Tunable Emission Properties of Indolizine-Based Aggregation-Induced Emission Luminogens for White-Light Emission

Abstract

Unlike conventional fluorophores experiencing aggregation-caused quenching (ACQ) in the solid state, aggregation-induced emission (AIE) luminogens (AIEgens) exhibit enhanced fluorescence upon aggregation. This property renders AIEgens promising for various technological applications. However, developing red-emitting AIEgens with high fluorescence efficiency remains challenging due to structural constraints. Compact red-light-emitting AIEgens hold great potential, as their small, simple fluorescent frameworks enable facile modification and precise color tuning. This study presents a novel compact indolizine-based AIE system with solid-state emission tunable from cyan (496 nm) to deep red (669 nm), with an average molecular weight (MW) of 312.8 g/mol. The deep-red-emitting derivative is uncommon among short-conjugation, low-molecular-weight (∼280 g/mol) systems and demonstrates exceptional photophysical characteristics. Through carbonyl functional groups, we modulated molecular packing, resulting in distinct optical properties and emission tunability in solid and aggregated states. Structural investigations using single-crystal X-ray diffraction (SCXRD) and computational modeling revealed that molecular packing and intermolecular interactions significantly influence emission behavior. Integration of these AIEgens into blue-light-emitting diode chips facilitated white-light production with CIE coordinates (0.360, 0.362) and a color temperature of 4533 K. These findings underscore the potential of indolizine-based AIEgens for next-generation emissive materials, providing insights for their rational design and broader technological applications.

1. Introduction

Solid-state emitting fluorescent materials have garnered significant attention in recent years due to their remarkable properties and wide-ranging applications. These substances exhibit enhanced photoluminescence efficiency and stability in solid form, rendering them exceptionally suitable for integration into various advanced technological applications. These materials function as highly sensitive and specific fluorogenic bioprobes in biological imaging and detection, − enabling researchers to monitor cellular processes and identify specific biomolecules , with unprecedented precision. Furthermore, their distinctive optical properties have proven crucial in the development of optoelectronic devices, such as organic light-emitting devices, , photovoltaic cells, , and chemical sensors. ,

Among these applications, white-light-emitting diode (WLED) have emerged as particularly promising due to their high efficiency, low power consumption, and environmental sustainability. − White light can be generated through various methodologies, including the combination of red, green, and blue (RGB) primary colors or the integration of complementary hues such as yellow and blue. ,− Consequently, substantial research efforts have been dedicated to developing single-color RGB fluorescent materials and tunable fluorescence systems capable of producing variable color outputs.

Despite these advancements, numerous conventional fluorophores exhibit aggregation-caused quenching (ACQ), in which strong intermolecular interactions in the aggregated state significantly reduce fluorescence efficiency, particularly in the solid state environments. ,,− Addressing this limitation necessitates the development of alternative molecular systems capable of maintaining robust fluorescence in both aggregated and solid-state environments. Aggregation-induced emission (AIE) luminogens (AIEgens) present a promising solution by exhibiting enhanced fluorescence upon aggregation. Unlike conventional fluorophores, which frequently experience ACQ, AIEgens undergo restriction of intramolecular motion (RIM), including rotation (RIR) and vibration (RIV), because molecular packing constrains their dynamic movements in the aggregated state. − By suppressing these nonradiative decay pathways, AIEgens exhibit strong solid-state emission, distinguishing them from traditional fluorophores. Since Tang et al. introduced pentaphenylsilole as an AIE-active luminogen in 2001, AIE-based materials have been extensively investigated for a diverse range of applications, including their incorporation into polymers, , metal–organic frameworks, , and small molecules. , Their versatility extends across multiple scientific disciplines, demonstrating significant utility in optoelectronics, , biomedical imaging, , chemical sensing, − and stimuli-responsive applications.

Among AIEgens, achieving efficient red emission is particularly crucial for both optoelectronic and biomedical applications. Red-emissive AIE materials are essential for producing high-quality white light for lighting and display technologies. − Furthermore, their long-wavelength fluorescence offers substantial advantages in biomedical imaging by reducing background interference and tissue damage. , Nevertheless, the creation of red-emissive materials presents a significant challenge due to inherent structural limitations. Long-wavelength emission typically requires narrowing the energy gap (E _g ) between the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) through π-conjugation extension or donor–acceptor interactions. − Although these strategies effectively shift emission to the red spectrum, they also promote strong π-π stacking or dipole–dipole interactions, frequently leading to ACQ and diminished fluorescence efficiency in solid form. − Even when red emissive AIEgens are successfully designed to counteract ACQ, their complex structure and large molecular size often make synthesis difficult, hindering large-scale production. Additionally, the inherent hydrophobicity of these compounds requires structural alterations to improve water solubility for biological applications. , Consequently, red-light-emitting AIEgens with compact molecular structures exhibit great potential. Their small and structurally simple fluorescent frameworks offer significant advantages for the development of novel optoelectronic devices, owing to their facile structural modification and straightforward synthesis, thereby enabling precise color tuning. Moreover, these compounds exhibit improved solubility and biocompatibility, making them promising candidates for biomedical applications. ,,

To address these challenges, we present novel indolizine-based AIEgens with a structurally simple design and tunable emission ranging from cyan (497 nm) to deep red (669 nm) in both aggregated and solid states, despite their small size, with an average molecular weight (MW) of 312.8 g/mol (Figure ). Through the attachment of phenyl groups to the C3 position of the indolizine core, we synthesized two derivatives, 2 and 3, which exhibit tunable aggregation-induced fluorescence. They form cyan (496 nm) and orange (603 nm) aggregates, respectively, as f _w increases. The AIE mechanism is governed by the interplay between RIR and intramolecular charge transfer (ICT). RIR enhances fluorescence by limiting molecular movement in the solid state, while ICT modulates AIE intensity through charge-transfer dynamics. Further modifications of 3, achieved through the addition of various carbonyl functional groups, yielded additional derivatives (4, 6, 13, and 17) with distinct optical properties in the solid and aggregated states. Although these derivatives demonstrated similar absorption (∼430 nm) and emission (∼590 nm) wavelengths in tetrahydrofuran (THF) solution, their solid-state emissions (spanning from 559 to 669 nm) varied significantly. Notably, compound 13 exhibited deep-red emission at 669 nm in the solid state, an uncommon characteristic among short-conjugation, low-molecular-weight (∼280 g/mol) systems, underscoring its exceptional photophysical properties. Single-crystal X-ray diffraction (SCXRD) and computational modeling further revealed that these structural alterations influenced molecular packing, with distinct distances of π-π interactions, molecular alignment and dipole–dipole interactions in the excimer ultimately determining their fluorescence behaviors in both aggregated and solid states. To demonstrate the practical applications of these materials, a functional WLED was fabricated by incorporating a mixture of AIEgens into commercially available blue-light-emitting diode chips. The facile synthesis, compact size, and tunable emission render indolizine-based AIEgens highly promising for diverse applications, including optoelectronics, bioimaging, and sensing, as well as other emerging fields.

1.

Study of Indolizine-based AIEgens with their photos. Schematic illustration of the fluorescence properties in the solid state and the AIE mechanisms of indolizine-based AIEgens presented in this paper, along with their applications in white-light emission.

2. Results and Discussion

2.1. Synthesis of Phenyl-Subsituted Indolizine-Based AIEgen

The synthetic pathway for the indolizine-based AIEgens investigated in this study is illustrated in Scheme . The acetyl- and nitro-substituted phenyl-indolizine derivatives 2 and 3 were synthesized via a Suzuki–Miyaura coupling from the previously reported indolizine 1. Subsequently, the ester group of 3 was converted to an aldehyde (4) using a one-pot synthesis as reported by Jeon et al., and further hydrolyzed to obtain carboxylic acid 5 through a standard saponification procedure with aqueous NaOH. The cyclohexyl ester 6 was then synthesized from 5 via a Steglich-type esterification. For the synthesis of compounds 13 and 17, we commenced with dihydroindolizine 7, which was also previously reported. Compound 7 was first hydrolyzed to yield carboxylic acid 8, which was subsequently converted to the Weinreb amide 9. A Grignard reaction with 9 yielded ketone derivatives 10 and 14. These intermediates were subjected to regioselective bromination using N-bromosuccinimide (NBS), followed by Suzuki–Miyaura coupling to produce intermediates 12 and 16. Finally, oxidative dehydrogenation using manganese dioxide (MnO₂) afforded the target AIEgens 13 and 17.

1. Synthetic Pathway for the Preparation of Indolizine-Based AIEgen.

2.2. Photophysical Properties of Synthesized AIEgen

In our investigation into the development of novel compact and low-molecular-weight AIEgens with tunable solid-state emission, we synthesized and analyzed a series of indolizine-based fluorophores, which resulted in the identification of their distinctive AIE properties and emission tunability. Specifically, 2 and 3 were synthesized through the introduction of phenyl groups at the C3 position of the indolizine core (Figure a,b). Their structures were confirmed utilizing ¹H and ¹³C NMR spectroscopy and mass spectrometry, and comprehensive synthetic procedures, characterization data, and spectra are provided in the Supporting Information. To elucidate the AIE properties of these compounds, we analyzed 2 and 3 in water/acetonitrile (ACN) mixed solvents with varying water fractions (f _w ) from 0% to 90%. As shown in Figure a, 2 exhibited intense cyan fluorescence in pure ACN. However, as f _w increased to 80%, its photoluminescence (PL) intensity gradually decreased, and its emission wavelength red-shifted from 495 to 550 nm. At 90% f _w , aggregation occurred, resulting in a fluorescence intensity increase and a blue-shifted emission at 496.4 nm (Table S2). This observation is supported by the leveling-off of the tail of the absorption spectrum of 2 in the visible region at 90% f _w (Figure S1), indicating aggregate formation. Dynamic light scattering (DLS) analysis confirmed the occurrence of aggregation, revealing an average particle size of approximately 1.14 μm (Figure S2).

2.

AIE properties of 2 and 3. Molecular structure and their photographs of the powders under 365 nm UV light (top) and emission spectra in ACN/water mixed solutions and their photographs under 365 nm UV light (bottom) of (a) 2 and (b) 3. (c) PL intensities of 2 and 3 at their maximum emission wavelengths at 90% f _w in different ACN/water mixed solutions.

In contrast, 3 exhibited negligible fluorescence emission in pure ACN until f _w reached 70%. At 80% f _w , aggregation occurred, leading to orange fluorescence under 365 nm UV light (Figure b). The emission wavelength of 3 increased from 596 nm at 80% f _w to 603 nm at 90% f _w . Similar to 2, the tails of the absorption spectra of 3 at 80% and 90% f _w exhibited leveling-off in the visible region, confirming the formation of aggregates (Figure S1). DLS analysis corroborated this observation, revealing an average aggregate particle size of approximately 409 nm at 90% f _w (Figure S2). Additionally, the AIE enhancement factor (α_AIE), calculated using the formula α_AIE = I/I ₀, where I is the emission intensity at 90% f _w and I ₀ is the emission intensity in pure ACN, significantly increased to 117.4 (Figure c). By comparison, 2 showed a much lower α_AIE value of 0.04 under similar conditions. This difference is attributed to the strong intrinsic fluorescence of 2 in pure ACN, which reduces its relative enhancement upon aggregation.

Beyond differences in α_AIE values, 3 exhibited a significant bathochromic shift (∼100 nm) in its fluorescence emission in the aggregated state compared to 2. While 2 displayed a solid-state emission at 497 nm (cyan emission), 3 exhibited a red-shifted emission at 605 nm (orange-red emission) (Figure a,b and S3). This pronounced bathochromic shift is attributed to the substitution of the acetyl group on the phenyl ring of the indolizine scaffold with a nitro group, which significantly influences the photophysical properties of the compound due to the differential electron-withdrawing characteristics of each functional group. , To elucidate the electron-withdrawing substituent effect, underlying these observations, the solvatochromism of 2 and 3 was investigated. A characteristic feature of donor–acceptor AIE luminogens is their reduced PL intensity in solvents, which is frequently influenced by ICT processes resulting from the charge transfer between the donor and acceptor. − As f _w increased from 0% to 80%, the PL emission of 2 red-shifted and decreased in intensity (Figure a), suggesting that the fluorophore is sensitive to changes in solvent polarity. This behavior indicates the potential involvement of ICT within the phenyl-substituted indolizine fluorophore. To validate this, we examined the solvatochromic properties of 2 and 3 in six solvents of varying polarities (Figure a and Table S1). 2 exhibited bright blue fluorescence at 460 nm in ether, but in methanol (MeOH), a more polar solvent, its intensity sharply decreased, and the emission shifted to 483 nm. This fluorescence quenching and bathochromic shift with increasing solvent polarity strongly suggest that 2 undergoes ICT. This solvent-dependent emission shift aligns with the previously observed blue-shift in the aggregated state, where the emission shifted from 80% f _w (551 nm) to 90% f _w (496 nm) (Figure a). Aggregate formation likely creates a hydrophobic environment that suppresses ICT, altering its photophysical properties and leading to hypsochromic emission shifts.

3.

Investigation of photophysical properties and the AIE mechanism. (a) Emission spectra of 2 and 3 in ether, THF, dichloromethane (DCM), dimethyl sulfoxide (DMSO), ACN, and methanol (MeOH) (c = 20 μM) (left); molecular orbital distributions and E _g of S₀ and S₁ obtained by TD-DFT cacluations using B3LYP/6-311++G** at gas phase (right). (b) Emission spectra of 2 (left) in different glycerol/MeOH fractions (f _g/M ) and 3 (right) in different glycerol/DMF fractions (f _g/D ). (c) Schematic illustration of the AIE mechanism of indolizine-based AIEgens.

3 also demonstrated solvatochromic behavior, emitting yellow fluorescence at 556 nm in ether and orange fluorescence at 590 nm in THF. Interestingly, no fluorescence was detected in the relatively nonpolar solvent dichloromethane (DCM) (Figure a). This distinct solvent-dependent behavior compared with that of 2 suggests that 3 exhibits a stronger ICT effect, attributed to the nitro group’s greater electron-withdrawing ability compared to the acetyl group. The stronger electron-withdrawing influence facilitates charge separation and enhances charge transfer between donor and acceptor components, intensifying ICT.

To validate the ICT behavior of 2 and 3, we performed density functional theory (DFT) calculations. As shown in Figure a, the electron densities of the S₀ in both compounds are predominantly localized on the indolizine core, whereas those of the S₁ shift to the phenyl substituent. This electron distribution aligns well with the experimental solvatochromic behavior, confirming the occurrence of ICT in the phenyl-substituted indolizine framework. The calculated E _g were 2.26 eV for 2 and 1.53 eV for 3, with the narrower E _g of 3 correlating with its red-shifted emission in both solution and the aggregated state. Moreover, the larger separation of the S₀–S₁ electron densities in 3 than in 2 suggests a significant enhancement in ICT strength. This increased separation facilitates charge transfer, which is consistent with the stronger solvent-dependent behavior of 3 observed experimentally. Enhanced charge transfer allows the molecule to react more sensitively to the polarity of its environment, resulting in a turn-off property under relatively nonpolar solvent conditions. Therefore, the stronger ICT in 3 likely contributes to its higher α_AIE compared with 2 by quenching its fluorescence emission in pure ACN.

Our solvatochromism study and DFT calculations collectively demonstrate that the introduction of the nitro group in 3 reduces the S₀–S₁ gap compared to 2 due to its strong electron-withdrawing effect. This narrowing of the energy gap leads to a pronounced bathochromic shift in fluorescence emission in both solution and solid states. Additionally, the stronger ICT effect in 3 increases its sensitivity to environmental polarity, facilitating fluorescence quenching in relatively nonpolar solvents and contributing to its turn-off behavior. The enhanced ICT in 3 also plays a crucial role in its higher α_AIE compared to 2, as fluorescence suppression in solution enhances its relative emission in the aggregated state. These findings underscore the importance of functional-group modifications in fine-tuning ICT strength and emission tunability, enabling precise control over both solvent-dependent fluorescence and AIE properties.

Various mechanisms, such as RIR and RIV, are known to induce AIE behavior. In solution, molecular structures with rotational or vibrational freedom facilitate the nonradiative stabilization of the excited state. However, upon aggregation, these motions become restricted, reducing nonradiative relaxation and leading to enhanced fluorescence emission. In our indolizine framework, the phenyl group is capable of free rotation, suggesting that the turn-on fluorescence characteristics of the aggregates primarily arise from the RIR mechanism. To test this hypothesis, we examined the fluorescence behavior of 2 in glycerol/MeOH mixed solutions with varying glycerol fractions (f _g/M ) because increases in viscosity restrict rotational motion. The solutions were excited at the maximum absorption wavelengths observed in MeOH, and the emission intensity and wavelength data are summarized in Table . In the case of 2, a red-shift of 47 nm was observed as f _g/M increased (Figure b and Table ). While the higher polarity of glycerol than that of MeOH likely induced ICT phenomena, resulting in red-shifted emission, the fluorescence intensity of 2 unexpectedly increased 7-fold in the mixed solutions compared with that in pure MeOH.

1. Viscochromism Study of 2 and 3 in Different Glycerol Fractions.

2			3
f g/M  [%]	λem [nm]	Int.	f g/D  [%]	λem [nm]	Int.
0	483.2	0.14	0	n.a.
10	493.6	0.12	10	n.a.
30	499.1	0.16	30	609.2	0.32
50	512.2	0.25	50	626.1	0.50
70	523.4	0.47	70	614.2	0.70
90	530.5	0.95	90	623.9	1.00
100	530	1.00	100	n.a.

^aGlycerol fraction in the glycerol/MeOH solution.

^bMaximum emission wavelength.

^cRelative PL intensity compared with the emission at 100% f _g/M .

^dGlycerol fraction in the glycerol/DMF solution.

^eRelative PL intensity compared with the emission at 90% f _g/D .

The fluorescence behavior of 3 was investigated in glycerol/DMF mixed solutions with varying glycerol fractions (f _g/D ) due to its poor solubility (Figure b and Table ). The solutions were excited at the maximum absorption wavelengths observed in DMF (Figure S4). Since 3 exhibits significant sensitivity to environmental polarity, the overall emission intensity was negligible, making it difficult to observe the gradual red-shift behavior. Nevertheless, as the f _g/D increases, an enhancement in PL intensity can be observed even in a polar solvent environment (Table ). This viscochromic behavior strongly supports the RIR mechanism, underscoring its significant contribution to the AIE fluorescence observed in the indolizine-based framework.

Taken together, the results thus far indicate that the AIE property in the indolizine-based framework arises from the complementary effects of RIR and ICT processes. When in solution, the rotation of phenyl groups and ICT contribute to a decrease in fluorescence intensity. In contrast, the solid state experiences enhanced fluorescence intensity due to RIR, which restricts nonradiative decay, while ICT modulates AIE strength by altering charge-transfer dynamics (Figure c). This dual contribution underscores the complementary roles of structural constraints and electronic properties in determining the photophysical behavior of these molecules, offering valuable insights for molecular design.

2.3. Color Tunability in Solid State via Functional Group Modification

Building on the favorable AIE properties of 3, we sought to explore its structural modifications to further optimize its photophysical properties. Specifically, we replaced the ester group at the C7 position of the molecule with other electron-withdrawing groups, such as acetyl and aldehyde, to investigate their effect on its fluorescence behavior. To evaluate these effects, we synthesized two derivatives, 4 and 13, bearing an aldehyde group and an acetyl group, respectively, instead of the ester group (Figure a).

4.

Investigated photophysical properties of the indolizine-based AIEgens 3. (a) Molecular structures (left), emission spectra in ACN/water mixed solutions (bottom), and photographs under 365 nm UV light (right) of 4, 13, 6, and 17. (b) Emission spectra of the derivatives in the solid state (left) and their photographs under 365 nm UV light (right). (c) Emission spectra of the derivatives in 20 μm THF and photographs of 3 and 17 under 365 nm UV light. (d) Molecular orbital distributions and calculated S₀–S₁ E _g of 3, 4, 13, 6, and 17 at gas phase.

We evaluated the AIE and ICT behaviors of 4 and 13 by monitoring their fluorescence emissions in both aggregated and solution states, following the previously described procedure. As shown in Figure a, both derivatives exhibited negligible fluorescence in pure ACN or at low f _w . As f _w increased, aggregation was induced and the fluorescence emission intensity gradually increased, consistent with the behavior of 3. The maximum emission wavelengths of 4 and 13 are 651 and 657 nm, respectively. In the solid state, both compounds exhibited deep-red fluorescence under 365 nm UV light, with emission wavelengths of 660 nm for 4 and 669 nm for 13 (Figure b). To the best of our knowledge, such deep-red emissions from short-conjugation systems with low molecular weights (4, 266.26 g/mol, 13, 280.28 g/mol) are extremely rare. Interestingly, although the aggregated-state emissions of 4 and 13 were significantly red-shifted compared with that of 3 (λ_aggregate: 3, 603 nm; 4, 651 nm; 13, 657 nm), their photophysical properties in the solution state were nearly identical (Figure S6). In THF solution, all three compounds exhibited similar maximum absorption (∼430 nm) and emission (∼590 nm) wavelengths (Figure c and Figure S7). This finding indicates that replacing the ester group in 3 with stronger electron-withdrawing groups, such as aldehyde (in 4) or acetyl (in 13) groups, has little to no effect on the S₀–S₁ E _g of the indolizine core in solution. To demonstrate this phenomenon further, we performed DFT calculations on the S₀–S₁ E _g values at gas phase of all three compounds. As shown in Figure d, the calculated E _g values averaged at 1.57 eV. This consistency indicates the uniform absorption and emission wavelengths of the compounds in solution. Therefore, changing the electron-withdrawing characteristics of the carbonyl group has a negligible effect on the emission wavelengths of the solution state.

To elucidate the unique red-shifting behaviors of these molecules in the solid state, we investigated fluorescence emission mechanisms and molecular packing effects. Several recent studies have reported correlations between fluorescence emission and molecular states such as aggregation and crystal packing. For instance, a previous study demonstrated that a fluorophore can adopt multiple packing conformations during crystal growth or Z/E isomerization during sublimation, each leading to distinct fluorescence emissions due to variations in π–π interactions. Furthermore, variations in slip angles between adjacent molecules lead to distinct aggregate patterns, specifically H- and J-type aggregates. H-type aggregation typically results in dimer excitation with blue-shifted emission relative to the emission of the monomer, whereas J-type aggregation induces red-shifted emission, a phenomenon that has been well documented. , These differences in molecular packing likely account for the observed variations in fluorescence behavior between the aggregated and solid states. This highlights the critical role of functional-group modifications in altering intermolecular interactions, thereby diversifying packing states and enabling precise tuning of optical properties.

2.4. Mechanistic Studies of Solid-State Fluorescence

To confirm our hypothesis regarding the role of molecular packing in influencing the photophysical properties of the compounds, we synthesized two additional derivatives, 6 and 17, by introducing a bulky cyclohexyl group to 3 and 13, respectively, thereby inducing differences in their packing conformation in the solid state (Figure a). Interestingly, these derivatives also exhibited distinct emission properties in the aggregated state. 6 displayed a maximum emission wavelength of 560 nm, which is significantly blue-shifted compared with that of 3 (603 nm). Similarly, 17 exhibited an emission maximum at 596 nm, which is blue-shifted compared with that of 13 (657 nm). These shifts highlight the effect of the bulky cyclohexyl group on the packing arrangement and subsequent photophysical properties of the molecules in the aggregated state. By contrast, 6 and 17 exhibited absorption and emission wavelengths in solution that were nearly identical to those of 3, 4, and 13 (Figures c, S6, and S7). This consistency suggests that the introduction of the cyclohexyl group and variations in carbonyl substituents do not significantly affect the emission of the solution state. To verify the minimal impact of these modifications on the S₀–S₁ E _g of these molecules in solution, we performed further calculations for 6 and 17. As shown in Figure d, the E _g values of the derivatives were nearly identical to those of the other derivatives, measuring approximately 1.55 eV. This uniformity in E _g aligns with the observed similarity in their solution-phase absorption and emission wavelengths. These findings confirm that the carbonyl substituent and bulky cyclohexyl group primarily influence intermolecular interactions and packing in the aggregated state, with negligible effects on the emission wavelengths of the solution state.

To further validate these findings, we conducted SCXRD studies on 3, 4, and 6, and present the crystallographic data in Table S4. As shown in Figure , the dihedral angles between the phenyl and indolizine moieties are 44.4° for 3, 41.3° and 37.3° for 4, and 30.2° for 6, indicating that all three compounds adopt twisted conformations in their aggregated and solid states. This twisted geometry induces RIR, contributing to the aggregate and solid-state emissions observed in the viscochromism experiments. The crystal-packing states of the three compounds were also investigated. 3 crystallizes in an orthorhombic structure (Pna2₁), 4 adopts a monoclinic structure (Pn), and 6 exhibits a monoclinic structure (P2₁/c). This demonstrates that the crystal packing of the molecules is influenced by the substitution of the carbonyl group across these derivatives. Specifically, 3 forms cofacial and parallel stacking along the b-axis, with a centroid distance of 3.84 Å and mean planar distance 3.48 Å, which facilitates π-π interactions between phenyl–phenyl and indolizine-indolizine units resulting in H-type conformation (Figures and S12). Similarly, 4 also forms cofacial and parallel stacking along the b-axis, with a centroid distance of 3.74 Å and mean planar distance of 3.38 Å and 3.43 Å, which is slightly closer (0.1 Å and 0.05 Å) than the 3 (Figure and S12). In contrast to 3 and 4, 6 exhibits cofacial, twisted, and antiparallel stacking along the b-axis (Figures and S13). The bulky cyclohexyl groups create van der Waals interactions between parallel molecules, resulting in oppositely directed cofacial π-π interactions among adjacent molecules. Noteworthy, the cofacial and antiparallel orientations contribute to the twisted π-π interactions and an H-type oblique conformation, complicating the accurate measurement of the mean planar distance between adjacent molecules (Figure S13).

5.

Crystal structure analysis of 3, 4, and 6. Crystal packing diagram and photograph of crystal under UV light (3) or fluorescence microscopic imaging (4 and 6) (left), crystal structure analysis, definition of dihedral angles (θ), and calculated θ between phenyl and indolizine (middle), and the centroid distance between adjacent molecules (right) of (a) 3; (b) 4; and (c) 6. All hydrogen atoms have been omitted for clarity.

To experimentally examine the influence of variations in molecular packing arrangements on fluorescence emission characteristics, a comparative analysis was conducted on the emissions of the solid-state and monomeric forms of 3, 4, and 6. Considering that monomeric emissions in solvents typically exhibit a broad wavelength range depending on solvent polarity (Figure S6), the fluorescence of these compounds in their monomeric states was measured using hexane, a nonpolar solvent selected specifically to minimize ICT contributions. As shown in Figure S15, the fluorescence spectra of the monomeric forms in hexane exhibited a wavy emission pattern and were blue-shifted relative to their solid-state emissions. Notably, this observed behavior diverges from the typical H-type aggregation emission, which generally involves dimer excitations resulting in emissions that are blue-shifted compared to the monomeric state, suggesting the presence of additional factors influencing the observed emission patterns.

To resolve this apparent discrepancy, we investigated the excited-state dynamics of the H-type aggregates with respect to exciton splitting, as delineated by Kasha’s exciton model and excimer formation (Figure a). − Upon excitation of an organic molecule, the electron in the LUMO and the hole in the HOMO form a bound pair through Coulombic attraction, termed an exciton. This exciton may become delocalized across adjacent molecules in the aggregated state, resulting in dimer formation. , According to Kasha’s model, Coulombic interactions between the transition dipole moments in such dimers lead to the emergence of two energy levels, a phenomenon known as exciton splitting. Based on selection rules, only the in-phase combination of dipoles facilitates an allowed radiative transition.

6.

Proposed mechanism for distinct solid-state fluorescence emission. (a) Simplified illustration of the energy diagram of monomer and H-type dimers. (b) Schematic illustration of the dimeric energies of the ground, excited, and excimer states of 3, 4, and 6. The dipole moments of each molecule are represented by blue (ground state) and red (excited state) arrows.

According to this rule, H-type aggregate dimers experience exciton splitting into two distinct excited states due to dipole interactions between monomers. In these aggregates, identical dipole alignments of each monomer yield a higher-energy excited state (E₂), whereas opposite dipole alignments yield a lower-energy excited state (E₁) due to Coulomb interactions. Notably, in the E₁ state, the opposing dipoles between the monomers lead to a net-zero transition dipole moment, resulting in a forbidden radiative transition. Therefore, H-type dimers typically exhibit unstable relaxation from the higher-energy E₂ state directly to the ground state, resulting in emission wavelengths that are blue-shifted compared to the monomeric state. However, when excimer formationdefined as an interaction between a ground-state monomer and an excited-state monomeroccurs, it introduces an alternative, energetically stabilized excited state. Consequently, emission from this excimer is red-shifted compared to that of the monomer, circumventing the forbidden E₁ transition. The fluorescence behaviors of 3 and 4 are consistent with the characteristic excimer properties of typical H-type aggregate fluorophores. Conversely, 6 exhibits an oblique crystal packing conformation that prevents dipole cancellation in the E₁ state, enabling relaxation directly to the ground state. Thus, the solid-state fluorescence emission of 6 peaks at 559 nm, corresponding to direct relaxation from the E₁ state, accompanied by a shoulder around 590 nm, indicative of partial relaxation from an excimer state (Figure S15). Overall, these findings imply that excimer formation (or exciton splitting) significantly influences the solid-state emission properties of 3, 4, and 6.

To further elucidate the influence of dimer conformations on the observed solid-state fluorescence properties, computational calculations were conducted to examine the relevant energy states, with a particular focus on binding energies and dipole–dipole interactions within the dimers. Although the dipole moments of dimers of 3 and 4 increased markedlyby approximately 1.9-fold (3: from 3.94 to 7.55 D) and 1.8-fold (4, from 5.83 to 9.85 and 10.95 D)6 experienced a dramatic, nearly 20-fold reduction in dipole moment (7.19 to 0.36 D) (Figure S16). This substantial difference arises because the dipoles in the dimers of 3 and 4 are aligned in parallel orientations, resulting in enhanced dipole moments and thus greater Coulomb repulsion. Conversely, the antiparallel orientation of dipoles in the dimer of 6 leads to dipole cancellation, inducing Coulomb attraction and increased stabilization of its ground state. Consistent with these dipole interactions, binding-energy calculations performed using the M062X/6-311++G** method demonstrated that dimers of 3 (−11.58 kcal/mol) and 4 (−10.16 and–10.66 kcal/mol) are less stable compared to the dimer of 6 (−16.45 kcal/mol). Although the dimer conformation of 6 exhibits an oblique-antiparallel configuration with a slight deficiency in π-π interactions, this observation suggests that the Coulomb attraction of the dimers contributes to additional stabilization of the ground state. Consequently, this stabilization increases the E _g between the excited and ground states, ultimately leading to the observed blue-shifted solid-state emission (∼47 nm) of 6 compared to 3 (Figure b).

Meanwhile, the minimal difference in binding energies between the dimers of 3 and 4 can be attributed to a balance between enhanced π-π stacking interactions and increased Coulomb repulsion due to their similarly oriented dipoles. Specifically, 4 exhibits slightly reduced mean planar distances compared to 3, thereby strengthening π-π interactions. However, the parallel alignment of molecular dipoles simultaneously increases Coulomb repulsion, counterbalancing the net stabilization from π-π stacking. Notably, significant differences emerge when comparing their excimer states. The dipole moment of the excited monomer within an excimer is hypothesized to exhibit an orientation opposite to that of its ground state monomer. This is consistent with the observed shift in electron density from the S₀ to the S₁ state (Figure d). Consequently, the dipole orientations in the excimer state are inverted relative to each monomer, thereby converting Coulombic repulsion into attraction. Consequently, the excimer state of 4 benefits more significantly from both increased π-π interactions and Coulomb attraction due to its reduced intermolecular distances, thereby stabilizing the excimer energy relative to 3. This additional stabilization results in a narrower E _g and accounts for the pronounced red shift (∼55 nm) in the solid-state emission of 4 relative to 3. Collectively, these results (summarized in Figure b) clearly demonstrate the profound influence of subtle conformational variations within dimers on solid-state emission behaviors. In particular, molecular orientations dictate dipole–dipole interactions that intricately interplay with π-π stacking interactions, ultimately determining the photophysical characteristics of these indolizine-based AIE fluorophores. Lastly, we conducted powder X-ray diffraction (PXRD) analysis on each compound in both crystalline and amorphous states to assess whether the proposed solid-state fluorescence mechanism, derived from SCXRD analysis, is applicable to the amorphous samples. As shown in Figure S17, the simulated PXRD patterns based on SCXRD data closely corresponded with those of both crystalline and amorphous samples, thereby indicating that the proposed mechanism is valid for interpreting the solid-state fluorescence behavior of each compound. These insights into structural and electronic relationships will significantly inform the rational design of novel fluorophores with precisely tunable emission characteristics.

2.5. Application of AIEgens in WLE Device

Finally, we investigated the potential of these compounds as light-conversion materials for WLEDs. The detailed fabrication process is described in the Supporting Information. As shown in Figure a, the phosphor films were synthesized by incorporating 2 or 3 into a polydimethylsiloxane (PDMS) polymer matrix. These compounds were selected after evaluating multiple combinations, as they demonstrated the most effective color conversion and balanced emission properties for achieving white-light generation (Figure S18). These films convert the blue light emitted from the light-emitting diode (LED) into longer wavelengths, allowing the human eye to perceive the overall light as white. To evaluate their light-conversion efficiency, we analyzed the emission spectra of each film. The experimental setup, as shown in Figure a, consists of a 3D-printed holder with 5.0 mm-radius circular openings at the top, housing a blue LED and the PDMS film. This setup was designed to minimize external light interference in spectral measurements. The emission spectra reveal distinct color shifts depending on the incorporated compound. The 2-PDMS film converted the blue LED light into cyan emission, peaking at near 500 nm. In contrast, the 3-PDMS film exhibited an orange-colored emission with a central peak near 600 nm. These results indicate that a proper combination of these three colorsblue from the LED, cyan, and orange from the converted lightcan generate white light, as they collectively stimulate the three photoreceptors for blue, green, and red.

7.

Application of indolizine-based AIEgens 2 and 3 to a white-light device. (a) Photograph of 2-PDMS and 3-PDMS film (left), schematic illustration of preparing the emission using a 3-D printed holder, fabricated films, and blue LED (middle), and their luminescence emission spectra with its photograph (right). (b) Principle of white-light emission. (c) Photograph of the white-light emission (left), luminescence emission spectra (middle), and the color coordinates of emitted white light from the prepared white-light device, along with the determined color rendering index (CRI) (right).

To achieve white-light emission, we fabricated a multistack layer consisting of three layers of 2-PDMS film and one layer of 3-PDMS film, as illustrated in Figure b, and additional results from different phosphor film combinations are provided in Figure S19. In this configuration, the cyan emission from the 2-PDMS film not only contributes to the final spectrum but also serves as an excitation source, enhancing the orange light emission from the 3-PDMS film. Notably, the 3-PDMS film also exhibited near 600 nm emission when excited by cyan-colored LED illumination, as shown in Figure S19. The resulting emission spectrum from this multistack layer, excited by the blue LED, is displayed in Figure c. The transmitted light exhibited cyan and orange emissions at near 500 and 600 nm, respectively, along with blue light transmission at 425 nm. These combined emissions produce white light, as confirmed by visual perception. To further characterize the white-light emission, we analyzed its color temperature using the CIE 1931 chromaticity diagram. As shown in Figure c, the white emission had color coordinates (x = 0.360, y = 0.362), corresponding to a color temperature of 4533 K, which falls within the neutral white-light range. Hence, we successfully demonstrated the practical application of 2-PDMS and 3-PDMS films as efficient light-conversion layers for white-light generation in WLEDs. Furthermore, the additional fabrication of the three-color combined (2-PDMS, 3-PDMS, and 4-PDMS) experiment demonstrated a visible white light emission with CIE coordinates of (0.388, 0.405) and a color temperature of 4004 K (Figure S18). This work provides a new strategy for designing efficient light-conversion materials through molecular engineering of AIE fluorophores. Further optimization of phosphor film composition and stacking configurations could enhance color rendering performance for next-generation WLED applications.

3. Conclusion

In conclusion, this study presents the development of novel indolizine-based fluorophores with AIE properties, characterized by compact molecular frameworks and tunable fluorescence emissions, making them suitable for WLED applications. The initial derivatives, 2 and 3, exhibited aggregation-induced fluorescence ranging from green (496 nm) to orange (603 nm) as f _w increased, with their AIE behaviors influenced by the complementary roles of RIR and ICT. Subsequent structural modifications at the ester position of 3, through the introduction of various carbonyl substituents, resulted in additional derivatives (4, 6, 13, and 17). These modified derivatives, with an average molecular weight of 313.9 g/mol, exhibited distinct and tunable emission wavelengths ranging from yellow (559 nm, 6) to deep red (669 nm, 13) in both aggregated and solid states, despite showing minimal changes in their monomeric emission properties. Crystallographic analysis and computational studies revealed that these emission variations primarily stem from differences in molecular packing, intermolecular π–π stacking, and dipole–dipole interactions, particularly within the excimeric states. Notably, subtle changes in molecular conformation significantly affected the dipole orientations and Coulomb interactions within dimers, thereby influencing solid-state fluorescence. Finally, the WLED was successfully fabricated by integrating these phosphors into a 425 nm blue LED, achieving CIE coordinates (0.360, 0.362) and a color temperature of 4533 K, demonstrating the potential of these AIEgens in optoelectronic device applications. Overall, this study offers significant insights into the impact of functional group modifications and molecular packing on emission properties, thereby providing a rational framework for the precise design of fluorophores with specific photophysical characteristics. Building on these findings, future research will aim to develop novel AIE materials with exceptional optical properties for potential applications in advanced technologies, including bioimaging, sensing, and next-generation optoelectronic devices.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/jacsau.5c00426.

• General experimental information, experimental procedures, Tables S1–S5, Figures S1–S19, synthetic procedures, and absorption and emission spectra of 2, 3, 4, 6, 13, and 17, and ¹H/¹³C NMR spectra of new compounds (PDF)

• C₁₇H₁₄N₂O₄ (CIF)

• C₁₅H₁₀N₂O₃ (CIF)

• C₂₁H₂₀N₂O₄ (CIF)

T.K. and J.K. conceived and designed the project; T.K. and E.B. performed the synthesis; T.K. performed photophysical investigation for AIEgens; J.H. performed solid fluorescence investigation; T.K. solved the crystal structure; K.-H.C. performed the DFT calculations; J.H., H.K., Y.L., and H.O. performed the WLE tests, including the fabrication of the WLE device; T.K. and J.K. wrote the manuscript. All the authors discussed the results and commented on the manuscript.

The authors declare no competing financial interest.