Enhancement of the triplet photosensitizing ability of BODIPY dyes by controlled red-shifted H-aggregate formation using ionic liquids: insights from spectroscopy, microscopy, DFT and biological studies

Abstract

We report an innovative, straightforward, and effective supramolecular method for converting fluorophores into photosensitizers using symmetrical imidazolium ionic liquids (ILs) via the aggregation of BODIPY derivatives. By utilizing “red-shifted H-aggregates” of 2,6-diethyl-4,4-difluoro-1,3,5,7-tetramethyl-8-phenyl-4-bora-3a,4a-diaza-s-indacene (R6-BDP), along with both shorter and longer cationic alkyl chain symmetrical ILs, we enhanced the conversion efficiency of the R6-BDP aggregates from fluorophores to photosensitizers. The optical properties of R6-BDP in both the molecular and aggregated states were investigated through UV-visible, fluorescence and time-resolved fluorescence spectroscopy techniques. Notably, the absorption data indicated the formation of “red-shifted H-aggregates”, which is an unusual phenomenon. The emission intensity of the R6-BDP aggregates decreased by approximately 65-fold compared to their molecular form, suggesting that the aggregation-caused quenching effect (ACQ) occurred in the aggregated state. The average lifetimes of the molecular and aggregate forms were estimated to be 4.52 ns and 1.02 ns, respectively. Photophysical studies suggested that the addition of a shorter alkyl chain IL (C₂C₂im-Br) promotes the formation of R6-BDP aggregates, whereas a longer alkyl chain IL (C₈C₈im-Br) induces their dissociation, returning the aggregates to their molecular form. No significant effect on aggregation was observed with the medium-length alkyl chain IL (C₄C₄im-Br). FESEM (field emission scanning electron microscopy) analysis revealed that in the presence of C₂C₂im-Br, the particle size of the R6-BDP aggregates increased, with a corresponding change in shape from round to rod-like structures. However, the particle size decreased with the addition of C₈C₈im-Br, and the particles appeared circular in shape. Theoretical density functional theory (DFT) studies provided further confirmation of the formation of H-type aggregates for the R6-BDP molecules. Biological studies revealed that the addition of a short alkyl chain IL to the R6-BDP aggregates significantly enhanced singlet oxygen generation, as evidenced by increased green fluorescence in HeLa cells under light irradiation, indicating an improvement in their photodynamic therapy (PDT) performance. The current work offers a novel approach for the research community to realize photodynamic therapy.

---

1. Introduction

Significant attention has recently been directed toward BODIPY (4,4-difluoro-4-bora-3a,4a-diaza-s-indacene) dyes because of their remarkable luminescent properties.^1,2 These organic dyes exhibit high fluorescence quantum yields, sharp absorption bands and intense emission peaks across the visible to infrared range.^1–4 One of the main features of BODIPY derivatives that has garnered significant attention is their tunable fluorescence capabilities.^5,6 The wavelengths at which these dyes absorb and emit light can be adjusted as per the requirement by modifying the structure of the BODIPY core via the incorporation of various functional groups. This tunable fluorescence property is highly useful to design BODIPY derivatives with specific fluorescence characteristics suitable for a variety of applications.^1,2,7–10 However, their inherently low intersystem crossing (ISC) rate makes them inefficient triplet photosensitizers (PSs), i.e. restricts their use as PSs in photodynamic therapy (PDT) of cancers. Consequently, research has been dedicated to increasing their triplet yields for developing efficient PDT agents based on BODIPY dyes.

The ISC rate constant (k_ISC) for the singlet excited state (S_m) and triplet excited state (T_n) can be written as follows:^8,11,12

According to this equation, k_ISC is proportional to the square of the spin–orbit coupling (SOC) matrix element (H_so represents the SOC Hamiltonian) and inversely proportional to the square of the energy gap between S_m and T_n (ΔE_Sm–Tn). Thus, various heavy atom-attached BODIPY dyes have been synthesized, which effectively increased the SOC to enhance their photosensitizing ability.⁹ However, heavy atoms enhance the dark toxicity and increase the cost and pollution, making them unsuitable for practical applications. Alternatively, heavy atom-free PSs can be synthesized by designing low ΔE_Sm–Tn molecules.¹³ The planar conjugated BODIPY molecule has a high energy gap between the singlet and triplet excited states (ΔE_Sm–Tn) due to the very strong spatial overlap of its molecular orbitals. Thus, the ISC process is inefficient in BODIPY, showing a very low triplet quantum yield. Various strategies have been adopted to reduce ΔE_Sm–Tn such as molecular twisting and generation of charge transfer states, which enhance the ISC rate.¹⁴ Thus far, all the strategies adopted to the improve triplet yield are based on the synthetic modification of BODIPY dyes, which are tedious, time-consuming and resource intensive. Instead, new strategies that enable control of the fluorescence of BODIPY to enhance its ISC rate without synthesizing new molecules would be highly interesting. One strategy involves the formation of dye aggregates, which can help control the fluorescence properties.^15,16 The aggregation state of dye molecules causes the fluorescence intensity to decrease, suggesting that the released energy is transferred through nonradiative processes such as internal conversion and intersystem crossing. If the ISC process is enhanced, the chromophore can act as a triplet sensitizer, transferring its energy from the excited triplet state to molecular oxygen, producing singlet oxygen (¹O₂). Kang et al.¹⁷ reported for the first time an innovative supramolecular method to transform fluorescent organic molecules into effective photosensitizers via aggregation. This strategy to create photosensitizers through aggregation can be helpful for PDT in cancer treatment. Based on the discussion above, the current work aimed to develop a strategy to control the fluorescence intensity of aggregated BODIPY derivatives, and also to improve the triplet conversion of BODIPY dyes, which inherently have a very low ISC conversion rate.

It is commonly known that in aqueous solution, dye molecules containing polynuclear aromatic moieties strongly prefer to form colloidal aggregates.¹⁸ Thus, hydrophobic BODIPY dyes also have a tendency to form aggregates in water because of their intermolecular interactions and π–π stacking of the planar π conjugated BODIPY chromophore.¹⁹ Because of the formation of different aggregate structures in the presence of water, the emission properties of the BODIPY molecule in the aggregated state in most cases are significantly quenched.¹⁹

According to previous reports, micelles can serve as a great medium for encapsulating hydrophobic molecules, and also can aid in making them more soluble in aqueous solutions.^20,21 The preparation of micelles commonly involves the use of surfactants such as sodium dodecyl sulphate and TRITON-X.^20,22 However, micelles have also been formed in aqueous media using surface active ionic liquids (SAILs).^20,22 Because SAILs have long alkyl chains, which give them surface-active qualities, they have low toxicity, very low vapor pressure, great thermal stability, and other desirable characteristics.²³ It has been demonstrated that SAILs can function as more efficient micelle-forming agents than traditional surfactants.^24,25 Dandpat et al.²⁶ demonstrated that SAIL can serve as an effective medium for dissociating organic aggregates. Martins and co-workers^27,28 used N-alkyl isoquinolinium, N-alkyl pyridinium, and 1-alkyl-3-methyl benzimidazolium chloride ionic liquids to induce the aggregation of asphaltene and understand the influence of these ILs on the aggregation of asphaltene. Although some reports in the literature address the impact of ILs on organic aggregates of various small drug molecules,^29,30 currently there are no reports available specifically examining their effects on BODIPY derivatives. Moreover, researchers have primarily focused on the effect of imidazolium-based asymmetrical ionic liquids on the aggregation of organic molecules,^26,28 leaving the impact of symmetrical imidazolium-based ILs unstudied. Notably, the physicochemical properties and structural organization of imidazolium-based symmetrical ILs differ significantly from that of their asymmetrical counterparts.^31–34 Therefore, it is of particular interest to investigate how symmetrical ionic liquids affect the aggregated state of BODIPY molecules. Specifically, this study aimed to explore how short and long cation alkyl-chain based symmetrical imidazolium ILs influence the aggregation state of BODIPY molecules. The van der Waals and electrostatic interactions among the ionic constituents of ILs vary with changes in their cation or anion chain length, thereby influencing both their overall properties and structural organization.^32–37 Thus, it is crucial to examine whether the cation chain length of symmetrical IL affects the formation of BODIPY aggregates.

For this purpose, we synthesized a BODIPY derivative, 2,6-diethyl-4,4-difluoro-1,3,5,7-tetramethyl-8-phenyl-4-bora-3a,4a-diaza-s-indacene (R6-BDP). Additionally, we prepared a series of ILs, including 1,3-diethylimidazolium bromide (C₂C₂im-Br), 1,3-dibutylimidazolium bromide (C₄C₄im-Br), and 1,3-dioctylimidazolium bromide (C₈C₈im-Br), to examine the influence of symmetrical ILs on the aggregates of R6-BDP. The molecular structures of all these compounds are provided in Chart 1. Initially, we examined the photophysical behavior of R6-BDP in both its molecular and aggregated forms. The effect of symmetrical ILs on the aggregation state of R6-BDP was also examined using various spectroscopic techniques, including UV-visible, fluorescence and time-resolved fluorescence spectroscopy. Their microscopic behaviors in the aggregated state, both in the presence and absence of ILs, were examined by field emission scanning electron microscopy (FESEM) and dynamic light scattering (DLS) methods. DFT-based theoretical calculations were also performed to evaluate the aggregation patterns and triplet photosensitizing properties of this molecule. Further, the singlet oxygen generation capabilities of the BODIPY dye aggregates were measured. Finally, biological studies were performed to confirm the singlet oxygen generation by the R6-BDP aggregates, both in isolation and in the presence of IL. The cytotoxicity of the R6-BDP aggregates, with and without ionic liquid, under light illumination conditions was evaluated using HeLa cells. The findings offer valuable insights into the crucial role of ILs in enhancing the PDT activity of BODIPY dyes. This information will be valuable for the practical application of ionic liquids in enhancing BODIPY-based PDT therapies for real-time cancer treatment.

Chart 1 Chemical structures of R6-BDP, C₂C₂im-Br, C₄C₄im-Br and C₈C₈im-Br.

2. Experimental

2.1. Materials

The compounds used for the synthesis of R6-BDP including 3-ethyl-2,4-dimethyl-1H-pyrrole, triethylamine, benzoyl chloride, and boron trifluoride etherate were obtained from Sigma-Aldrich. The compounds for the preparation of the ionic liquids (C₂C₂im-Br, C₄C₄im-Br, and C₈C₈im-Br) including ethyl imidazole, butyl imidazole, octyl imidazole, ethyl bromide, butyl bromide, and octyl bromide were also purchased from Sigma-Aldrich. 1,3-Diphenylisobenzofuran was used in the biological studies and obtained from Sigma-Aldrich. R6-BDP aggregates were prepared using double-distilled Milli-Q water.

2.2. Synthesis and characterisation

2.2.1. 2,6-Diethyl-4,4-difluoro-1,3,5,7-tetramethyl-8-phenyl-4-bora-3a,4a-diaza-s-indacene (R6-BDP)³⁸. 3-Ethyl-2,4-dimethyl-1H-pyrrole (1.10 mL, 8.13 mmol) and benzoyl chloride (0.47 mL, 4.06 mmol) were mixed in dry dichloromethane (25 mL) in a Schlenk flask under an argon (Ar) atmosphere, and then the reaction mixture was stirred for 24 h at 25 °C. Then the mixture was ice-cooled and triethyl amine (5.66 mL, 40.6 mmol) was slowly added to it and stirred for 15 min. After that, boron trifluoride etherate (BF₃·OEt₂) (6.0 mL, 48.72 mmol) was slowly added to this solution and the resultant mixture was stirred for 24 h at 25 °C. Then, the reaction was quenched with 50 mL of saturated aqueous sodium bicarbonate solution, extracted with 3 × 75 mL of dichloromethane, washed with 3 × 50 mL of water and 50 mL of brine and dried with Na₂SO₄. The organic mixture was concentrated in vacuo to obtain and the crude product, which was purified by flash column chromatography (silica gel, DCM/petroleum ether, 50 : 50) to furnish the R6-BDP (385 mg, 50%) dye as a brick red solid. The compound was characterised by ¹H and ¹³C NMR spectroscopy (Fig. S1 and S2).

¹H NMR (500 MHz, CDCl₃): δ (ppm) 7.47–7.46 (m, 3H), 7.29–7.27 (m, 2H), 2.53 (s, 6H), 2.30 (q, J = 7.5 Hz, 4H), 1.27 (s, 6H), 0.98 (t, J = 7.5 Hz, 6H).

¹³C NMR (125 MHz, CDCl₃): δ (ppm) 153.7, 140.2, 138.4, 135.8, 132.7, 130.8, 128.9, 128.7, 128.3, 17.1, 14.6, 12.5, 11.6.

2.2.2. 1,3-Dialkyl imidazolium bromide (C_nC_nim-Br). The ionic liquids, 1,3-dialkyl imidazolium bromide (C_nC_nim-Br, where n = 2, 4, 8), were synthesized by following the reported procedures.^31,39 To synthesize C_nC_nim-Br, 1-alkyl-imidazole (1.0 mmol) was added to 30 mL of acetonitrile in an RB (round-bottom) flask and the mixture was stirred. Then, 1-bromoalkane (1.10 mmol) was added dropwise to the reaction mixture. The mixture was refluxed under a nitrogen atmosphere for 72 h. The reaction mixture was concentrated in vacuo and the oily product was washed 5–6 times with diethyl ether, yielding a colorless liquid. Then, the compounds were dried under vacuum for two to three days to remove any water and ensure the high purity of the ILs. Characterization of the ILs was carried out using ¹H NMR spectroscopy, along with mass spectrometry (Fig. S3–S8).

C₂C₂im-Br: ¹H NMR (500 MHz, DMSO-d₆): δ (ppm) 9.48 (s, 1H), 7.90 (d, 2H), 4.22 (q, 4H), 1.41 (t, 6H).

ESI-MS (+ve): 125 m/z, [C₂C₂im]⁺

C₄C₄im-Br: ¹H NMR (500 MHz, DMSO-d₆): δ (ppm) 9.45 (s, 1H), 7.88 (d, 2H), 4.21 (t, 4H), 1.78 (m, 4H), 1.24 (m, 4H), 0.88 (t, 6H).

ESI-MS (+ve): 181 m/z, [C₄C₄im]⁺

C₈C₈im-Br: ¹H NMR (500 MHz, DMSO-d₆): δ (ppm) 9.61 (s, 1H), 7.95 (d, 2H), 4.23 (t, 4H), 1.79 (m, 4H), 1.20 (m, 20H), 0.81 (t, 6H).

ESI-MS (+ve): 293 m/z, [C₈C₈im]⁺

2.3. Preparation and characterization of R6-BDP aggregates

The preparation of R6-BDP aggregates was studied using the conventional re-precipitation method.^40,41 Usually, 30 μL of R6-BDP solution (0.01 M) in acetonitrile was quickly added to water (3 mL) under sonication. After 20 min of continuous stirring, colloidal aggregates were formed. The formation of R6-BDP aggregates was confirmed by characterizing the freshly prepared solution using UV-visible absorption, FESEM and DLS.

2.4. Instrumentation and methods

The ¹H and ¹³C NMR spectroscopic techniques were performed using a Bruker Avance NMR (500 MHz) spectrometer, with tetramethylsilane (TMS) serving as the internal standard. An Agilent QTOFG6545 spectrometer was used to obtain high-resolution mass spectrometry data at a resolution of 50 000 in ESI mode. A JASCO V-750 spectrophotometer was used to record all the steady-state absorption spectra, while fluorescence spectra were obtained with a JASCO FP-8200 spectrofluorometer. Time-resolved fluorescence studies were performed using a DeltaFlex time correlated single photon counting (TCSPC) spectrometer from HORIBA. The probe was excited by a laser diode source (λ_exc = 500 nm, FWHM = 81 ps) and an HPPD (HORIBA) detector was used. The EzTime decay analysis software was used to evaluate the fluorescence intensity decay through a non-linear least square's iteration process. The χ² (chi square) values were used to assess the fit quality, and fitting was used to determine the weighted deviation. Origin 8.0 was used for the plotting and analysis of all data. R6-BDP aggregates, with and without ILs, were analysed via imaging using a ZEISS FESEM.

2.5. Computational methods

The ground-state structures of R6-BDP in both the monomer and dimer were optimized to gain insight into the inter and intra-molecular interactions of R6-BDP, which helps to understand the aggregation behavior of the R6-BDP molecule. The optimization of the R6-BDP molecule was performed using density functional theory (DFT) calculations employing the B3LYP functional and the 6-31+G(d,p) basis set, as implemented in the Gaussian 09 software.^42,43 The ground-state geometries of the R6-BDP molecules and dimer of the R6-BDP molecule were initially optimized using DFT in acetonitrile, and then time-dependent DFT (TD-DFT)⁴⁴ was used to simulate the absorption (excitation) spectra.

2.6. Singlet oxygen (¹O₂) generation

To detect singlet oxygen (¹O₂) in various solutions, 1,3-diphenylisobenzofuran (DPBF) was employed as a chemical probe. Solutions of molecular R6-BDP, aggregate form, and R6-BDP aggregates with ionic liquid were each dispersed in 1 mL of an aqueous solution containing 60 μM DPBF. Subsequently, the mixtures were transferred to a cuvette and exposed to monochromatic light. Changes in absorption at 410 nm were monitored using a UV-vis absorption spectrophotometer.

2.7. Cell culture

The human cervical cancer cell line (HeLa) was maintained in Dulbecco's modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 10% antibiotic–antimycotic solution in a CO₂ incubator at 37 °C. The cells were cultured in 60 mm dishes until reaching approximately 80% confluence.

2.8. Cell viability assay

The viability of the HeLa cells was measured by means of the MTT ((3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide) assay. Briefly, the cells were seeded (1 × 10⁴ cells per well) in 96-well plates and maintained in a CO₂ incubator at 37 °C for 24 h. At approximately 80% confluence, the compound R6-BDP aggregates and aggregated form +C₂C₂im-Br were added at 0.2 μM, 2 μM, 4 μM and 8 μM concentrations, respectively. After two hours of incubation in the dark, one set was exposed to LED for 5 min. Then, the plates were incubated overnight at 37 °C in a CO₂ incubator. MTT (0.5 g L⁻¹) was added to each well, as described previously by Mosmann⁴⁵ to estimate mitochondrial dehydrogenase activity. After 4 h incubation, DMSO was added to each well to solubilize the MTT formazan crystals and the absorbance was recorded at 570 nm. The outcomes were reported as the percentage of cytotoxicity.

2.9. Measurement of intracellular reactive oxygen species

Briefly, cells were seeded (1 × 10⁴ cells per well) in black glass-bottom 96-well plates (BD Bioscience, USA) and incubated in a CO₂ incubator for 24 h at 37 °C. At approximately 80% confluence, the compound aggregated form of R6-BDP and aggregated form +C₂C₂im-Br were added at 0.2 μM, 2 μM, 4 μM and 8 μM concentrations, respectively. After 2 h of incubation in the dark, one set was exposed to an LED light for 5 min. The plates were incubated overnight in a CO₂ incubator. Reactive oxygen species (ROS) in HeLa cells were evaluated by utilizing a fluorescent probe, DCFH-DA, as reported previously by Cathcart et al.⁴⁶ Briefly, the cells were incubated with DCFH-DA (20 μM) in PBS (pH 7.4) for 20 min. After washing with PBS, the cells were imaged using a fluorescence microscope (Olympus IX83) equipped with FITC filters (excitation at 490 nm and emission at 525 nm). Fluorescence intensity was determined using the cellSens program.

3. Results and discussion

3.1. Photophysical studies of R6-BDP in molecular and aggregated form

The photophysical properties of R6-BDP were initially examined in both its molecular and aggregated forms. The absorption spectra of R6-BDP in both forms are presented in Fig. 1(a), while Table 1 summarizes the corresponding data. As seen in Fig. 1(a), a sharp absorption band was observed at approximately 522 nm in the molecular form, with a shoulder absorption band appearing at around 491 nm. The sharp absorption in the visible region is attributed to the 0 → 0 vibronic band for the S₀ → S₁ transition, while the shoulder corresponds to the 0 → 1 vibronic band for the same transition. It is worth mentioning that similar types of absorption bands have been published by other researchers when examining the absorption spectra of BODIPY molecules.⁸ Similar to its molecular form, the absorption profile of R6-BDP in the aggregated form displayed a sharp peak near 534 nm and a shoulder at around 502 nm (Fig. 1(a)).

Fig. 1 (a) Normalised absorption and (b) emission spectra of R6-BDP in the molecular and aggregated form.

Table 1 Absorption (λ^max_abs) and emission (λ^max_em) maxima of R6-BDP in its molecular and aggregated forms

R6-BDP	Absorption (λ^maxabs) (nm)	Emission (λ^maxem) (nm)
Molecular form	522 ± 2	536 ± 2
Aggregated form	534 ± 2	560 ± 2

---

Again, the absorption maximum of R6-BDP in its aggregated form is red shifted by over 12 nm compared to its corresponding molecular form, as shown in Fig. 1(a). According to Kasha's theory, molecular aggregates are classified as J or H aggregates depending on whether their absorption spectrum is red or blue shifted compared to their molecular form.⁴⁷ Thus, the observed red shift in the absorption spectrum indicates the formation of J-type aggregates for the R6-BDP molecule. However, some recent reports in the literature suggest that the red shift in the absorption spectrum is not necessarily induced by J-aggregation, but rather by the coupling of excitons and vibrations.³⁸ Spano et al.^48,49 demonstrated that the type of aggregation is independent of the absorption spectral shift. Instead, it is determined by the ratio of the absorbance intensities of the first two vibrational bands, A₁ and A₂, corresponding to the 0 → 0 and 0 → 1 transitions, respectively.^48,49 Again IA₁ denotes the absorption intensity of the  0 → 0 vibronic band, and IA₂ denotes the absorption intensity of the  0 → 1 vibronic band (Table 2). The decrease in the ratio of absorption intensities of these two vibronic bands in the aggregated state, which is related to the molecular state of R6-BDP, shows the formation of H-type aggregates.^48,49

The absorption intensity ratio of the first two vibronic bands for both the molecular and aggregated forms is provided in the Fig. S9, with the corresponding data given in Table 2. The data demonstrates a significant reduction in the ratio of the first two vibronic bands from the molecular to aggregated form of R6-BDP, indicating that the aggregates are predominantly H-aggregates. Interestingly, the investigation revealed that the aggregates of R6-BDP exhibit non-Kasha's behaviour, i.e. “red shifted H-aggregates”, which is quite uncommon (Fig. 2). The formation of “red-shifted H-aggregates” is propelled by the substantial quadrupole–quadrupole interaction.⁵⁰ These aggregates exhibit all the typical features of H-aggregates, such as suppressed radiative decay rates and vibronic signatures in their absorption and emission spectra, consistent with the Frenkel exciton models, except for the red shift in their absorption peak.⁵⁰

Table 2 Absorption position of vibronic bands and change in their intensity ratio

R6-BDP	A1 position (nm)	A2 position (nm)	IA1/IA2	% change
Molecular form	522	491	2.941	48%
Aggregated form	534	502	1.508

---

Fig. 2 Schematic of the allowed transitions for the H-aggregate (Kasha) and red-shifted H-aggregate (non-Kasha).

To gain more insight into the excited-state properties of both the molecular and aggregated forms, we performed fluorescence spectroscopy. The emission spectra of R6-BDP in molecular and aggregated form are shown in Fig. 1(b) and data are listed in Table 1. Similar to the absorption spectra, the emission maximum (λ^max_em) of the R6-BDP molecules upon photoexcitation at 500 nm shifted towards longer wavelengths (red shift) by 24 nm in the aggregated state relative to the molecular form. However, the emission intensity in the aggregated state of R6-BDP was significantly quenched compared to the molecular form of R6-BDP. In fact, the fluorescence intensity decreased by approximately 65-fold in the aggregated state relative to the molecular form due to the aggregation-caused quenching (ACQ) phenomenon (Fig. 1(b)). Several factors may contribute to the reduced fluorescence intensity in the aggregated state. One factor is the ACQ effect likely occurring due to a change in the emitting state upon aggregation.⁵¹ This effect is driven either by the strong π–π coupling of the planar π-conjugated structures of BODIPY molecules, which results in the formation of excimers or exciplexes, or by collisions between the excited and ground states of the fluorescent molecules at relatively high concentrations, resulting in non-radiative deactivation.⁵²

To gain a better understanding of the ACQ effect, we performed time-resolved fluorescence lifetime measurements of R6-BDP in both the molecular and aggregated states. The fluorescence lifetime data for both forms are provided in Table 3. Notably, the fluorescence intensity decay of the molecular form was best fitted by a mono-exponential model, while the aggregated form followed a tri-exponential decay. The tri-exponential fitting parameters for the aggregated form are provided in the Table S1. The fluorescence lifetimes of the molecular and aggregated forms of R6-BDP were estimated to be 4.54 ns and 1.02 ns, respectively. The data clearly indicate that the lifetime values are significantly lower in the aggregated state than in the molecular form, further supporting the ACQ phenomenon in the excited state.^48,53

Table 3 Average lifetime data for the molecular, aggregated and ionic liquids added to the aggregated form of R6-BDP

System	Average lifetime (ns)	Chi square (χ^2)
Molecular form	4.54 ± 0.23	1.05
Aggregated form	1.02 ± 0.05	1.18
Aggregated form + C2C2im-Br	1.00 ± 0.05	1.05
Aggregated form + C4C4im-Br	1.07 ± 0.05	1.02
Aggregated form + C8C8im-Br	5.12 ± 0.26	1.04

---

3.2. Effect of symmetrical ionic liquids on the R6-BDP aggregates

Over the past few years, researchers have thoroughly examined the development of very stable and organized supramolecular assemblies using oppositely charged surfactants. Thereafter, to create different supramolecular aggregates, the surfactants were replaced by surface-active ionic liquids, a unique family of ILs.^26–28,54 As a result, the structural elements of supramolecular aggregates-vesicular assemblies, micelles and mixed micelles have been altered from the oppositely charged SAIL/SAIL pair to surfactant/surfactant and surfactant/SAIL pairs.^26–28,54 The most recent addition to this ion pair combination are dye molecules with a hydrophilic ionic head group and a lengthy hydrophobic chain segment.^26–28,54 Hence, SAIL/dye or surfactant/dye pairs can also lead to the formation of different assemblies through electrostatic, hydrophobic or π–π stacking.^26–28,54 It is important to note that the chain length of the cation or anion in various ionic liquids plays a critical role in determining their properties.^26,27 By altering the chain length, the electrostatic and van der Waals forces among the components of the ionic liquids can be modified.^25,26 For instance, ILs with long alkyl chains, such as 1-dodecyl imidazole, exhibit behavior typical of surfactant-active ionic liquids (SAILs),²⁶ whereas ILs with very short alkyl chains do not display this behavior.⁵⁵ Therefore, it is crucial to investigate how ionic liquids with varying alkyl chain lengths influence the aggregate states of a drug molecule.

3.2.1. Effect of short alkyl chain ionic liquid. The effect of short alkyl chain ionic liquids on the aggregates of R6-BDP was investigated by gradually adding the cationic short alkyl chain IL C₂C₂im-Br. Fig. 3(a) shows the absorption spectra of the R6-BDP aggregates in the presence and absence of C₂C₂im-Br. As seen in Fig. 3(a), the absorption spectra shifted towards longer wavelengths (red shift), and the absorption intensity decreased upon the addition of C₂C₂im-Br to the R6-BDP aggregates. Interestingly, the absorption spectra became broadened as the concentration of C₂C₂im-Br increased. The FWHM (full width at half maximum) values of the absorption spectra for the R6-BDP aggregates and R6-BDP aggregates in the presence of 164.5 μM C₂C₂im-Br were determined to be 62 nm and 157 nm, respectively. The FWHM values suggest that the absorption spectra broadened significantly upon the addition of C₂C₂im-Br. The red shift and broadening of the absorption peaks in the presence of C₂C₂im-Br suggest that the addition of the short alkyl chain IL caused the R6-BDP aggregates to transition towards a more solidified form. Due to the presence of a small hydrophobic alkyl chain in C₂C₂im-Br, electrostatic interactions may dominate over van der Waals forces, causing the short alkyl chain ionic liquid to behave like a pure salt. This salt-like nature of C₂C₂im-Br facilitated the formation of organic aggregates of R6-BDP molecules. In this context, it is worth noting that Christau and coworkers⁵⁶ previously reported the salt-induced aggregation of citrate-coated gold nanoparticles (AuNPs) enclosed inside poly(N isopropyl-acrylamide) brushes implanted from plain substrates. They found that salt-induced AuNP aggregation resulted in nanoparticle assemblies that varied depending on the type of salt.⁵⁶ Thus, it is intriguing that short alkyl chain ionic liquids can behave like a salt,⁵⁷ which facilitates the formation of aggregates of R6-BDP molecules.

Fig. 3 (a) Absorption and (b) emission spectra of the aggregated state of R6-BDP with gradual addition C₂C₂im-Br.

Our interpretation of the mechanism of the absorption intensity quenching is further supported by the emission spectroscopy data for the R6-BDP aggregates. Quite interestingly, the fluorescence intensity of the R6-BDP aggregates gradually decreased with an addition of C₂C₂im-Br, as shown in Fig. 3(b). The broadening in the emission spectra and decrease in emission intensity upon the gradual addition of the short alkyl chain IL indicate that R6-BDP molecules were tending towards the solid form. The decrease in fluorescence intensity is primarily attributed to the intermolecular vibronic interactions, causing the nonradiative deactivation process, which possibly involves internal conversion, ISC or fluorescence quenching such as excitonic coupling and excimer formation.¹⁷ Enhancing the ISC process (S₁ → T_n) upon aggregation could allow the R6-BDP to act as a sensitizer, transferring its triplet excited state energy to molecular oxygen, thereby generating singlet oxygen (¹O₂), which would be very helpful for PDT.¹⁷ The generation of singlet oxygen in R6-BDP aggregates in the presence of C₂C₂im-Br is confirmed and discussed in Section 3.5.

Furthermore, the fluorescence quenching of the R6-BDP aggregates by C₂C₂im-Br was analysed using the Stern–Volmer plot.⁵⁸ This plot, which shows the relationship between F₀/F and C₂C₂im-Br concentration, is presented in Fig. 4. Based on the linear fit of the Stern–Volmer plot,⁵⁸ the quenching process appears to be either exclusively static or dynamic in nature (see Fig. 4). To distinguish whether the observed quenching behaviour is due to a static or dynamic process, fluorescence decay measurements of the R6-BDP aggregates in presence of C₂C₂im-Br were conducted using the TCSPC technique.

Fig. 4 Stern–Volmer plot illustrating the changes in the fluorescence intensity of R6-BDP aggregates with varying concentrations of the ionic liquid (IL). The black lines represent the error bars for the data points.

The representative fluorescence decay profiles of the R6-BDP aggregates, both with and without C₂C₂im-Br, are shown in Fig. 5, with the corresponding data provided in Table 3. The average lifetime values of the R6-BDP aggregates in the absence and presence of C₂C₂im-Br are 1.02 ns and 1.00 ns, respectively. These results suggest that the fluorescence lifetime of the R6-BDP aggregates did not change significantly upon the addition of the short alkyl chain IL, indicating that the quenching mechanism is purely static in nature.

Fig. 5 Fluorescence intensity decay profile of R6-BDP in the molecular form, aggregated form, and R6-BDP aggregates in the presence of C₂C₂im-Br and C₈C₈im-Br. Black lines represent the fitted data points.

3.2.2. Effect of medium and long alkyl chain ionic liquids. Further, we examined the effect of medium alkyl chain IL on the R6-BDP aggregates using a medium cationic alkyl chain symmetrical IL, C₄C₄im-Br. The absorption spectra of the R6-BDP aggregates with and without C₄C₄im-Br are presented in Fig. S10(a). Both the absorption intensity and absorption maxima of the R6-BDP aggregates did not change significantly upon the addition of C₄C₄im-Br as shown in Fig. S10(a). To better understand the role of C₄C₄im-Br, we also performed fluorescence spectroscopy. Fig. S10(b) represents the emission spectra of the R6-BDP molecules in aggregated form in the presence of C₄C₄im-Br. Similar to absorption, no significant changes in the emission maxima and emission intensity were observed for the R6-BDP aggregates upon the addition of C₄C₄im-Br. The data suggest that C₄C₄im-Br had no effect on the R6-BDP aggregates. This might be because it is a medium hydrophobic alkyl chain (n = 4) ionic liquid, which does not behave like a pure salt or it cannot form micelles⁵⁷ because both the electrostatic interaction and van der Waals interaction among the IL components may be balanced.

Moreover, to confirm this, we performed time-resolved fluorescence lifetime measurements. The average lifetime for the aggregated form and in the presence of C₄C₄im-Br was found to be 1.02 and 1.07, respectively (Table 3), which indicates that the change in fluorescence lifetime is not significant. This investigation suggests that medium alkyl chain ionic liquids do not have much effect on organic aggregates of R6-BDP.

Additionally, we extended our studies by using a longer cationic alkyl chain ionic liquid, C₈C₈im-Br, to investigate its effect on the R6-BDP aggregates. The representative absorption spectra of the R6-BDP aggregates, both with and without C₈C₈im-Br, are presented in Fig. 6(a). Upon the gradual addition of C₈C₈im-Br, the absorption intensity of the R6-BDP aggregates sharply increased and their absorption maximum shifted towards the shorter wavelength region (blue shift), as shown in Fig. 6(a). The observed increase in absorption intensity and the blue shift in absorption maximum indicate the dissociation of the aggregate molecules, transitioning towards a molecular form. Furthermore, we also performed a fluorescence study on the R6-BDP aggregates with the gradual addition of a longer alkyl chain IL (Fig. 6(b)). The addition of C₈C₈im-Br to the aggregated solution of R6-BDP was observed to cause a hypsochromic shift in its emission spectrum and its emission intensity increased by around 26-fold. This behaviour is likely due to the longer alkyl chain in C₈C₈im-Br, which enhances the hydrophobic interaction over the electrostatic interaction, thereby favouring the formation of micelles in the aqueous solution.⁵³ The significant blue shift in the emission spectrum and the increased emission intensity of the R6-BDP aggregates in the presence of C₈C₈im-Br suggest that the dissociated species are highly stable within the hydrophobic core of the micelles. The data also suggest that the R6-BDP aggregates dissociate and transition towards the molecular form in the presence of C₈C₈im-Br. This occurs due to the micellar effect of the longer alkyl chain IL. The presence of a long hydrophobic chain (n = 8) in C₈C₈im-Br allows greater compactness, enabling it to easily form micelles above its critical micelle concentration.⁵⁷ To determine the critical micelle concentration (CMC) value, we plotted a graph of fluorescence intensity versus the concentration of C₈C₈im-Br (Fig. S11). The inflection point of the sigmoidal curve in the graph is found to be approximately 10.3 μM, indicating that micelle formation occurred after the addition of 10.3 μM C₈C₈im-Br. This is due to the formation of micelles, which help in encapsulate the aggregates of R6-BDP molecules and break their aggregation.

Fig. 6 (a) Absorption and (b) emission spectra of aggregated state of R6-BDP with the gradual addition of C₈C₈im-Br.

To gain further insights into the increase in the fluorescence intensity of the R6-BDP aggregates, we also performed time-resolved fluorescence lifetime measurements. The fluorescence decay profile depicting the effect of a longer alkyl chain ionic liquid (C₈C₈im-Br) on the aggregates of R6-BDP is shown in Fig. 5. The fluorescence decay trace (excited at 500 nm) was fitted to triexponential functions. The lifetime value increased from τ = 1.02 ns to τ = 5.12 ns upon the addition of a longer alkyl chain ionic liquid (C₈C₈im-Br) to the R6-BDP aggregates, which is close to that of the molecular form of R6-BDP. This signifies that the aggregation was broken and it was tending towards the molecular form. The dissociation of aggregates into monomers is a key factor in enhancing the activity of drugs in biological systems, which makes this an intriguing observation.²⁶ Interestingly, the effect of the long alkyl chain IL significantly differed from that of the short alkyl chain IL. The short alkyl chain IL had the tendency of inducing aggregation, which we discussed earlier (Section 3.2.1). In contrast, the longer alkyl chain IL can able to form molecular species (Fig. 7). A dramatic colour change from pale pink to lime green under UV light was observed upon the addition of the longer alkyl chain IL (C₈C₈im-Br) to the R6-BDP aggregates, as shown in Fig. 7. In contrast, no colour change occurred when a medium alkyl chain ionic liquid (C₄C₄im-Br) was added to the R6-BDP aggregates. Interestingly, the colour transitioned completely from pale pink to colourless upon the addition of a short alkyl chain IL (C₂C₂im-Br) to the R6-BDP aggregates. Based on the above discussion, we conclude that the cationic alkyl chain length of symmetrical ILs plays a vital role in determining whether they facilitate the formation of organic aggregates or break them down into molecular forms. This strategy is particularly interesting given that it facilitates the formation of organic aggregates through the addition of short alkyl chain ILs, while also enabling their breakdown with the addition of long alkyl chain ILs. To further confirm our experimental observations, we extended our studies through microscopic analysis.

Fig. 7 Schematic of the self-assembly of R6-BDP upon the addition of C₂C₂im-Br, C₄C₄im-Br and C₈C₈im-Br.

3.3. Microscopic analysis of R6-BDP aggregates

To examine the morphology of the R6-BDP aggregates with and without ionic liquids, FESEM images were captured at ambient temperature (Fig. 8). The estimated particle size of the R6-BDP aggregates was found to be nearly 1.518 μm and the particles were round in shape with a rough surface (Fig. 8(a)). In the presence of the shorter alkyl chain IL, the particle size of the R6-BDP aggregates increased from 1.518 μm to 7.226 μm, with a corresponding change in shape from round to rod-like structures (Fig. 8(b)). These findings strongly support the previous observations from the photophysical studies (see Section 3.2.1), which indicated that the shorter alkyl chain IL promoted the formation of R6-BDP aggregates. Furthermore, the rod-shape structure of the R6-BDP aggregates in the presence of C₂C₂im-Br also confirmed that R6-BDP tended towards a more solidified structure in the presence of C₂C₂im-Br. However, the particle size of the R6-BDP aggregates was found to be around 2 μm, with no noticeable change in shape upon the addition of the medium alkyl chain ionic liquid. These findings suggest that C₄C₄im-Br had no substantial effect on the R6-BDP aggregates. Interestingly, the particle size of the R6-BDP aggregates declined from 1518 nm to 209 nm with the addition of the longer alkyl chain IL (C₈C₈im-Br). Additionally, the shape of the particles changed, becoming circular (Fig. 8(d)). This data indicates that the R6-BDP aggregates tended towards a molecular form in presence of C₈C₈im-Br. The results support the observations obtained from the spectroscopic investigation of the effect of the medium and long alkyl chain ILs on the R6-BDP aggregates (see Section 3.2.2).

Fig. 8 FESEM images of the (a) aggregation of R6-BDP. (b) Addition of C₂C₂im-Br to the aggregation of R6-BDP. (c) Addition of C₄C₄im-Br to the aggregation of R6-BDP. (d) Addition of C₈C₈im-Br to the aggregation of R6-BDP.

Additionally, to confirm the size of the particles obtained by FESEM, dynamic light scattering measurements were also performed for all these samples. The size distribution plots of the R6-BDP aggregates, both with and without liquids, are given in Fig. S12. It was observed that the size of the particle increased from 1557 nm to 2506 nm in the presence of the shorter alkyl chain ionic liquid (C₂C₂im-Br). Additionally, the polydispersity index (PDI) increased from 0.73 to 1.26 in the presence of C₂C₂im-Br, indicating that the particles grew in size with the addition of the short alkyl chain IL. This suggests that the R6-BDP molecules form larger aggregates in the presence of short chain ILs. However, on the addition of the medium alkyl chain IL (C₄C₄im-Br), the particle size changed from 1557 nm to 1799 nm. The PDI value also increased from 0.73 to 0.88, which is not significantly different for the change upon the addition of C₄C₄im-Br. Interestingly, the size of the particles decreased from 1557 nm to 528 nm and the PDI value also changed from 0.73 to 0.01 in the presence of the longer alkyl chain IL (C₈C₈im-Br), indicating the formation of molecular form. Similar observations were previously recorded based on FESEM data.

3.4. Theoretical calculations of the experimental results

The photophysical and aggregation behavior of molecules can be better understood by using theoretical calculations.^42,43 The ground-state structures of R6-BDP, including monomers, linear dimer and B–F⋯H linear dimers were completely optimized in the gas phase. Theoretical calculations were performed at the DFT level with the Gaussian 09 quantum mechanical package, employing the B3LYP functional in combination with the 6-31+G(d,p) basis set. The time-dependent DFT (TD-DFT)⁴⁴ formalism was used to estimate the excitation energies at the TD-B3LYP/6-31+G(d,p) level of theory.

The self-assembled structure of R6-BDP was studied by optimizing its monomer, linear dimers and B–F⋯H linear dimers to help understand its aggregation behaviour, as shown in Fig. 9. As seen in Fig. 9(c), two R6-BDP molecules are connected via B–F⋯H hydrogen bonding. In this case, the two molecules are interlinked through a head-to-head fashion. The hydrogen bond distance between B–F⋯H is 2.19 Å, indicating strong hydrogen bonding interaction. This hydrogen bond linkage further supports the assembly of H-type aggregates of the R6-BDP molecules, consistent with the experimental observations.

Fig. 9 Optimized geometries of R6-BDP, (a) monomers, (b) linear dimers, and (c) B–F⋯H linear dimers.

The excited-state energy difference was calculated for both the monomer and dimer of R6-BDP. Fig. 10 represents the energy gap between the singlet (S₁) and triplet (T₁) energy states for both the monomer and dimer of R6-BDP. According to Kasha's rule, when the S₁ state is populated either through direct photoirradiation or via internal conversion from higher excited singlet states, it is difficult for the monomer to transition from the S₁ state to the T₁ state because of the significant energy difference of 1.2786 eV between them, as shown in Fig. 10. Interestingly, in the dimer, the associated S₁–T₁ and S₁–T₂ energy difference is significantly reduced to 1.0758 eV and 1.0749 eV, respectively (see Table S2). According to the energy gap law,⁵⁹ inter system crossing (ISC) is more efficient when the singlet and triplet states have similar energy levels. However, based on the classical Marcus theory, the rate constants for ISC are inversely proportional to the square of the pertinent energy difference (ΔE) and directly proportional to the square of the spin–orbit coupling (H_if).⁶⁰ Given that heavy atom-free organic systems often have small spin–orbit couplings, the energy differences become the key factor governing the ISC efficiency.⁶¹ Thus, the reduction in the energy gap between the S₁ and T₁ states may account for the aggregation-induced increase in the ISC efficiency observed in the present study, as discussed vide infra.

Fig. 10 Energy level illustrations for the excited singlet (black) and triplet (blue) states calculated using TD-DFT for the R6-BDP monomer (left) and dimer (right) structures.

The computed λ_max values (Fig. 11) for the R6-BDP monomer and dimer forms are presented in Table 4. The calculated λ_max values for the monomer and dimer were found to be 448.68 nm and 456.81 nm, respectively. Although these values differ significantly from the experimental absorption maxima, both approaches show a consistent trend, i.e., a red shift in the absorption spectrum upon aggregation. Notably, similar discrepancies between the theoretical and experimental absorption maxima for both the molecular and aggregated forms have been reported previously.^40,62,63 These studies also emphasize that although the absolute values may vary, the observed trends remain consistent. Again, the S₀–S₁ transition is referred to as an optical transition. The theoretical results indicate that the hydrogen bonding between the R6-BDP molecules is primarily responsible for the observed red shift of 8.13 nm upon dimer formation. Furthermore, the fluorescence intensity is directly proportional to the oscillator strength (f).⁶⁴ The R6-BDP monomer exhibits a stronger oscillator strength than the dimer (Table 4), suggesting that the monomer has a higher fluorescence intensity than the dimer. Both findings align well with the experimental results.

Fig. 11 Computed λ_max for the R6-BDP monomer and dimer at the TD-B3LYP/6-31+G(d,p) level of theory.

Table 4 Computed λ_max for the monomer and dimer forms of R6-BDP

System	λ max (nm)	f (Oscillator strength)	Transitions
Monomer	448.68	0.586	H–L; 94.54%
			H-1–L; 5.45%
Dimer	456.81	0.020	H–L; 64.59%
			H-1–L-1; 35.41%

---

3.5. Singlet oxygen (¹O₂) generation and photostability

As discussed earlier in Section 3.2.1, the shorter alkyl chain ionic IL (C₂C₂im-Br) facilitates the formation of more solidified structures of R6-BDP aggregates by promoting their aggregation. This helps in decreasing the energy gaps among the singlet and triplet excited states, as proven by the theoretical study (see Section 3.4), which enhances the ISC efficiency. Due to the increase in the ISC efficiency of the R6-BDP aggregates upon the addition of the shorter alkyl chain IL, their singlet oxygen (¹O₂) generation efficiency should be enhanced, which was assessed by a photochemical technique. This study included the addition of C₂C₂im-Br (short alkyl chain IL) to the R6-BDP aggregates and DPBF was added as a singlet oxygen scavenger. The photodegradation of DPBF upon photoirradiation of the BODIPY aggregates was measured by monitoring the decrease in the absorbance of DPBF at 410 nm. It has been observed that the absorption of DPBF is significantly reduced in the presence of singlet oxygen.⁶⁵ Firstly, the photodegradation of DPBF was examined in the presence of the R6-BDP aggregates (Fig. 12(b)). The absorption peak intensity of DPBF in the presence of the R6-BDP aggregates showed a slight decrease under laser photoirradiation (Fig. 12(b)). However, its absorption peak intensity at 410 nm decreased rapidly upon addition of C₂C₂im-Br IL on the aggregation of R6-BDP (Fig. 12(c)), indicating that the R6-BDP aggregates generated singlet oxygen more efficiently in the presence of the short alkyl chain IL. The decrease in DPBF concentration over irradiation time was also evident through the drastic color change in the solution, shifting from green to colorless (see inset in Fig. 12(c)). Furthermore, no reaction was observed between the aggregated form +C₂C₂im-Br and DPBF as the absorption band of the photosensitizer at around 525 nm remained unchanged, as shown in Fig. 12(c). The decay rates of DPBF upon the addition of the aggregated form +C₂C₂im-Br are higher than that of the molecular and aggregated forms of R6-BDP (Fig. 12(d)), indicating the efficient photosensitization ability of the singlet oxygen generated by the R6-BDP aggregates in the presence of the short alkyl chain ionic liquid. This result demonstrates the excellent photostability of the aggregated form +C₂C₂im-Br, enabling its use for prolonged laser exposure (high dose) during PDT. Therefore, these results suggest that the impact of C₂C₂im-Br (short alkyl chain IL) on the enhancement of singlet oxygen generation by R6-BDP molecules may play an important role as a photosensitizer in photodynamic therapy.

Fig. 12 Absorbance spectra of (a) DPBF with time, (b) DPBF in the presence of R6-BDP aggregates, (c) DPBF in the presence of aggregated form +C₂C₂im-Br and (d) comparison of the decay rates of DPBF, R6-BDP molecular and aggregated forms, and the aggregated form +C₂C₂im-Br at 410 nm, with higher rates indicating more efficient photosensitization capabilities of ¹O₂.

3.6. In vitro photodynamic therapy evaluation in cancer cells

An in vitro photodynamic therapeutic investigation was performed using the MTT method both with and without illumination using 490 nm LED light. The cytotoxicity of the R6-BDP aggregates with the addition of the short alkyl chain IL (C₂C₂im-Br) on HeLa cells was assessed at concentrations of 0.2 μM, 2 μM, 4 μM and 8 μM. As shown in Fig. 13, the cytotoxicity was low for aggregates of R6-BDP even at the dose of 8 μM compared to aggregated form +C₂C₂im-Br in the dark as well as under light irradiation. Furthermore, the addition of the short alkyl chain IL to the R6-BDP aggregates incubated at concentrations of 0.2 μM, 2 μM, 4 μM and 8 μM in the dark caused 19%, 23%, 39% and 44% cell death in the HeLa cells, respectively. In contrast, a significant increase in cell death was observed after exposure to light, as shown in Fig. 13. Therefore, R6-BDP aggregation in the presence of C₂C₂im-Br could potentially enhance its role as a photosensitizer for photodynamic therapy.

Fig. 13 In vitro cytotoxicity of the aggregated form of R6-BDP and aggregated form +C₂C₂im-Br on HeLa cells in the dark and under light illumination.

3.7. Intracellular singlet oxygen (¹O₂) detection

To explore the therapeutic mechanism of the effect of ionic liquids on R6-BDP aggregate-mediated PDT, we further verified the generation of singlet oxygen in HeLa cells by fluorescence microscopy imaging using 2,7-dichlorofluorescein diacetate (DCFH-DA) as an reactive oxygen species (ROS) indicator.⁴⁶ Intracellular esterase can hydrolyse DCFH-DA to create DCFH, which resists cell membrane penetration and allows ROS within the cell, subsequently oxidizing DCFH to create fluorescent DCF (2,7-dichlorofluoresce) to emit bright green fluorescence.⁶⁶ HeLa cells were incubated with the DCFH-DA + aggregated form of R6-BDP and DCFH-DA+ aggregated form of R6-BDP + C₂C₂im-Br in a dose-dependent manner at concentrations of 0.2 μM, 2 μM, 4 μM and 8 μM with and without light irradiation.

As shown in Fig. 14, the HeLa cells exhibited high DCF green fluorescence in the presence of the aggregated form of R6-BDP following 490 nm irradiation, indicating that the cells contained an abundance of ¹O₂. The green fluorescence was greater at a higher dose (4 μM and 8 μM) under irradiation with light, indicating more singlet oxygen generation at the doses of 4 μM and 8 μM. In contrast, the green fluorescence was less in the dark compared to that in the light. This is an indication of an efficient photosensitizer that is helpful for PDT. After the incubation of HeLa cells with aggregated form +C₂C₂im-Br, the production of singlet oxygen was greater under light, showing more green fluorescence than that in the dark. In addition, more singlet oxygen was generated than that by the R6-BDP aggregates. The generation of more singlet oxygen after the addition of the short alkyl chain ionic liquid (C₂C₂im-Br) to the R6-BDP aggregates indicates the formation of a more solidified form of R6-BDP. This helps in transferring energy to molecular oxygen, producing singlet oxygen due to the low energy gap between the singlet and triplet states. These data clearly indicate that ionic liquids have potential to increase the intracellular generation of singlet oxygen via the aggregation of R6-BDP for PDT treatment.

Fig. 14 Confocal fluorescence images of intracellular ¹O₂ production in HeLa cells using DCFH-DA as an indicator.

4. Conclusion

In this study, we explored the aggregation behaviour of a BODIPY molecule derivative (R6-BDP) through various spectroscopic, microscopic, and computational techniques. We also examined the influence of the cationic chain length of symmetrical ionic liquids on the aggregation of R6-BDP. Three ionic liquids with varying alkyl chain lengths on the cationic moiety were synthesized to examine how increasing the cationic alkyl chain length affects the aggregation of R6-BDP. Photophysical studies were conducted on both the molecular and aggregated forms of the R6-BDP molecule. The absorption profile of the R6-BDP aggregates exhibited a red shift compared to its molecular form, and the spectral data suggested the formation of J-aggregates. Notably, we observed the formation of “red-shifted H-aggregates”, which is an uncommon phenomenon in this context. Further investigation revealed that the addition of a shorter alkyl chain IL (C₂C₂im-Br) promoted the stronger aggregation of R6-BDP, decreasing the energy gap between its singlet and triplet excited states. This enhanced aggregation improved the generation of singlet oxygen, which can have important implications for photodynamic therapy in cancer treatment. In contrast, the presence of a longer alkyl chain IL (C₈C₈im-Br) led to dissociation of the aggregates, converting them back to the molecular form. Notably, no significant effect on R6-BDP aggregation was observed with a medium-length alkyl chain IL (C₄C₄im-Br). These findings highlight the significant role of the cation chain length of ILs in modulating the aggregation state of R6-BDP, with potential applications in PDT and drug delivery. Short alkyl chain ionic liquids appear to favour the formation of more stable organic aggregates, while longer chains disrupt this aggregation. The theoretical DFT and TD-DFT results also supported the experimental findings. Biological studies revealed that the addition of a shorter alkyl chain IL to the R6-BDP aggregates enhanced singlet oxygen generation, improving their photodynamic therapeutic performance. The solidified form of R6-BDP aggregates, resulting from the addition of a short alkyl chain IL, boosted singlet oxygen production and photocytotoxicity. These findings demonstrate the potential of ionic liquids in optimizing R6-BDP aggregates for effective PDT treatment. Further research is required to expand our understanding of these effects and explore their therapeutic potential.

Conflicts of interest

There are no conflicts to declare.

Data availability

¹H NMR spectrum of R6-BDP, ¹³C{¹H} NMR spectrum of R6-BDP, ¹H NMR spectrum of C₂C₂im-Br, HR-MS analysis of C₂C₂im-Br, ¹H NMR spectrum of C₄C₄im-Br, HR-MS analysis of C₄C₄im-Br, ¹H NMR spectrum of C₈C₈im-Br, HR-MS analysis of C₈C₈im-Br, absorption spectra of R6-BDP in molecular and aggregated forms, ultraviolet-visible absorption and fluorescence spectra of aggregated state of R6-BDP with gradual addition C₄C₄im-Br, fluorescence intensity of R6-BDP aggregates vs concentration of IL (C₈C₈im-Br), DLS-based particle size data analysis of R6-BDP aggregates, aggregation + C₂C₂im-Br, aggregation + C₄C₄im-Br, aggregation + C₈C₈im-Br, and tri-exponential fitting parameters and average lifetimes for the aggregated form and aggregated form in the presence of C₂C₂im-Br, C₄C₄im-Br and C₈C₈im-Br. Excitation energies (eV) of the S₁, S₂, S₃, T₁, T₂, and T₃ states of monomer and dimers of R6-BDP. See DOI: https://doi.org/10.1039/d5cp01890k

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

Acknowledgements

This work was supported by the Department of Science and Technology (DST), Government of India, New Delhi (award no. DST/INSPIRE/04/2022/002706). N. N. Das and Dr D. Majhi express their gratitude to the DST INSPIRE Faculty Program for the fellowship/research grant and financial support. Dr D. Majhi also thank the National Institute of Technology, Tiruchirappalli (NITT), for providing the necessary infrastructure facilities and the ‘Central Facility for Time-Resolved Spectroscopy and Microscopy’ at IIT Bombay for the time-resolved fluorescence decay measurements.

References

1. J. Bañuelos, Chem. Rec., 2016, 16, 335–348.
2. S. Mula, BODIPY: A Unique Dye for Versatile Optical Applications, Springer Nature, 2024, vol. 1, pp. 369–399.
3. A. Loudet and K. Burgess, Chem. Rev., 2007, 107, 4891–4932.
4. M. Gupta, S. Mula, M. Tyagi, T. K. Ghanty, S. Murudkar, A. K. Ray and S. Chattopadhyay, Chem. – Eur. J., 2013, 19, 17766–17772.
5. M. Liu, S. Ma, M. She, J. Chen, Z. Wang, P. Liu, S. Zhang and J. Li, Chin. Chem. Lett., 2019, 30, 1815–1824.
6. H. Lu, J. Mack, Y. Yang and Z. Shen, Chem. Soc. Rev., 2014, 43, 4778–4823.
7. G. Ulrich, R. Ziessel and A. Harriman, Angew. Chem. Int. Ed., 2008, 47, 1184–1201.
8. S. Mula and M. Koli, Chem. Med. Chem., 2024, 19, e202400041.
9. I. S. Yadav and R. Misra, J. Mater. Chem. C, 2023, 11, 8688–8723.
10. M. K. Choudhary and S. Mula, New J. Chem., 2023, 47, 9045–9049.
11. Y. Dong, P. Kumar, P. Maity, I. Kurganskii, S. Li, A. Elmali, J. Zhao, D. Escudero, H. Wu, A. Karatay, O. F. Mohammed and M. Fedin, Phys. Chem. Chem. Phys., 2021, 23, 8641–8652.
12. J. Zhao, K. Chen, Y. Hou, Y. Che, L. Liu and D. Jia, Org. Biomol. Chem., 2018, 16, 3692–3701.
13. M. Koli, S. Gupta, S. Chakraborty, A. Ghosh, R. Ghosh, A. P. Wadawale, T. K. Ghanty, B. S. Patro and S. Mula, Chem. – Eur. J., 2023, 29, e202301605.
14. X. Zhang, Z. Wang, Y. Hou, Y. Yan, J. Zhao and B. Dick, J. Mater. Chem. C, 2021, 9, 11944–11973.
15. D. Simionesie, G. O’Callaghan, J. R. H. Manning, T. Düren, J. A. Preece, R. Evans and Z. J. Zhang, Polycyclic Aromat. Compd., 2023, 43, 3790–3809.
16. Y. S. Marfin, E. A. Banakova, D. A. Merkushev, S. D. Usoltsev and A. V. Churakov, J. Fluoresc., 2020, 30, 1611–1621.
17. Y. F. Kang, W. K. Chen, K. X. Teng, L. Y. Wang, X. C. Xu, L. Y. Niu, G. Cui and Q. Z. Yang, CCS Chem., 2022, 4, 3516–3528.
18. D. Simionesie, G. O’Callaghan, J. R. H. Manning, T. Düren, J. A. Preece, R. Evans and Z. J. Zhang, Polycyclic Aromat. Compd., 2023, 43, 3790–3809.
19. Y. S. Marfin, E. A. Banakova, D. A. Merkushev, S. D. Usoltsev and A. V. Churakov, J. Fluoresc., 2020, 30, 1611–1621.
20. K. Kataoka, G. S. Kwon, M. Yokoyama, T. Okano and Y. Sakurai, J. Controlled Release, 1993, 24, 19–132.
21. K. Kataoka, A. Harada and Y. Nagasaki, Adv. Drug Delivery Rev., 2012, 64, 37–48.
22. J. Chen, S. Li, Z. Wang, Y. Pan, J. Wei, S. Lu, Q. W. Zhang, L. H. Wang and R. Wang, Chem. Sci., 2021, 12, 7727–7734.
23. N. Rapoport, Prog. Polym. Sci., 2007, 32, 962–990.
24. S. Ghosh, C. Ghatak, C. Banerjee, S. Mandal, J. Kuchlyan and N. Sarkar, Langmuir, 2013, 29, 10066–10076.
25. S. Mandal, J. Kuchlyan, S. Ghosh, C. Banerjee, N. Kundu, D. Banik and N. Sarkar, J. Phys. Chem. B, 2014, 118, 5913–5923.
26. S. S. Dandpat and M. Sarkar, Phys. Chem. Chem. Phys., 2015, 17, 13992–14002.
27. L. G. Celia-Silva, R. N. Martins, A. J. P. Carvalho, J. P. P. Ramalho, P. Morgado, E. J. M. Filipe and L. F. G. Martins, Energy Fuels, 2022, 36, 9048–9065.
28. R. N. Martins, L. G. Celia-Silva, J. P. P. Ramalho, P. Morgado, E. J. M. Filipe and L. F. G. Martins, Energy Fuels, 2024, 38, 3693–3712.
29. S. Rani, D. Bagchi, U. Pal, M. Kumari, M. Sharma, A. Bera, J. Shabir, S. K. Pal, T. Saha-Dasgupta and S. Mozumdar, ACS Omega, 2020, 5, 25582–25592.
30. C. Ghatak, V. G. Rao, S. Mandal, S. Ghosh and N. Sarkar, J. Phys. Chem. B, 2012, 116, 3369–3379.
31. D. Majhi, A. V. Komolkin and S. V. Dvinskikh, Int. J. Mol. Sci., 2020, 21, 1–14.
32. S. P. Verevkin, D. H. Zaitsau, R. N. Nagrimanov, Y. Zhang, E. J. Maginn and A. Stark, J. Mol. Liq., 2024, 395, 123850.
33. W. Zheng, A. Mohammed, L. G. Hines, D. Xiao, O. J. Martinez, R. A. Bartsch, S. L. Simon, O. Russina, A. Triolo and E. L. Quitevis, J. Phys. Chem. B, 2011, 115, 6572–6584.
34. D. Majhi, A. Pabbathi and M. Sarkar, J. Phys. Chem. B, 2016, 120, 193–205.
35. E. Binetti, A. Panniello, L. Triggiani, R. Tommasi, A. Agostiano, M. L. Curii and M. Striccoli, J. Phys. Chem. B, 2012, 116, 3512–3518.
36. D. Majhi and M. Sarkar, Phys. Chem. Chem. Phys., 2017, 19, 23194–23203.
37. A. Ghosh, C. K. De, T. Chatterjee and P. K. Mandal, Phys. Chem. Chem. Phys., 2015, 17, 16587–16593.
38. S. Gorai, R. Agrawal, R. Ghosh and S. Mula, Chem. – Eur. J., 2024, 30, e202402669.
39. A. Dimitrijević, N. Zec, N. Zdolšek, S. Dožić, A. Tot, S. Gadžurić, M. Vraneš and T. Trtić-Petrović, J. Ind. Eng. Chem., 2016, 40, 152–160.
40. D. Majhi, S. K. Das, P. K. Sahu, Md. S. Pratik, A. Kumar and M. Sarkar, Phys. Chem. Chem. Phys., 2014, 16, 18349–18359.
41. M. Z. K. Baig, D. Majhi, R. N. P. Tulichala, M. Sarkar and M. Chakravarty, J. Mater. Chem. C, 2017, 5, 2380–2387.
42. Y. Kim, T. Jun, S. V. Mulay, S. T. Manjare, J. Kwak, Y. Lee and D. G. Churchill, Dalton Trans., 2017, 46, 4111–4117.
43. N. Gupta, S. I. Reja, V. Bhalla, M. Gupta, G. Kaurb and M. Kumar, J. Mater. Chem. B, 2016, 4, 1968–1977.
44. A. H. Pandith, N. Islam, Z. F. Syed, S. Rehman, S. Bandaru and A. Anoop, Chem. Phys. Lett., 2011, 516, 199–203.
45. T. Mosmann, J. lmmunol. Methods, 1983, 65, 55–63.
46. R. Cathcart, E. Schwiers and N. B. Ames, Anal. Biochem., 1983, 134, 111–116.
47. M. Kasha, H. R. Rawls and M. A. Bayoumi, Pure Appl. Chem., 1965, 11, 371–392.
48. N. J. Hestand and F. C. Spano, Chem. Rev., 2018, 118, 7069–7163.
49. F. C. Spano, Acc. Chem. Res., 2010, 43, 429–439.
50. C. Zheng, C. Zhong, C. J. Collison and F. C. Spano, J. Phys. Chem. C, 2019, 123, 3203–3215.
51. Z. Liu, Z. Jiang, M. Yan and X. Wang, Front. Chem., 2019, 7, 1–16.
52. S. Ghosh, B. Jana, A. Ghosh, D. M. Guldi and A. Patra, J. Phys. Chem. Lett., 2021, 12, 3424–3430.
53. R. Dutta, S. Kundu and N. Sarkar, Biophys. Rev., 2018, 10, 861–871.
54. C. K. De, A. Ghosh and P. K. Mandal, J. Phys. Chem. B, 2022, 126, 1551–1557.
55. M. N. Vaghela, V. N. Sastry and K. V. Aswal, Colloid Polym. Sci., 2011, 289, 309–322.
56. S. Christau, T. Moeller, J. Genzer and R. Koehler, Macromolecules, 2017, 50, 7333–7343.
57. M. Blesic, M. H. Marques, N. V. Plechkova, K. R. Seddon, L. P. N. Rebelo and A. Lopes, Green Chem., 2007, 9, 481–490.
58. M. H. Gehlen, J. Photochem. Photobiol. C Photochem. Rev., 2020, 42, 100338.
59. M. C. Marian, Annu. Rev. Phys. Chem., 2021, 72, 617–640.
60. J. T. Penfold, E. Gindensperger, C. Daniel and C. M. Marian, Chem. Rev., 2018, 118, 6975–7025.
61. L. Estergreen, A. R. Mencke, D. E. Cotton, N. V. Korovina, J. Michl, S. T. Roberts, M. E. Thompson and S. E. Bradforth, Acc. Chem. Res., 2022, 55, 1561–1572.
62. F. Sabuzi, M. Stefanelli, D. Monti, V. Conte and P. Galloni, Molecules, 2020, 25, 133.
63. Q. Wan, J. Xia, W. Lu, J. Yang and C. M. Che, J. Am. Chem. Soc., 2019, 141, 11572–11582.
64. K. Magesh, L. Y. Hsun, S. P. Wu and S. Velmathi, Appl. Bio Mater., 2024, 7, 5679–5688.
65. Y. C. Lai, S. Y. Su and C. C. Chang, ACS Appl. Mater. Interfaces, 2013, 5, 12935–12943.
66. N. M. Idris, M. K. Gnanasammandhan, J. Zhang, P. C. Ho, R. Mahendran and Y. Zhang, Nat. Med., 2012, 18, 1580–1585.

---