Introduction

Atropisomers, a class of stereoisomers arising from restricted single-bond rotation due to steric hindrance, exhibit distinct conformational stability dictated by their rotational energy barriers1. These isomers exhibit broad applications across life sciences to advanced materials. For instance, in pharmaceutical science, atropisomers function as targeted enzyme inhibitors by leveraging conformation-selective binding to modulate biological activity, with their stereochemical configuration critically influencing drug efficacy, pharmacokinetics, and bioavailability2,3. Similarly, in materials science, atropisomerism underpins the design of chiral molecules4,5 and artificial molecular motors6,7, driving asymmetric catalysis and dynamic nanotechnology8,9,10. Despite these advances, current research has predominantly focused on monoaxial atropisomeric frameworks, particularly Type I systems such as biphenyl and binaphthyl derivatives (Fig. 1a)11,12. While notable progress has been achieved in monoaxial Type II and III atropisomers13,14,15,16,17,18, the study of naturally occurring multi-axis atropisomers involving higher-order stereochemical complexity remains in its infancy10,19,20. Thus, investigating these systems is essential, as it promises to expand fundamental stereochemical theory and offers insights into the conformational dynamics governing biological and functional materials.

Fig. 1: Types of atropisomers and thermodynamic/kinetic control.
Fig. 1: Types of atropisomers and thermodynamic/kinetic control.
Full size image

a Representative types of atropisomers. b Chemical structures of 22-NB, 11-NB, and 11-NB-8DMe. c Single-crystal structures and dihedral angles of these atropisomers. d Free energy differences (ΔG) between syn and anti conformers and interconversion half-lives (t1/2). e, f Temperature-dependent NMR and syn/anti ratios of 11-NB. For clearer visualization, hydrogen atom counts are reduced based on symmetry considerations, where red represents 11-NB-syn and blue represents 11-NB-anti.

Meanwhile, the structural elucidation of atropisomers typically relies on single-crystal X-ray diffraction (XRD)21, nuclear magnetic resonance (NMR) spectroscopy22, and circular dichroism (CD)23. However, these techniques are inherently limited by requirements such as high crystallinity, well-resolved NMR signals, or strong chiroptical activity, which restrict their broader applicability. In contrast, fluorescence-based methods offer ultrasensitivity and versatility to overcome such constraints. Recent studies verify that noncovalent through-space conjugation (TSC) in sterically hindered twisted molecules modulates electronic structures via spatial orbital overlap, generating conformation-dependent fluorescence signals24,25,26,27. Moreover, leveraging the crowded topology of atropisomers, this TSC-driven fluorescence approach may emerge as a powerful alternative for structural analysis and real-time monitoring of dynamic conformational equilibria.

In this work, a series of biaxial atropisomers (namely, Type IV systems) by integrating two naphthyl rotors with a phenyl-based linker are constructed (Fig. 1a, Supplementary Figs. 1–32 and Supplementary Table 1). Through precise modulation of rotor steric hindrance, three LaPlente-classified atropisomers with different TSC features are obtained28,29. Precise control over the thermodynamic and kinetic interconversions of their conformers in 22-NB, 11-NB, and 11-NB-8DMe is also achieved, elucidating the correlation among molecular dynamics, electronic TSC, and fluorescent signals. Notably, complete separation of stable syn and anti conformers is achieved for 11-NB-2DMe, and its conformer-dependent fluorescence variations induced by TSC successfully enable quantitative analysis for resolving syn and anti populations in dilute solution. This developed fluorescent spectroscopy approach further realizes real-time visualization of crystallization-driven conformer redistribution, providing high resolution of atropisomer dynamics. Beyond expanding the multidimensional stereochemical framework of atropisomers, this work establishes a versatile in-situ methodology for mapping atropisomeric populations with high precision.

Results

Thermodynamic and kinetic control of atropisomers

The biaxial atropisomers, namely 22-NB and 11-NB, were synthesized, and their chemical structures and purity were confirmed through XRD and high-performance liquid chromatography (HPLC) analyses, respectively (Fig. 1b, c, Supplementary Fig. 30). However, NMR characterization reveals divergent properties between these structurally similar compounds. 22-NB displays a single set of well-resolved signals perfectly matching its expected structure, whereas 11-NB exhibits two distinguishable sets of peaks from its two conformers (Supplementary Figs. 8–11). To investigate the underlying mechanism, potential energy surface scans of the naphthyl rotor and phenyl linker for both 22-NB and 11-NB were performed (Supplementary Figs. 33, 34). The computational analysis reveals striking differences in their rotational profiles: 22-NB displays a relatively flat energy landscape with a maximum rotational barrier of 31.2 kJ/mol, while 11-NB exhibits two well-defined energy minima corresponding to stable syn and anti conformations, separated by a substantial rotational barrier of 88.2 kJ/mol. This pronounced energy difference provides a quantitative explanation for their distinct NMR characteristics, with a higher barrier of 11-NB enabling the observation of two stable conformers.

Quantitative analysis of the free energy and conformational transition half-life (t1/2) of the syn and anti conformers also reveals differences between these atropisomeric systems. For 22-NB, the calculated t1/2 for syn and anti interconversion are short (101 ns for anti to syn and 186 ns for syn to anti), which is consistent with its low rotational barrier (Fig. 1d and Supplementary Fig. 38). In contrast, 11-NB exhibits significantly prolonged t1/2 of 1.8 min and 6.0 min, demonstrating effective kinetic stability of both conformers. This kinetic disparity indicates that rapid single-bond rotation in 22-NB results in NMR signal averaging, while the slower conformational transition rate in 11-NB enables resolution of two distinct conformers. Furthermore, thermodynamic analysis confirms the syn conformation as energetically favored, corresponding to a larger population ratio.

Variable-temperature 1H NMR analysis of 11-NB provides direct evidence of temperature-dependent conformer equilibria. As the temperature decreased from 320 K to 240 K, an increase of the syn conformer population from 60.0% to 63.2% was observed (Fig. 1e, f), demonstrating the dynamic equilibrium toward the thermodynamically favored conformation at a low temperature. In contrast, parallel experiments on 22-NB show no observable spectral changes across the same temperature range, consistent with its rapid conformer interconversion (Supplementary Fig. 22). The dynamic interconversion of 11-NB is also realized and validated by its NMR spectra across diverse solvent environments (Supplementary Fig. 23). Expanding the investigation to a higher complexity system, a triaxial atropisomer (namely 111-NB) with three conformers could also be obtained (Supplementary Figs. 12, 13). Critically, this result demonstrates that the conformational equilibrium observed in 11-NB is not an isolated occurrence but rather a generalizable phenomenon.

To further investigate the steric hindrance effect on the biaxial atropisomers, 11-NB-8DMe with two additional methyl groups was subsequently designed. Surprisingly, this derivative exhibits a single stable conformer, contrasting sharply with the dynamic behavior of its parent compound of 11-NB (Fig. 1b, c, Supplementary Figs. 15, 16). Computational analysis of free energy and t1/2 provided mechanistic insights into this observation. Although both syn and anti conformations exist in 11-NB-8DMe, the free energy of the syn conformer is 16.3 kJ/mol higher than the anti conformer, resulting in the thermodynamically stable anti conformer (Fig. 1d, Supplementary Figs. 36, 38). This substantial energy difference originates from severe intramolecular steric hindrance between adjacent methyl groups in the syn conformer, while the anti conformer remains sterically unencumbered. Moreover, the kinetic barrier leads to more than a 700-fold disparity in interconversion rates, resulting in a rapid transition from syn to anti (t1/2 = 3.3 min) but a slow reverse transition from anti to syn (t1/2 = 1.7 day), effectively quenching the population of the syn conformer at equilibrium.

The above thermodynamic and kinetic regulation of atropisomers reveals two essential principles for achieving complete conformer separation in atropisomeric systems: (i) Minimal free energy differences to ensure thermodynamic coexistence without a significant population bias toward either isomer. (ii) Sufficient interconversion barriers to trap the conformers in their respective states, thereby suppressing thermal equilibration over experimentally relevant timescales.

Electronic TSC in atropisomers

The photophysical properties of 22-NB, 11-NB, and 11-NB-8DMe-anti were further systematically investigated (Supplementary Figs. 39–49 and Supplementary Table 3). All three atropisomers show the absorption maximum ≤290 nm, a typical feature of isolated naphthalene rings (Supplementary Figs. 39–40). To elucidate whether through-bond conjugation (TBC) or TSC dominates the photoluminescence (PL) process in biaxial atropisomer systems, three model compounds were characterized. As shown in Supplementary Fig. 45a, temperature-dependent PL spectra in dilute 2-methyltetrahydrofuran (2-Me-THF) reveal that naphthalene (Naph) exhibits intrinsic emission at 321/336 nm. The emission of 2-phenylnaphthalene (2-NB) redshifts to 358 nm relative to Naph (Supplementary Fig. 45b), attributable to extended TBC. In contrast, 1-phenylnaphthalene (1-NB) displays a blueshifted emission at 345 nm (Supplementary Fig. 45c). This shift indicates that steric hindrance between naphthalene and benzene rings in 1-NB induces a large dihedral angle, disrupting side-by-side orbital overlap and causing significant blueshift (consistent with our previous finding that twisted conformations disrupt TBC26).

In biaxial atropisomer system, 22-NB exhibits isolated Naph emission at 341 nm (Fig. 2a), showing a blueshift relative to 2-NB (358 nm). This spectral shift indicates that steric hindrance between the naphthalene rings increases their torsion angle relative to the connecting benzene ring, thereby disrupting TBC. Meanwhile, this configuration fails to establish TSC30. On the contrary, 11-NB displays an emission peak at 354 nm at 320 K (Fig. 2b), which is redshifted compared to that of 1-NB at 345 nm, suggesting the formation of TSC between two naphthalene fragments linked by the phenyl unit. Furthermore, the emission peak further redshifts to 383 nm with the decreased temperature, indicating the gradual enhancement and stable existence of TSC due to the rigid external environment upon cooling. Compared with most reported systems that require aggregation for TSC stabilization31, the single-molecule state TSC of 11-NB indicates strong intramolecular steric hindrance that stabilizes the conformation. Meanwhile, 11-NB-8DMe-anti displays a temperature-independent emission peak at 395 nm, demonstrating its highly stable and strong TSC (Fig. 2c). Furthermore, the aggregation-dependent PL studies in acetonitrile/water mixtures corroborated these findings, showing consistent behavior with temperature-variable experiments (Supplementary Figs. 43, 44)32,33. These results established a clear structure-property relationship for these atropisomers, where steric engineering controls the intensity and stability of TSC, enabling precise modulation of their luminescent wavelength.

Fig. 2: Photophysical properties and intramolecular through-space conjugation (TSC).
Fig. 2: Photophysical properties and intramolecular through-space conjugation (TSC).
Full size image

Temperature-dependent photoluminescence (PL) spectra of a 22-NB, b 11-NB, and c 11-NB-8DMe-anti in 2-Me-THF, c = 10 μM. The inset displays solution photographs captured at room temperature (RT) and 200 K under 300 nm excitation. d The excited-state geometries of these three compounds were analyzed using IGMH (isovalue = 0.018 a. u.) to characterize the interactions between the two naphthalene rings, which are highlighted within red-dashed boxes to visualize their specific locations and features. The colors represent the interaction characteristics and intensities on the isosurface around the critical points, as shown in the color scale. e The relationship between the flexibility of the molecular skeleton and TSC.

Additionally, as described above, 22-NB adopts an averaged conformation while 11-NB-8DMe-anti maintains a single stable conformation. Both compounds exhibit nearly invariant emission profiles in temperature-dependent PL spectra (Fig. 2a, c), confirming that the spectral characteristics of emission in single-conformation systems remain thermally stable under temperature perturbation. Conversely, 11-NB shows temperature-dependent conformational populations (Fig. 1e, f), with synchronized PL emission shifts occurring over identical temperature ranges (Fig. 2b). This correlation establishes fluorescence as a sensitive probe for conformational dynamics. Collectively, these findings demonstrate that PL spectroscopy has the potential to serve as an effective tool for quantifying atropisomer conformation populations—a principle substantiated in subsequent sections (Fluorescence mapping of atropisomer population and crystallization kinetics, Fig. 4).

To elucidate the origin of the observed photophysical difference in 22-NB, 11-NB and 11-NB-8DMe-anti, intramolecular interactions of their excited-state geometries using the independent gradient model based on Hirshfeld partitioning (IGMH) were performed (Fig. 2d)34,35,36. 22-NB shows a complete absence of intramolecular interactions and TSC, which is consistent with its temperature-independent and short-wavelength emission from isolated naphthalene. However, the IGMH surface of 11-NB (both syn and anti conformers) and 11-NB-8DMe-anti exhibit clear evidence of nonconvalent intramolecular interactions37, which aligns closely with their redshifted emission from intramolecular TSC between two naphthalene units.

Accordingly, since 22-NB is identified as a Class 1 atropisomer under the LaPlente classification (ΔG < 84 kJ/mol) with a low rotational barrier that permits unrestricted 360° rotation between its naphthalene and phenyl units29, its pronounced flexibility markedly destabilizes the transient TSC (Fig. 2e). In contrast, 11-NB falls into Class 2 atropisomer (84 kJ/mol <ΔG < 117 kJ/mol), where slowed interconversion between syn and anti conformations enables temperature-responsive and switchable TSC. Further restricting molecular motion by methyl groups in 11-NB-8DMe-anti yields a rigid skeleton, facilitating the formation of exceptionally stable TSC38. Together, these results establish a direct correlation between single-molecule dynamics and electronic TSC in the solution state, providing critical mechanistic insights into the unexplored TSC behavior of atropisomers.

Separation and typical features of syn and anti conformers

Following the above principles for achieving conformer separation in atropisomeric systems, 11-NB-2DMe was precisely designed for isolating and studying its syn and anti conformers (Fig. 3a, Supplementary Figs. 6, 7). Compared to 11-NB-8DMe, the changed substitution positions of adjacent methyl groups decreases the steric hindrance of the syn conformer. As a result, 11-NB-2DMe satisfies the two essential criteria for realizing complete separation: (i) minimal free energy difference between conformers (ΔGsyn-anti = −2.8 kJ/mol, Supplementary Fig. 38); (ii) slow interconversion kinetics, with t1/2 of 4464 years for syn and 1442 years for anti conformer, respectively, qualifying it as Class 3 atropisomer (ΔG > 117 kJ/mol) under the LaPlente classification. Thus, the isolation of pure syn and anti conformers via fractional crystallization is successfully achieved. XRD, NMR, and HPLC analyses confirm the high purity and stereostructure of the obtained 11-NB-2DMe-syn and 11-NB-2DMe-anti conformers (Supplementary Figs. 18–21, 31). Photophysical studies reveal that while the syn and anti conformers exhibit nearly identical absorption spectra with characteristic peaks at 221 and 285 nm, their PL behavior differed (Fig. 3b, c). 11-NB-2DMe-syn displays dual emission, where the structured bands at 330 nm and 342 nm originate from the isolated naphthalene ring, and another broad long-wavelength peak around 372–383 nm is attributed to TSC. Upon cooling, the intensity of the TSC emission increases significantly due to the restricted intramolecular motions. Meanwhile, 11-NB-2DMe-anti exhibits dominant TSC emission at 378 nm, which is of significantly greater intensity than its residual naphthalene emission at 330 nm.

Fig. 3: Separating and characterizing syn and anti conformers of 11-NB-2DMe.
Fig. 3: Separating and characterizing syn and anti conformers of 11-NB-2DMe.
Full size image

a Chemical structures, single-crystal structures, and calculated free energies and half-lives of 11-NB-2DMe. Absorbance and temperature-dependent PL spectra of b 11-NB-2DMe-syn and c 11-NB-2DMe-anti in 2-Me-THF, c = 10 μM. d, e Plots of δginter versus sign(λ2)ρ. The inset displays the optimized excited-state geometry of 11-NB-2DMe, where interactions between its two naphthalene rings, analyzed by IGMH (isovalue = 0.018 a.u.), are circled by red dashed lines. f, g Plots of reorganization energy versus normal mode wavenumber of 11-NB-2DMe in the gas phase. Inset: Proportions of bond length, bond angle, and dihedral angle contributing to total reorganization energy. h, i Schematic diagram of the photophysics of 11-NB-2DMe atropisomers. Excited-state electronic structures are analyzed using the hole-electron method.

Although IGMH analysis reveals noncovalent interactions between the two naphthalene rings (green regions) in both 11-NB-2DMe-syn and 11-NB-2DMe-anti, the anti conformer exhibits more extensive interaction regions than its syn counterpart, consistent with the enhanced TSC emission observed in the PL spectra of the anti conformer (Fig. 3d, e). Furthermore, the reorganization energy (λ), which functions as a quantitative measure of inherent geometric changes upon photoexcitation and the contribution of intramolecular motion to non-radiative decay, was calculated39. For 11-NB-2DMe-syn, it exhibits a substantially high total λ of 6130 cm−1 compared to 11-NB-2DMe-anti of 3190 cm−1. Notably, the dihedral angle torsional motions (yellow region), dominating the non-radiative relaxation process, contribute 63.85% and 51.41% to the total reorganization energy for the syn and anti conformers, respectively (Fig. 3f, g). Interestingly, 11-NB-2DMe-syn exhibits a pronounced butterfly-like swinging motion that substantially contributes to its larger λ (Supplementary Fig. 50), a feature absent in 11-NB-2DMe-anti. These results prove the flexible skeleton and strong intramolecular motions upon photoexcitation of the syn conformer, resulting in its unstable TSC and short-wavelength emission.

Unlike conventional H- and J-aggregates40, the intramolecular dimers of naphthalene in 11-NB-2DMe preserve the ground-state electronic structure of isolated naphthalene (evidenced by unchanged absorption peaks) while significantly altering excited-state properties. Hole-electron analysis based on excited-state geometries confirms the presence of orbital overlap between naphthalene rings in both syn and anti conformers of 11-NB-2DMe (Supplementary Fig. 51)41, indicating the existence of TSC. However, these conformers exhibit markedly different intensity ratios of emission from TBC and TSC, revealing the mutually competing radiative transition channels of TBC and TSC. Specifically, face-to-face H-dimer stacking suppresses the low-level radiative transition of TSC, whereas staggered J-dimer stacking enhances its radiative efficiency (Fig. 3h, i, Supplementary Fig. 52). Thus, the emission intensity ratios between TBC and TSC, which are strongly modulated by naphthalene stacking patterns, could be a potential ultra-sensitive approach for differentiating two conformers and even mapping their populations in complex environments.

Fluorescence mapping of atropisomer population and crystallization kinetics

For 11-NB-2DMe, the separation and purification of its syn and anti conformers pose significant challenges. Although fractional crystallization successfully achieves isolation of these conformers, the underlying crystallization kinetics remain poorly unexplored42,43,44,45. Direct mapping of conformers and thus visualizing the crystallization dynamics may provide mechanistic insights into atropisomeric formation and separation.

To verify the ability of fluorescence signals to differentiate atropisomeric conformers, the PL spectra of 11-NB-2DMe with varying syn/anti ratios is characterized (Fig. 4a). A near-perfect linear correlation (R2 = 0.9808) between the relative PL intensity (I365/I342) and syn/anti ratio is observed, where I365 and I342 represent characteristic emission intensity of 11-NB-2DMe-anti (365 nm) and 11-NB-2DMe-syn (342 nm) conformers, respectively (Fig. 4d). This robust correlation ensures ratiometric fluorescence as a powerful real-time probe for mapping populations of atropisomeric conformers during crystallization processes. Hence, to qualitatively probe the crystallization kinetics, absorption spectra of 11-NB-2DMe-syn and 11-NB-2DMe-anti are acquired in dilute solutions across a concentration gradient (Fig. 4b, c). Both conformers exhibit strict adherence to the Lambert-Beer law (R2 > 0.99) at 285 nm within the tested concentration range (Fig. 4e, f), validating absorbance spectroscopy as a quantitative method for determining solute concentrations during crystallization.

Fig. 4: Visualizing crystallization kinetics of 11-NB-2DMe.
Fig. 4: Visualizing crystallization kinetics of 11-NB-2DMe.
Full size image

a PL spectra of 11-NB-2DMe in acetonitrile (ACN) with varying syn and anti fractions, c = 10 μM. Absorption spectra of b 11-NB-2DMe-syn and c 11-NB-2DMe-anti in ACN with different concentrations. d Plot of relative PL intensity (I365/I342) versus syn fraction in panel (a). Plots of absorbance of e 11-NB-2DMe-syn and f 11-NB-2DMe-anti at 285 nm versus concentration in panels (b, c). g PL spectra of 11-NB-2DMe during its crystallization in ACN at various crystallization times. Plots of h population ratios and i mass (m) of syn and anti conformers during the crystallization. j Schematic diagram of three stages of 11-NB-2DMe crystallization.

Based on the above results, the complete crystallization process of 11-NB-2DMe is further monitored in its entirety (Supplementary Fig. 53). Fluorescence analysis shows that the I365/I342 ratio remains stable at ~0.69 during the initial phase (0–3 h). Subsequently, this ratio exhibits a progressive increase over time, reaching 1.06 at 24 h (Fig. 4g, Supplementary Table 4a, b). Combined with absorption measurements that reflect solution concentration, these optical analyses enable quantification of the real-time relative populations and mass for two specific conformers in solution. This integrated approach reveals three distinct kinetic stages throughout the crystallization process (Fig. 4h, i). (1) During the initial stage (0−3 h), the solution volume decreases without changes in conformer populations or total solute mass of 11-NB-2DMe-syn and 11-NB-2DMe-anti, indicating solvent evaporation leading to saturation prior to nucleation. (2) The second stage (3−21 h) displays a selective reduction in the syn-conformer population concurrent with accumulation of the anti conformer in solution, a dynamic driven solely by preferential crystallization of syn conformers. During this stage, the syn-nuclei formation and crystal growth generated harvestable high-purity 11-NB-2DMe-syn crystals. (3) Upon reaching the 21-hour mark, the third crystallization stage represents concurrent crystallization of both conformers until process completion, as evidenced by simultaneous mass reduction of their solutes. The third stage leads to mixed crystal growth, precluding high-purity isolation of 11-NB-2DMe-anti due to inseparable blending of syn- and anti-crystalline domains. Reproducibility was confirmed through replicate experiments (Supplementary Figs. 54, 55).

Throughout the sequential crystallization process from concentration (stage I), selective crystallization (stage II), and competitive crystallization (stage III), ratiometric fluorescence approach enables precise and real-time mapping of syn/anti conformer populations in 11-NB-2DMe (Fig. 4j). This non-destructive nature and molecular-level sensitivity of fluorescence mapping provides complete visualization of crystallization kinetics, establishing it as a powerful tool for mapping complex crystallization processes in atropisomeric systems.

Therefore, this study successfully constructs and characterizes a series of biaxial atropisomers using naphthalene rotors and phenyl linkers, focusing on their complex stereoconformations and dynamic interconversion mechanisms. Through precise structural modulation, three LaPlente-type atropisomers are synthesized with tailored thermodynamic and kinetic properties. The correlation between molecular dynamics and TSC is elucidated, demonstrating that TSC-induced fluorescence variations enable real-time mapping of atropisomer populations. Notably, the syn and anti conformers of 11-NB-2DMe are completely separated and characterized. Fluorescence mapping provides insights into the dynamic conformational population during crystallization, revealing temporally sequential crystallization where syn conformers undergo preferential nucleation during the initial stage, while anti conformers exhibit delayed crystallization behavior. This work significantly advances atropisomer science by establishing a fluorescence-enabled in-situ platform for real-time mapping of atropisomeric dynamics, thereby deepening fundamental stereochemical theory and opening possibilities for studying molecular dynamics in diverse matter and biological systems.

Methods

Materials

All reagents were purchased from commercial suppliers (Leyan, Energy Chemical, Macklin, Aladdin, J&K Scientific, or Bidepharm) and used without further purification. Specifically, tetrakis(triphenylphosphine)palladium was obtained from Leyan (Cat No.1021232, Leyan, Shanghai, China). All the final products used in experiments were purified through a silica gel column and recrystallization at least three times. Purification of all samples was checked by high-performance liquid chromatography (HPLC). HPLC-grade acetonitrile (ACN), toluene (Tol), and 2-methyltetrahydrofuran (2-Me-THF) were used for synthesis and photophysical measurements.

Instrumentation and analytical techniques

1H and 13C NMR spectra were recorded on Bruker AVANCE NEO 400 MHz and 500 MHz spectrometers. Deuterated solvents, including CD2Cl2, THF-d8, CD3CN, DMSO-d6, and CDCl3, were used without further purification. Variable-temperature 1H NMR spectra were acquired on a Bruker AVANCE NEO 600 MHz instrument. UV-Vis absorption spectra were measured using a Shimadzu UV-2600 spectrophotometer. Photoluminescence spectra, absolute fluorescence quantum yields, and fluorescence lifetimes were obtained on a Shimadzu RF-6000 spectrofluorophotometer coupled with an Edinburgh Instruments FLS1000 spectrometer. Single crystals of the target compounds were mounted directly on a Bruker D8 Venture diffractometer. Data collection was performed at 250.00 K. Structures were solved using ShelXT 2018/2 (dual-space algorithm) and refined with ShelXL 2019/2 (full-matrix least-squares on F2) via the Olex2 1.5 interface. HPLC measurements were conducted on a Waters 2690 separations module with a C18 reverse-phase column. High-resolution mass spectra of these compounds were all obtained through fourier transform mass spectrometry (FTMS), and test instrument is Thermo Scientific Exactive GC Orbitrap (Thermo Scientific).

Computational details

All the compounds were fully optimized with the density functional theory (DFT) method by using the B3LYP density functional and 6–31 G(d,p) basis set46 London-dispersion effects were also taken into consideration using Grimme’s DFT-D3BJ correction47,48 Analytical frequency calculations were also performed at the same level of theory to confirm that the optimized structures were at a minimum point. Time-dependent density functional theory (TD-DFT)49 was utilized at the same level of theory to calculate optimized excited singlet-state (S1) geometries and energy levels. All the above quantum chemical calculations were carried out using the Gaussian 16 program50. Shermo was employed to calculate Gibbs free energy extraction from Gaussian-derived checkpoint files (.fchk), applying ZPVE scaling (factor = 0.9806) at 298 K51. The Eyring equation

$${k}=\kappa \frac{{{k}}_{{B}}{T}}{{h}}{{e}}^{-\frac{{\Delta {G}}^{\ne }}{{RT}}}$$

defines the reaction rate constant k through transition state theory, where κ is the transmission coefficient (typically unity), kB is the Boltzmann constant, T is the absolute temperature in Kelvin (298 K in this work), h is the Planck’s constant, ΔG≠ is the Gibbs free energy of activation in kJ/mol, and R is the universal gas constant. For first-order reactions, the half-life follows as

$${t}_{1/2}={{{\mathrm{ln}}}}(2)/k$$

Reorganization energy analysis based on normal-mode-projected displacements and Duschinsky rotation matrices under curvilinear coordinates was performed using the Molecular Materials Property Prediction Package (MOMAP) based on the formatted checkpoint file (.fchk) obtained from Gaussian39,52,53. IGMH analyses34 were calculated by Multiwfn 3.835, and were displayed using Visual Molecular Dynamics (VMD)54 (Isovalue: 0.018). Hole-electron analyses41 were calculated by Multiwfn 3.8, and were displayed using IQmol molecular viewer package (Isovalue: 0.001).