimental and simulated absorption bands of the studied compound in acetone solution. In the latter case, the AHAS method was employed to simulate the S₁ state of selected merocyanine isomers. In contrast, for the spiropyran, the VG approach was used to simulate the S₃ state absorption band.

Table 3.

Vertical excitation energies (ΔE in eV) along with the corresponding λ_max (in nm) and oscillator strengths (f) of the selected isomers of the studied compound calculated using the PBE0-D3BJ/def2-TZVP/CPCM method (acetone) at the corresponding optimized geometries. The last two columns report the maxima of the simulated AHAS bands and the corresponding experimental band tentatively assigned to the TTC isomer.

Isomer	ΔES1	λ	f	λmax(1)	λmax(exp)
CCC	2.54	489	0.41	575
CCT	2.77	447	0.47	521
CTC	2.70	459	1.26	541
CTT	2.55	486	1.18	577
TCC	2.74	453	0.23	484
TTC	2.68	462	1.32	542	569
TTT	2.54	488	1.21	574

⁽¹⁾Absorption maximum of AHAS simulated band.

The ring-opening reaction was induced by irradiation with UV light at 368 nm. According to our calculations (shown schematically in Fig. 8), this leads to the absorption of the S₃ state of SP, which may be characterized as the HOMO-1 to LUMO ¹ππ^* transition located on the benzopyran fragment. The photoexcited SP undergoes rapid non-radiative deactivation to the S₂ state (having HOMO to LUMO + 1 ¹ππ^*) and then to the S₁ state (HOMO to LUMO ¹ππ^*), in which the C-O bond is eventually broken at the MECP, returning to the ground state. Both S₂ and S₁ states exhibit a clear indene-to-benzopyran charge-transfer character. The located MECPs of S₃/S₂ and S₂/S₁ states retain the SP form, showing only slight distortions with respect to the equilibrium geometry (cf. superimposed structures in Fig. 8). The most noticeable change is the rotation of the -NO₂ moiety by 90° at the S₂/S₁ MECP. Our results are consistent with previous theoretical studies of model spiropyrans³³.

Fig. 8.

Plot of the located stationary points on the ground and excited states surfaces. The first row shows the orbital character of the S₃ state of spiropyran at the equilibrium geometry. The next row shows the geometry of the S₃/S₂ MECP, superimposed on the equilibrium geometry and orbitals that indicates the character of the S₂ state. The third row displays analogous data for the S₂/S₁ MECP. In the last row on the left, the superimposed geometries of all the above MECPs and the ground state are shown. On the right, the S₁/S₀ MECP is depicted, featuring an open structure similar to CCT.

Conclusions

We have presented a comprehensive experimental and theoretical study of the solvent-dependent photochromism in 5’-methoxy-1',3',3'-trimethyl-6-nitrospiro[chromene-2,2'-indoline], focusing on the influence of solvent polarity. Experimental results have shown that the UV-induced ring-opening and the thermally driven ring-closing reactions occurred efficiently in all tested solvents. Increasing solvent polarity led to a progressive blue shift of the merocyanine (MC) absorption maximum, confirming the presence of a negative solvatochromic effect. The influence of solvent polarity was also evident visually: before irradiation, all solutions were colorless, whereas after UV exposure, they appeared blue in dioxane (the least polar solvent), purple in dichloromethane (intermediate polarity), and pink in acetone (the most polar solvent).

The pronounced influence of solvent polarity was also evident in the kinetics of thermal relaxation. Increasing solvent polarity markedly slowed the back reaction: the ring-closing process in acetone proceeded more than twenty times slower than in dioxane. Experimental data further showed that the activation energy of the thermal ring-closing reaction rises with solvent polarity, indicating that relaxation requires greater energy in more polar environments. Ab initio calculations quantitatively reproduced these trends, demonstrating that polar solvents preferentially stabilize the merocyanine form relative to the transition state. The close agreement between experimental and theoretical activation energies provides the first direct correlation between solvent polarity and kinetic barriers in this class of photochromic molecules. Theoretical results also revealed that UV excitation at 368 nm populates the S₃ state of spiropyran, characterized by the ¹ππ* transition localized on the 2H-1-benzopyran fragment. Calculations indicate that during nonradiative deactivation, the spiropyran molecule retains its closed-ring structure through conical intersections with lower excited states, and the ring-opening occurs at the intersection with the ground state, from which the system relaxes to the most stable merocyanine isomer.

Methods

Materials and synthesis

The spiroindoline derivative (SP) was synthesized according to a three-step protocol based on the literature procedure²¹, starting from commercially available 3-methoxyphenylhydrazine hydrochloride (≥ 98%, Sigma-Aldrich), 3-methylbutan-2-one (≥ 99%, Sigma-Aldrich), and 2-hydroxy-5-nitrobenzaldehyde (≥ 98%, Alfa Aesar). All reagents were used as received without further purification. The synthesis involved the formation of an intermediate hydrazone, followed by the Fischer indole synthesis to afford the corresponding indole derivative, which subsequently underwent an intramolecular cyclization with 2-hydroxy-5-nitrobenzaldehyde under acidic conditions to yield the target SP compound. A detailed reaction scheme has been shown in Fig. 9, and the synthetic details are provided in the Supplementary Information. The overall yield of the target compound was 20%. The final product was isolated as a yellow crystalline material after recrystallization from ethanol.

Fig. 9.

Synthesis of the spiroindoline derivative (SP).

Structural characterization

¹H NMR spectra were recorded at 600 MHz using a Bruker Avance 600 spectrometer. Deuterated chloroform (CDCl₃, Sigma-Aldrich, ≥ 99.8% D) was used as the solvent. Chemical shifts (δ) are reported in parts per million (ppm) relative to tetramethylsilane (TMS) as an internal standard, and coupling constants (J) are given in hertz (Hz).

High-resolution mass spectrometry (HRMS) measurements were carried out, using a Bruker maXis Impact Q-TOF mass spectrometer (Bruker Daltonics, Bremen, Germany) equipped with an electrospray ionization (ESI) source operating in positive-ion mode. The time-of-flight (TOF) mass analyzer was used for ion selection. External calibration was performed with a sodium formate solution. Data were processed using Bruker Compass DataAnalysis 4.4 (Build 200.55) (https://www.bruker.com).

Single crystals suitable for X-ray diffraction (XRD) were obtained by slow evaporation from an ethanol solution. The crystal was examined under a polarized light microscope and mounted on a glass fibre. Diffraction data were collected at room temperature on an Oxford Diffraction four-circle single crystal diffractometer equipped with a CCD detector, using graphite-monochromatized MoKα radiation (λ = 0.71073 Å). Data reduction was performed using the CrysAlis Data Reduction program (version 1.171.42.49), and corrections for Lorentz and polarization effects were applied. The crystal structure was solved by direct methods using SHELXS-2018/3 and refined by full-matrix least-squares on F² using SHELXL-2019/2^34–36. Non-hydrogen atoms were refined anisotropically. Hydrogen atoms were located on Fourier difference maps, placed geometrically, and refined using a riding model with isotropic displacement parameters set to 1.2 × U_eq of the parent atom for CH groups and 1.5 × U_eq for methyl groups. Methyl groups were treated as rotating groups to maximize electron density during refinement.

UV–Vis spectroscopic studies of photochromic reactions

UV–Vis absorption spectra, the photochromic behavior, and reaction kinetics of the thermal ring-closing reaction of SP were studied using a Shimadzu UV 2600i spectrophotometer. The spectrophotometer was modified with a built-in light source (an electroluminescent diode) that enables initiation of the photochromic reaction directly in the apparatus. The ring-opening reaction was induced by irradiating the SP compound solution with UV light at a wavelength of λ = 368 nm. Dioxane, methylene chloride (DCM) and acetone (all analytically pure, from Stanlab and Honeywell) were used for solutions preparation. Measurements of absorption spectra, changes in the absorption spectra due to the light-induced ring-opening reaction and the thermal ring-closing reaction, and determination of the thermal relaxation reaction rate constants were carried out at ambient temperature (298 K) in solutions of 10^–5 – 10^–4 mol/L compound concentration. Kinetic measurements of the activation energy of thermal relaxation were conducted in the same apparatus, in the temperature range from 293 to 343 K; the cuvette temperature was stabilized by a thermostat connected to the spectrophotometer.

Theoretical calculations

The equilibrium geometries of the title compound and its selected merocyanine isomers were optimized using the PBE0-D3BJ/def2-TZVP method, assuming the conductor-like polarizable continuum (CPCM) solvent model³⁷ for acetone, DCM, and 1,4-dioxane. The same approach was used in the following calculations unless otherwise noted. The transition states between the selected isomers were located using the Nudged Elastic Band (NEB) approach, and the relative free energies were calculated using the DLPNO-CCSD(T)/def2-TZVPP/CPCM method³⁸, assuming geometries and thermochemical corrections calculated at the KS-DFT level. The absorption bands were simulated assuming either the Vertical Gradient (VG) or the Adiabatic Hessian After a Step (AHAS) models³⁹. All the above calculations were performed using the ORCA 6.0.1 package⁴⁰. The minimum energy crossing points (MECPs) between excited states were located at the TDDFT level in the ORCA package. In contrast, the S₁/S₀ crossing was located assuming the unrestricted spin-flip TDDFT BH&HLYP/def2-TZVP/CPCM approach and penalty-constrained optimization algorithm developed by Levine et al.⁴¹, as implemented in the Q-Chem 6.2 package⁴².

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

A.S., R.G, E.W and S.B. formed the idea of the studies, A.S. analyzed the data, wrote the paper; R.G. performed theoretical calculations and analyzed the data, wrote the part of the paper devoted to these results, contributed to editing; M.K. performed spectroscopic studies and analyzed the results, participated in writing and editing of the paper, E.W., G.S. and H.J. synthesized of the compound, performed and analyzed NMR spectra; M.D. performed the XRD experiment and analyzed the results, wrote the part devoted to them; S.B. designed the experiments, analyzed the data, and supervised the project. All authors reviewed the manuscript.

Funding

The research was funded by the own funds of the Faculty of Chemistry, Wroclaw University of Science and Technology.

Data availability

The datasets generated and analyzed during the current study are available from the corresponding authors on reasonable request.

Declarations

Competing interests

The authors declare no competing financial interests.