BOPHYs versus BODIPYs: A comparison of their performance as effective multi-function organic dyes

Abstract

The computationally-aided photophysical and lasing properties of a selected battery of BOPHYs are described and compared to those of related BODIPY counterparts. The present joined theoretical-experimental study helps to put into context the weaknesses and strengths of both dye families under different irradiation conditions. The chemical versatility of the BOPHY scaffold has been also comparatively explored to modulate key photonic properties towards the development of red-emitting dyes, chiroptical dyes and singlet oxygen photosensitizers. Thus, BOPHY BINOLation by fluorine substitution with enantiopure BINOLs endows the BOPHY chromophore with chiroptical activity, as supporting by the simulated circular dichroism, decreasing deeply its fluorescent response due to the promotion of fluorescence-quenching intramolecular charge transfer (ICT). Interestingly, the sole alkylation of the BOPHY core strongly modulates the promotion of ICT, allowing the generation of highly bright BINOL-based BOPHY dyes. Moreover, 3,3′-dibromoBINOLating BOPHYs can easily achieve singlet-oxygen photogeneration, owing to spin-orbit coupling mediated by heavy-atom effect feasible in view of the theoretically predicted disposition of the bromines surrounding the chromophore. From this background, we have established the master guidelines to design bright fluorophores and laser dyes, photosensitizers for singlet oxygen production and chiroptical dyes based on BOPHYs. The possibility to finely mix and balance such properties in a given molecular scaffold outstands BOPHYs as promising dyes competing with the well-settled BODIPY dyes.

1. Introduction

The modern avenues in organic chemistry have boosted a renewed and challenging interest in dye chemistry [1,2]. Currently, organic chromophores deserve a great deal of attention, and many efforts are being made to design organic dyes with tailor-made molecular structures and photophysical properties, to be applied in advanced (bio) photonic applications [3,4]. Indeed, the main objective in many active research projects is the development of organic molecular fluorophores with tunable key photophysical properties (e.g., absorption and emission wavelengths or fluorescence response) and high stability for longlasting operative lifetime. Besides, specific functionalities for additional capabilities are searched, such as recognition sites for binding target species in sensing applications [5], chiral moieties towards chiroptical applications [6], or heavy atoms to promote intersystem crossing (ISC) allowing singlet-oxygen photogeneration for therapeutic applications [7], among others. The chemical versatility of the organic chromophores makes them suitable scaffolds to achieve such tailor-made multi-functionalization in a single molecular-dye structure. In fact, organic fluorophores are intensively applied as molecular probes in fluorescence bioimaging [8], one of the most powerful techniques for diagnosis, as well as for tracking biochemical processes in real time, even at the single-molecule level, thanks to the latest technological advances on fluorescence microscopy (super-resolution microscopy or nanoscopy) [9–11].

Among all the available organic dyes with adjustable key properties fulfilling the specific demands of the different application fields, those known as BODIPYs (boron dipyrromethenes; see Fig. 1) are in the forefront [12]. Indeed, a quick survey of the bibliography reveals their excellent versatility for their use in a myriad of (bio)technological fields, from diagnosis [13], biosensing or therapy in (bio)medicine [14], to energy conversion in photovoltaic devices [15] or lasers [16]. The key of the success of the BODIPY dyes relies mainly on the chemical versatility of their boron-dipyrrin core [17,18], easy to obtain and readily amenable to a wide pool of synthetic transformations allowing not only the fine modulation of key photophysical signatures, but also the consecution of additional functions for advanced photonic applications [19].

Fig. 1.

BOPHYs vs. BODIPYs: Related synthetic access from similar pyrrolebased precursors, and related structures and electronic distributions (for computed electrostatic-potential maps, negative vs. positive charges are highlighted in red and blue, respectively).

In the last years, a new family of fluorescent dyes have emerged with remarkable impact: the BOPHYs (bis(difluoroboron) 1,2-bis [(pyrrol-2-yl)methylene)hydrazines; see Fig. 1) [20,21]. These new dyes are structurally close to BODIPYs, but they have two chelating positions amenable for linking boron. Thus, the synthetic access to both types of related dyes involves boron chelation of a π-conjugated di-pyrrole-based system, but, in the BOPHY case, it includes a central hydrazine-based spacer (Fig. 1).

The close similarity between BODIPYs and BOPHYs (see Fig. 1), as well as the novelty of the latter dyes (first reported in 2011), explains the growing interest in BOPHY dyes [22,23]. In this regard, their photonic performance in different issues previously exploited in BODIPYs has been reported (e.g., non-linear optics [24], chirality [25], sensing [26], red emission [27], energy transfer [28] or solar energy conversion [29]), but not its final competing application in a specific photonic tool. However, one should expect key differences in the photonic performances in both types of dyes, due to significant differences in their structures (e.g., compare the C_2H- vs. C_2v-symmetric electronic distribution in BOPHYs vs. flat BODIPYs, in Fig. 1). Therefore, against this background, it is advisable to compare the photophysical behavior of both types of dyes, mainly their performance as fluorophores, laser dyes, chiroptical dyes and singlet-oxygen photo-sentitizers, since these are the most demanded photonic capabilities from organic dyes.

This aim will be firstly addressed in this work by revisiting the photophysical signatures of alkylated BOPHYs (Fig. 2) [20,21], and then contributing with new experimental measurements and computational simulations, including the unprecedented study of their behavior as laser dyes. In fact, dye photostability under strong irradiation, which is key in lasing performance, is also a fundamental parameter for any practical photonic applications. Secondly, the work will focus on exploring the application to BOPHYs of two of the most important transformations (chromophoric π-extension and BINOL-derivatization at boron) used in BODIPYs to finely modulate key (chiro)photophysical properties, such as emission wavelength, fluorescence, singlet-oxygen photogeneration, lasing, circular dichroism and circularly polarized luminescence. To this aim, a selected battery of π-extended and BINOL-derived BOPHYs (Fig. 2) has been synthesized and thoroughly studied in comparison to analogous BODIPYs.

Fig. 2.

Molecular structures of the studied BOPHYs.

From such exhaustive photophysical, quantum mechanical, laser and chiroptical study, we intend to establish master guidelines to optimize the performance of BOPHYs as laser dyes, fluorescent dyes, chiroptical dyes and photodynamic therapy agents. The purpose is to identify the weaknesses and strengths of these dyes in comparison with related BODIPYs, to settle which one is the best current multi-function organic chromophore.

2. Experimental

2.1. Materials and methods

2.1.1. Synthesis

Common solvents were dried and distilled by standard procedures. All starting materials and reagents were obtained commercially and used without further purifications. Elution flash chromatography was conducted on silica gel (230–400 mesh ASTM). Thin layer chromatography (TLC) was performed on silica gel plates (silica gel 60 F254, supported on aluminum). Melting points were measured with a Gallekamp apparatus. NMR spectra were recorded in CDCl₃ on a Brucker DPX 300 MHz spectrometer at 20 °C. Chemical shifts are reported in units of ppm relative to the solvent residue peaks (CDCl₃, δ = 7.260 ppm for ¹H, 77.16 for ¹³C and CD₂Cl₂, δ = 7.320 ppm for ¹H, 54.00 for ¹³C). DEPT-135 NMR experiments were used for the assignation of the type of carbon nucleus (C, CH, CH₂, CH₃). Complex spin-system signals were additionally simulated using MestRe-C (Cobas, C.; Cruces, J.; Sardina, J., MestRe-C program version 2.3). FTIR spectra were recorded with Bruker Alpha-7 spectrometer from neat samples using ATR technique. IR bands are given in cm⁻¹. High-resolution mass spectrometry (HRMS) was performed using the ESI technique for the ionization and ion tramp (positive mode) for the detection. Optical rotations in chloroform solution (dye concentration, c, expressed in g/100 mL) were recorded at 293 K on an Anton Paar MCP 100 polarimeter.

2.1.2. Spectroscopic measurements

The photophysical properties were registered using quartz cuvettes with optical pathways of 1 cm in diluted solutions (around 2·10⁻⁶ M), prepared by adding the corresponding solvent to the residue from the adequate amount of a concentrated stock solution in acetone, after vacuum evaporation of this solvent. Ultraviolet–visible (UV–vis) absorption and fluorescence spectra were recorded on a Varian model CARY 4E spectrophotometer and an Edinburgh Instruments spectrofluorimeter (model FLSP920), respectively. Fluorescence quantum yields (ϕ) were obtained using PM650 as reference (Exciton, ϕ^r = 0.1 in ethanol [30]) for red-shifted BOPHYs 4a and 5a, and Coumarine 1 (Kodak, ϕ^r = 0.75) and Coumarine 152 (Kodak, ϕ^r = 0.2) in ethanol [31] for the rest of the herein tested BOPHYs. Radiative decay curves were registered with the time correlated single-photon counting technique, as implemented in the aforementioned spectrofluorimeter. Fluorescence emission was monitored at the maximum emission wavelength, by means a microchannel plate detector (Hamamatsu C4878) of picosecond time-resolution (20 ps), after excitation with a Fianium pulsed laser (time resolution of around 150 ps). The fluorescence lifetime (τ) was obtained after the deconvolution of the instrumental response signal from the recorded decay curves by means of an iterative method. The goodness of the exponential fit was controlled by statistical parameters (chi-square) and the analysis of the residuals. Radiative (k_fl) and non-radiative (k_nr) rate constants were calculated as follows:

$$k_{\text{fl}}=\varphi /\tau ;k_{\text{nr}}=\left(1-\varphi \right)/\tau .$$

Laser efficiency was evaluated from concentrated solutions (milli-molar) of dyes in ethyl acetate contained in 1-cm optical-path rectangular quartz cells carefully sealed to avoid solvent evaporation during experiments. The liquid solutions were transversely pumped with a wavelength tunable Optical Parametric Oscillator (OPO) coupled to the third harmonic (355 nm) of a Q-switched Nd:YAG laser (Lotis TII 2134) at a repetition rate of 1 Hz. The exciting pulses were line-focused onto the cell using a combination of positive and negative cylindrical lenses (f = 15 cm and f = −15 cm, respectively) perpendicularly arranged. The plane parallel oscillation cavity (2 cm length) consisted of a 90% reflectivity aluminium mirror acting as back reflector, and the lateral face of the cell acting as output coupler (4% reflectivity). The pump and output energies were detected by an Ophir power meter. The photostability of the dyes in ethyl acetate solution was evaluated by using a pumping energy and geometry exactly equal to that of the laser experiments. We used spectroscopic quartz cuvettes with 0.1 cm optical to allow for the minimum solution volume (VS = 40 μL) to be excited. The lateral faces were grounded, whereupon no laser oscillation was obtained. Information about photostability was obtained by monitoring the decrease in laser-induced fluorescence (LIF) intensity. In order to facilitate comparisons independently of the experimental conditions and sample, the photostability figure of merit was defined as the accumulated pump energy absorbed by the system (E_dose), per mole of dye, before the output energy falls to a 90% its initial value. In terms of experimental parameters, this energy dose, in units of GJ mol⁻¹, can be expressed as:

$$E_{\mathit{\text{dose}}}^{90\%}=\frac{E_{\mathit{\text{pump}}}\cdot \left(1-{10}^{-\epsilon CL}\right)\cdot {\sum }_{\#\text{pulses}}f}{C{V}_{\text{S}}}$$

where E_pump is the energy per pulse in units of GJ, C is the molar concentration, ε is the molar absorption coefficient in units of M⁻¹ cm⁻¹, L is the depth of the cuvette expressed in cm, V_S is the solution volume, in liters, within the cuvette, and f is the ratio between the LIF intensity after # pulses and the LIF intensity in the first pulse. It can be shown that Σf accounts for the reduction in pump absorption due to species photo-degradation. To speed up the experiment, the pump repetition rate was increased up to 10 Hz. The fluorescence emission and laser spectra were monitored perpendicular to the exciting beam, collected by an optical fiber, and imaged onto a spectrometer (USB2000 + from Ocean Optics).

The photoinduced production of singlet oxygen (¹O₂) was determined by direct measurement of the luminescence at 1276 nm with a NIR detector integrated in the aforementioned spectrofluorimeter (InGaAs detector, Hamamatsu G8605–23). The ¹O₂ signal was registered in front configuration (front face), 40° and 50° to the excitation and emission beams, respectively and leaned 30° to the plane formed by the direction of incidence and registration in cells of 1 cm. The signal was filtered by a low cut-off of 850 nm. ¹O₂-generation quantum yield (ϕ_Δ) was determined using the following equation:

$${\varphi }_{\Delta }={{\varphi }_{\Delta }}^{\text{r}}\cdot \left({\alpha }^{\text{r}}/{\alpha }^{\text{Ps}}\right)\cdot \left({\text{Se}}^{\text{Ps}}/{\text{Se}}^{\text{r}}\right)$$

where ϕ_Δ^r is the quantum yield of ¹O₂ generation for the used reference (in our case, phenalenone). Factor α = 1–10^−Abs, corrects the different amount of photons absorbed by the sample (α^Ps) and reference (α^R). Factor Se is the intensity of the ¹O₂ phosphorescence signal of the sample (Se^Ps) and the reference (Se^r) at 1276 nm. Phenalenone in chloroform was used as reference for visible irradiation (420 nm). The singlet-oxygen quantum yield of phenalenone in chloroform used as reference was ϕ_Δ = 0.98 [32]. ¹O₂ quantum yields were averaged from 5 concentrations between 10⁻⁶ M and 10⁻⁵ M in chloroform (spectroscopic grade).

CD spectra were recorded on a Jasco (model J-715) spectro-polarimeter using standard quartz cells of 1-cm optical-path length in chloroform solution, unless otherwise indicated, at a dye concentration of ca. 4·10⁻⁶ M. Circularly polarized luminescence (CPL) and total luminescence spectra were recorded at 295 K in degassed chloroform solution (nitrogen was bubbled into the solution), unless otherwise indicated, at a dye concentration of ca. 1.5·10⁻³ M, on an instrument described previously, [33] operating in a differential photon-counting mode.

2.1.3. Computational simulations

Ground state energy minimizations were performed using range-separated hybrid wb97xd functional, within the Density Functional Theory (DFT), using the triple valence basis set with a polarization function (6–311g*). Such functional is better suited to describe long-range interactions, [34] as those promoted by the chelation of BINOL at the boron bridge of BOPHY. The absorption profile and first singlet excited state was simulated and optimized, respectively, with the Time Dependent (TD-DFT) method using the same calculation level and basis set. The optimized geometries were taken as a true energy minimum using frequency calculations (no negative frequencies). The corresponding cartesian coordinates of all the optimized geometries are collected in Tables S4–S7 in Supporting Information. The Polarizable Continuum Model (PCM) considered solvent effect (cyclohexane) in all the calculations. All the calculations were performed in Gaussian 16, using the “arina” computational resources provided by the UPV-EHU.

2.2. Synthetic procedures and characterization data

BOPHYs 1a, 2a and 3a were prepared following previously reported methods [20,21].

2.2.1. BOPHY 1b

Unless stated otherwise, the reaction was performed in a flame-dried flask. A mixture of BOPHY 1a (25 mg, 0.09 mmol, 1 mol equiv.) and aluminum chloride (59 mg, 0.44 mmol, 5 mol equiv.) in dry CH₂Cl₂ (DCM, 5 mL) was refluxed under argon atmosphere until reaction completion (reaction monitored by TLC). The mixture was cooled down to room temperature and, then, a solution of (R)-BINOL ((R)-1,1′-binapht-2-ol, 101 mg, 0.35 mmol, 4 mol equiv.) in anhydrous acetonitrile (2 mL) was added dropwise. The resulting mixture was stirred at r.t. for additional 6 h. After filtration and solvent evaporation under reduced pressure, the obtained residue was purified by flash chromatography (hexane/DCM 7:3) to afford 1b (24 mg, 35%) as a yellow solid. R_f = 0.27 (hexane/DCM 1:1). Mp: 250 °C (decomp.). [α]_D²⁰ -1666 (c 0.055 CHCl₃). ¹H NMR (300 MHz, CDCl₃)δ 8.01–7.92 (m, 6H), 7.86 (d, J = 7.9 Hz, 2H), 7.81 (d, J = 8.8 Hz, 2H), 7.57 (d, J = 8.8 Hz, 2H), 7.46 (d, J = 8.4 Hz, 4H), 7.43–7.35 (m, 4H), 7.39 (d, J = 8.7 Hz, 2H), 7.33–7.23 (m, 4H), 7.19 (m, 2H), 6.79 (dd, J = 4.1, 1.2 Hz, 2H), 6.34 (dd, J = 4.1, 2.1 Hz, 2H) ppm. ¹³C NMR (75 MHz, CDCl₃) δ 152.8 (C), 152.1 (C), 140.5 (CH), 137.8 (CH), 133.3 (C), 133.1 (C), 130.7 (C), 130.6 (C), 130.3 (CH), 129.8 (CH), 128.5 (CH), 128.2 (CH), 127.4 (CH), 127.3 (CH), 127.0 (CH), 126.0 (CH), 125.8 (CH), 124.3 (CH), 124.2 (C), 124.1 (CH), 123.2 (C), 122.9 (CH), 122.6 (CH), 122.5 (C), 116.6 (CH) ppm. FTIRν 1589, 1436, 1316, 1251, 1222, 1191, 1071, 1035, 982, 895 cm⁻¹. HRMS: 775.2704 (calcd. for C₅₀H₃₃B₂N₄O₄: 775.2688).

2.2.2. BOPHY 1c

Following a similar procedure to that used for BOPHY 1b, 1a (20 mg, 0.07 mmol) was reacted with (R)-3,3′-dibromoBINOL (125 mg, 0.28 mmol). The reaction crude was purified using flash chromatography (hexane/DCM 1:1) to obtain 1c (29 mg, 38%) as a yellow solid. R_f = 0.45 (hexane/DCM 1:1). Mp: 200 °C (decomp.). [α]_D²⁰ -1220.1(c 0.142 CHCl₃). ¹H NMR (CD₂Cl₂, 300 MHz)δ 8.25 (s, 2H), 8.19 (s, 2H), 7.94 (s, 2H), 7.85 (m, 4H), 7.44 (m, 4H), 7.30–7.18 (m, 8H), 6.96 (dd, J = 4.1, 1.2 Hz, 2H), 6.86 (m, 2H), 6.39 (dd, J = 4.1, 2.1 Hz, 2H) ppm. ¹³C NMR (CD₂Cl₂, 75 MHz) 150.1 (C), 149.2 (C), 140.9 (CH), 138.6 (CH), 133.4 (CH), 133.0 (CH), 133.0 (C), 132.7 (C), 131.5 (C), 131.4 (C), 128.6 (CH), 127.9 (CH), 127.3 (CH), 127.3 (CH), 126.8 (CH), 126.7 (CH), 125.7 (CH), 125.7 (CH), 125.5 (C), 124.1 (C), 124.0 (C), 118.1 (C), 117.9 (C), 117.4 (CH) ppm. FTIRν 1594, 1400, 1334, 1252, 1220, 1195, 1034, 985, 897 cm⁻¹. HRMS: 1108.8953 (calcd. For C₅₀H₂₈B₂Br₄N₄O₄Na: 1108.8928).

2.2.3. BOPHY 2b

Following a similar procedure to that used for BOPHY 1b, 2a (23 mg, 0.07 mmol) was reacted with (R)-BINOL (76 mg, 0.26 mmol). The reaction crude was purified using flash chromatography (hexane/DCM 1:1) to obtain 2b (18 mg, 32%) as a yellow solid. R_f = 0.27 (hexane/DCM 1:1). Mp: > 300 °C. [α]_D²⁰ -412 (c 0.03 CHCl₃). ¹H NMR (CDCl_3, 300 MHz) δ 7.93 (d, J = 8.9 Hz, 2H), 7.92 (s, 2H), 7.88 (d, J = 8.1 Hz, 2H), 7.82 (d, J = 8.0 Hz, 2H), 7.70 (d, J = 8.8 Hz, 2H), 7.50 (d, J = 8.7 Hz, 2H), 7.43–7.27 (m, 8H), 7.25–7.15 (m, 4H), 6.87 (d, J = 8.7 Hz, 2H), 5.83 (s, 2H), 1.86 (s, 6H), 1.51 (s, 6H) ppm. ¹³C NMR (CDCl_3, 75 MHz) δ 153.7 (C), 153.0 (C), 151.8 (C), 140.2 (C), 135.6 (CH), 133.7 (C), 133.3 (C), 130.3 (C), 130.3 (C), 129.9 (CH), 129.6 (CH), 128.2 (CH), 128.1 (CH), 127.4 (CH), 126.9 (CH), 125.9 (CH), 125.6 (CH), 124.1 (CH), 123.8 (CH), 123.7 (C), 123.2 (CH), 122.6 (CH), 122.2 (C), 122.1 (C), 118.0 (CH), 14.4 (CH₃), 10.6 (CH₃) ppm. FTIRν 1594, 1514, 1467, 1298, 1250, 1200, 1029, 975, 943 cm⁻¹. HRMS: 831.3329 (calcd. for C₅₄H₄₀B₂N₄O₄: 831.3314).

2.2.4. BOPHY 2c

Following a similar procedure to that used for BOPHY 1b, 2a (20 mg, 0.06 mmol) was reacted with (R)-3,3′-dibromoBINOL (105 mg, 0.24 mmol). The reaction crude was purified using flash chromatography (hexane/CH₂Cl₂ 1:1) to obtain 2c (27 mg, 40%) as a yellow solid. R_f = 0.33 (hexane/DCM 1:1). Mp: > 300 °C. [α]_D²⁰ -1885 (c 0.035 CHCl₃). ¹H NMR (CDCl_3, 300 MHz) δ 8.22 (s, 2H), 8.10 (s, 2H), 7.80 (d, J = 8.2 Hz, 2H), 7.76 (d, J = 8.2 Hz, 2H), 7.53 (s, 2H), 7.43–7.32 (m, 4H), 7.25–7.13 (m, 8H), 5.93 (s, 2H), 1.66 (s, 6H), 1.54 (s, 6H) ppm. ¹³C NMR (CDCl_3, 75 MHz) δ 151.7 (C), 150.1 (C), 149.8 (C), 140.7 (C), 135.0 (CH), 132.7 (C), 132.6 (CH), 132.5 (C), 132.3 (CH), 130.7 (C), 130.6 (C), 127.3 (CH), 127.2 (CH), 127.2 (CH), 126.8 (CH), 126.3 (CH), 126.1 (CH), 125.1 (CH), 124.9 (CH), 124.2 (C), 123.4 (C), 118.9 (CH), 118.4 (C), 118.4 (C), 14.4 (CH₃), 10.5 (CH₃) ppm. FTIRν 1593, 1515, 1250, 1203, 1035, 942 cm⁻¹. HRMS: 1164.9518 (calcd. for C₅₄H₃₆B₂Br₄N₄O₄Na: 1164.9554).

2.2.5. BOPHY 3b

Following a similar procedure to that used for BOPHY 1b, 3a (17 mg, 0.04 mmol) was reacted with (R)-BINOL (50 mg, 0.18 mmol). The reaction crude was purified using flash chromatography (hexane/CH₂Cl₂ 1:1) to obtain 3b (22 mg, 58%) as a yellow solid. R_f = 0.33 (hexane/DCM 1:1). Mp: > 300 °C. [α]_D²⁰ -128 (c 0.03 CHCl₃). ¹H NMR (CD₂Cl_2, 300 MHz) δ 8.07 (s, 2H), 8.02 (d, J = 8.8 Hz, 2H), 7.94 (d, J = 8.1 Hz, 2H), 7.82 (d, J = 8.1 Hz, 2H), 7.64 (d, J = 8.8 Hz, 2H), 7.60 (d, J = 8.8 Hz, 2H), 7.39 (ddd, J = 8.2, 6.6, 1.3 Hz, 2H), 7.36–7.28 (m, 4H), 7.27–7.14 (m, 6H), 6.66 (d, J = 8.7 Hz, 2H), 2.22 and 2.29 (ABX₃ system, AB part, J_AB = 14.5, J_AX = J_BX = 7.5 Hz, 4H), 2.03 (s, 6H), 1.36 (s, 6H), 0.88 (ABX₃ system, X part, J_AX = J_BX = 7.5 Hz, 6H) ppm. ¹³C NMR (CDCl_3, 75 MHz) δ 153.9 (C), 153.2 (C), 150.1 (C), 136.5 (C), 134.6 (CH), 133.7 (C), 133.3 (C), 130.6 (C), 130.3 (C), 130.3 (C), 129.9 (CH), 129.5 (CH), 128.1 (CH), 128.1 (CH), 127.4 (CH), 126.9 (CH), 125.8 (CH), 125.5 (CH), 124.0 (CH), 123.7 (CH), 123.3 (CH), 123.0 (C), 122.8 (CH), 122.2 (C), 17.2 (CH₃), 14.6 (CH₃), 12.2 (CH₃), 8.7 (CH₃) ppm. FTIRν 1594 1480, 1334, 1297, 1271, 1206, 1173, 945 cm⁻¹. HRMS: 887.3968 (calcd. for C₅₈H₄₉B₂N₄O₄: 887.3940).

2.2.6. BOPHY 3c

Following a similar procedure to that used for BOPHY 1b, 3a (20 mg, 0.05 mmol) was reacted with (R)-3,3′-dibromoBINOL (90 mg, 0.20 mmol). The reaction crude was purified using flash chromatography (hexane/CH₂Cl₂ 1:1) to obtain 3c (32 mg, 52%) as a yellow solid. R_f = 0.36 (hexane/DCM 1:1). Mp: > 300 °C. [α]_D²⁰ -1882 (c 0.02 CHCl₃). ¹H NMR (CDCl_3, 300 MHz) δ 8.21 (s, 2H), 8.08 (s, 2H), 7.78 (m, 4H), 7.46 (s, 2H), 7.43–7.31 (m, 4H), 7.25–7.13 (m, 8H), 2.22 (m, 4H), 1.57 (s, 6H), 1.51 (s, 6H), 0.98 (t, J = 7.5 Hz, 6H) ppm. ¹³C NMR (CDCl_3, 75 MHz) δ 150.4 (C), 150.0 (C), 149.9 (C), 136.9 (C), 134.0 (CH), 132.7 (C), 132.53 (C), 132.49 (CH), 132.2 (CH), 131.4 (C), 130.7 (C), 130.5 (C), 127.3 (CH), 127.2 (two CH), 126.8 (CH), 126.2 (CH), 126.0 (CH), 125.0 (CH), 124.8 (CH), 123.5 (C), 123.44 (C), 123.42 (C), 118.61 (C), 118.59 (C), 17.4 (CH₂), 14.5 (CH₃), 12.2 (CH₃), 8.6 (CH₃) ppm. FTIRν 1597, 1450, 1299, 1251, 1206, 1043, 944 cm⁻¹. HRMS: 1199.0405 (calcd. for C₅₈H₄₅B₂Br₄N₄O₄: 1199.0360).

2.2.7. BOPHY 4a

A solution of p-methoxybenzaldehyde (185 mg, 1.36 mmol, 3 mol equiv.) in piperidine (1 mL) was added over a solution of 2a (150 mg, 0.44 mmol, 1 mol equiv.) and p-toluenesulfonic acid (30 mg, 0.17 mmol, 0.4 mol equiv.) in dry toluene (5 mL). The mixture was heated at 140 °C, under Argon atmosphere, until the solvent was fully evaporated. The residue was purified using flash chromatography (hexane/DCM 1:1) to obtain 4a (108 mg, 42%) as a black solid. R_f = 0.18 (hexane/AcOEt). Mp: > 300 °C. ¹H NMR (CDCl_3, 700 MHz) δ 7.94 (s, 2H), 7.52 (d, J = 8.4 Hz, 4H), 7.29 and 7.25 (AB system, J_AB = 16.3 Hz, 4H), 6.92 (d, J = 8.4 Hz, 4H), 6.74 (s, 2H), 3.85 (s, 6H), 2.38 (s, 6H) ppm. ¹³C NMR (CDCl_3, 176 MHz) δ 160.7 (C), 150.4 (C), 140.5 (C), 136.5 (CH), 133.1 (CH), 129.2 (C), 129.0 (CH), 124.8 (C), 115.8 (CH), 114.8 (CH), 114.5 (CH), 55.6 (CH₃), 11.4 (CH₃) ppm. FTIRν 1575, 1501, 1292, 1111, 1087, 939, 814 cm⁻¹. HRMS: 597.2232 (calcd. for C₃₀H₂₈B₂F₄N₄O₂Na: 597.2232).

2.2.8. BOPHY 5a

Following a similar procedure to that used for BOPHY 4a, 3a (150 mg, 0.38 mmol) was reacted with p-methoxybenzaldehyde (156 mg, 1.15 mmol). The residue was purified using flash chromatography (hexane/DCM 1:1) to obtain 3c (108 mg, 45%) as a black solid. R_f = 0.25 (hexane/AcOEt). Mp: > 300 °C. ¹H NMR (CDCl_3, 700 MHz) δ 7.94 (s, 2H), 7.53 (d, J = 8.7 Hz, 4H), 7.31 and 7.29 (AB system J_AB = 16.6 Hz, 4H), 6.93 (d, J = 8.7 Hz, 4H), 3.86 (s, 6H), 2.72 (q, J = 7.6 Hz, 4H), 2.28 (s, 6H), 1.22 (t, J = 7.6 Hz, 6H) ppm. ¹³C NMR (CDCl_3, 176 MHz) δ 160.5 (C), 147.2 (C), 137.6 (C), 136.1 (CH), 132.9 (CH), 132.3 (C), 129.9 (C), 128.8 (CH), 124.2 (C), 116.1 (CH), 114.4 (CH), 55.6 (CH₃), 18.5 (CH₂), 14.3 (CH₃), 9.2 (CH₃) ppm. FTIRν 1576, 1455, 1295, 952 cm⁻¹. HRMS: 653.2855 (calcd. For C₃₄H₃₆B₂F₄N₄O₂Na: 653.2858).

3. Results and discussion

3.1. Alkylated and red-emitting BOPHYs

The chromophoric core of the BOPHY dyes reminds to that of the BODIPYs, but involving two boron-pyrromethene units (see Fig. 1). Thus, the BOPHY core comprises four fused π-conjugated rings, which should result in a planar (pseudo aromatic) structure similar to that of the BODIPY dyes, which involve three fused π-conjugated rings. However, computational geometry optimization reveals that the BOPHY framework is not fully planar (neither in the ground state, nor in the excited state). Thus, each one of the BOPHY boron-pyrromethene moieties stands in a different plane, with a deviation angle of ca. 30° from the expected planarity, giving place to a butterfly-like (C₂-symmetric) structure, which is similar in both the ground and the excited state (Fig. 3). Interestingly, such a lack of planarity does not prevent the extended delocalization of the involved π-system, as supported by the computed contour maps for the BOPHY frontier orbitals: the π-system spreads throughout the whole BOPHY tetracycle, with the expected exception of the boron centers (Fig. 3). Nonetheless, this bent arrangement does lead to a less efficient delocalization of the π-system, as reflected in the spectral band positions of the simplest BOPHY dye (1a in Fig. 1) when compared to the corresponding BODIPY counterpart (see Fig. 1) [35]. Thus, the spectral bands of said BODIPY counterpart are ca. 50 nm red-shifted with respect to those of BOPHY 1a, in spite of the lesser π-extension of the BODIPY core (Fig. S1 in Supporting Information). Accordingly, the absorption probability of BOPHY 1a (evaluated by the molar absorption) is also lower than in its BODIPY counterpart (4–5·10⁴ M⁻¹cm⁻¹ vs.7–9·10⁴ M⁻¹cm⁻¹ [35], respectively; see Table 1). The profile of the spectral bands of BOPHYs is also different to that of related BODIPYs, being the spectral bands of BOPHYs broader and with a marked vibrational resolution (Fig. 4 and Fig. S1 in Supporting Information). In the particular case of the BOPHY absorption, the band features two main peaks with similar intensity, albeit their relative intensity depends on the surrounding environment (e.g., the vibrational resolution is clearer in non-polar media; e.g., see spectra of BOPHY 1a in different solvents in Fig. S2 in Supporting Information) and on the substituents decorating the chromophoric core (e.g., see Fig. 4) [20,21]. Besides, as the alkylation increases (2a and 3a), the spectral bands are shifted bathochromically and the absorption long-wavelength shoulder becomes dominant (Fig. 4). This is due to the electron-donating inductive effect exerted by the alkyl groups increasing mainly the HOMO energy, and hence, reducing the energy gap (as confirmed theoretically in Fig. 3). In contrast, in the fluorescence profile, the high-energy peak always prevails and the band shape is almost independent of the above said factors. Therefore, the absorption and fluorescence spectra are not mirror images, which is especially clear in non-alkylated BOPHY 1a (Fig. 3).

Fig. 3.

Optimized ground- and excited-state geometries, corresponding electronic contour maps and energies of the frontier orbitals (PCM-wb97xd/6–311g*), and theoretically predicted absorption spectra (PCM-TD-wb97xd/6–311g*) for BOPHYs 1a, 2a and 3a.

Table 1.

Photophysical (diluted μM solutions) and laser signatures (concentrated mM solutions) of BOPHYs 1a, 2a and 3a in different solvents.

	λab (nm)	εmax 10−4 (M−1 cm−1)	λfl (nm)	ϕ	τ (ns)	kfl 108 (s−1)	knr108 (s−1)	λla (nm)	%Eff	Edose90% (GJ mol−1)
1a
c-hex	424.0 (444)	4.0 (3.5)	463.0	0.85	2.38	3.57	0.63
AcOEt	417.5 (436)	4.5 (3.8)	460.5	0.94	2.23	4.21	0.27	–	0	–
2a
c-hex	469.0 (444)	4.6 (4.4)	482.5	1.00	2.58	3.87	0.00
AcOEt	441.0 (462)	4.3 (4.2)	480.5	0.95	2.58	3.68	0.19	566	15	1.11
MeOH	440.0 (460)	4.3 (4.2)	479.5	0.88	2.56	3.43	0.46
3a
c-hex	482.0 (454)	6.4 (5.7)	493.5	1.00	2.66	3.76	0.00
AcOEt	474.0 (449)	6.2 (6.0)	490.5	0.90	2.68	3.35	0.37	571	18	4.41
MeOH	473.0 (449)	6.3 (6.0)	489.5	0.96	2.69	3.57	0.15

Maximum absorption (λ_ab) and fluorescence (λ_fl) wavelengths; maximum molar absorption (ε_max); fluorescence quantum yield (ϕ) and lifetime (τ); radiative (k_fl) and non-radiative (k_nr) rate constants; lasing wavelength (λ_la) and efficiency (%Eff) at optimal concentration; photostability as the required energy to reduce the laser output by 10% (E_dose^90%). c-hex: cyclohexane; AcOEt: ethyl acetate; MeOH: methanol. BOPHY 1a was unstable upon irradiation in methanol and the solution bleached quickly.

Fig. 4.

Absorption and fluorescence spectra of BOPHYs 1a, 2a and 3a in cyclohexane, as well as laser spectra of 2a and 3a in ethyl acetate (red-shifted narrow bands; scaled to the corresponding fluorescence spectra).

In spite of the aforementioned theoretically-predicted bended geometry, the fluorescence efficiency of the alkylated BOPHYs (1a-3a) in solution is extremely high, reaching fluorescence quantum yields ca. 100% regardless of the solvent polarity (see 2a and 3a in Table 1) [20,21], thereby slightly improving the fluorescence efficiency of the corresponding BODIPY counterparts (fluorescence quantum yields up to ca. 90% [30,35]). Thus, the two boron bridges involved in the BOPHY structure seem to compensate the deleterious effect produced by its lack of planarity, providing high enough rigidity as to avoid the internal conversion processes associated to the geometrical distortions [31]. Moreover, lifetimes are analyzed as mono-exponentials for the studied BOPHYs (Table 1), being faster than those for related BODIPYs (ca. 2.5 ns vs. 4–6 ns [31,34], respectively) and leading to higher fluorescence deactivation rate constants, too (ca. 3–4·10⁸ s⁻¹ for BOPHYs vs. 1–2·10⁸ s⁻¹ for BODIPYs [30,35]). These results demonstrate that BOPHYs are faintly brighter fluorophores than related BODIPYs.

On the other hand, the chromophoric organic scaffold of the BOPHY dyes is expected to be as chemically versatile as that of the related BODIPY dyes. In fact, BOPHYs should be readily amenable to most of the chemical transformations used in BODIPYs to finely modulate key photophysical signatures. This is the case of the Knoevenagel-like reaction of the acidic methyl groups allocated at the α-pyrrolic positions with aromatic aldehydes, which allows π-extending the chromophore, to easily generate red-shifted dyes in both BODIPYs [36] and BOPHYs [23]. Nonetheless, possible differences have not been reported to date. To this aim, we have synthesized two π-extended BOPHYs (4a and 5a in Fig. 2) by reacting the corresponding methylated BOPHY precursors (2a and 3a) with 4-methoxybenzaldehyde (see Fig. 2 and Experimental section). These extended BOPHYs are expected to exhibit bright fluorescence response, owing to the excellent fluorescence capability of the parent precursors (Table 1), but deeply shifted to the red-edge of the visible electromagnetic spectrum, as it occurs when conducting the same transformation in related BODIPY counterparts [36]. In fact, the spectral bands of π-extended BOPHYs 4a and 5a are well red-shifted, being allocated at the red edge of the visible spectrum (Fig. 5), as predicted also theoretically (Fig. S3 in Supporting Information). However, these red-emitting dyes (emission maxima ca. 610–620 nm) are characterized by lower fluorescence quantum yields (40–50%; Table 2) and faster lifetimes (lower than 2 ns; Table 2) than related BODIPYs bearing two styryls at the α-pyrrolic positions [37]. Thus, the reduction of the energy gap implies a significant increase of the non-radiative rate constant (likely internal conversion owing to the more feasible vibrational coupling between ground and excited state in this spectral region), since the radiative rate constant remains high (cf. precursor 2a and 3a vs. π-extended 4a and 5a in Tables 1 and 2). Such a non-radiative pathway seems to be more feasible in BOPHYs, likely owing to its distorted geometry affording a higher vibrational motion.

Fig. 5.

Normalized absorption and fluorescence spectra of red-emitting π-extended BOPHYs (red solid lines) vs. respective precursors (black dashed lines) in ethyl acetate. A): 4a vs. 2. B): 5a vs. 3a.

Table 2.

Photophysical properties of red-emitting BOPHYs 4a and 5a.

	λab (nm)	εmax 10−4 (M−1 cm−1)	λfl (nm)	ϕ	τ (ns)	kfl 108 (s−1)	knr108 (s−1)
4a
THF	578.5 (546)	8.9 (8.3)	612.0	0.55	1.71	3.21	2.63
AcOEt	573.5 (540)	7.8 (7.2)	605.0	0.52	1.71	3.04	2.80
ACN	568.0 (538)	8.8 (8.4)	610.5	0.41	1.32	3.10	4.47
5a
THF	584.5 (555)	5.2 (5.0)	623.5	0.48	1.67	2.87	3.11
AcOEt	577.0 (548)	6.1 (6.0)	619.5	0.43	1.66	2.59	3.43
ACN	≈ 560	4.0	619.5	0.33	1.28	2.58	5.23

Maximum absorption (λ_ab) and fluorescence (λ_fl) wavelengths; maximum molar absorption (ε_max); fluorescence quantum yield (ϕ) and lifetime (τ); radiative (k_fl) and non-radiative (k_nr) rate constants. THF: tetrahydrofurane; AcOEt: ethyl acetate; ACN: acetonitrile. BOPHYs 4a and 5a were poorly soluble in cyclohexane and unstable in methanol.

Such high probability of fluorescence deactivation and notable fluorescence response prompted us to test the performance of these BOPHYs as active media of dye lasers. It must be noted here that, to the best of our knowledge, the behavior of BOPHYs as laser dyes has been unreported to date. Strikingly, the simplest BOPHY 1a does not render laser signal, in spite of its high fluorescence efficiency (Table 1). In contrast, alkylated BOPHYs 2a and 3a afford laser emission around 565–570 nm (Fig. 3), displaying laser efficiency ca. 20% (Table 1) and good tolerance under prolonged and intense lasing pumping (especially for fully alkylated 3a, where up to 4.41 GJ mol⁻¹ were required to decrease the output energy of the laser emission to 90% of its initial value (E_dose^90%); Table 1). However, red-emitting BOPHYs 4a and 5a provided a weak laser emission (laser efficiency ca. 1% at 650 nm), quickly undergoing photobleaching under the required strong laser pumping.

In order to compare rigorously the lasing behavior of BOPHYs with that of BODIPYs, the laser performance of the well-known lasing alkylated BODIPYs PM546 and PM567 (commercial laser dyes) [30] was tested under the same experimental conditions used in our labs to test the lasing performance of related BOPHYs 2a and 3a (i.e. pumping at the corresponding maximum-absorption wavelength with a wavelength-tunable OPO coupled to the third harmonic of the Nd:YAG laser; see Experimental section). Under such conditions, both BODIPYs display higher laser efficiencies than related BOPHYs 2a and 3a (around 60% for PM546 and PM567; up to 18% for BOPHY 3a), but similar photostabilities in terms of E_dose^90% (around 2.5–3.3 GJ mol⁻¹ for PM546 and PM567; up to 4.4 GJ mol⁻¹ for BOPHY 3a).

In addition to the laser efficiency reduction, the laser emission of the BOPHYs is more red-shifted than that expected from the positions of the corresponding absorption and fluorescence bands (the laser emission in BODIPYs is red-shifted ca. 30 nm with regard to the corresponding fluorescence band [38], whereas such a shift reaches ca. 90 nm in BOPHYs; Fig. 4). The foregoing trends clearly point to the presence of much stronger excited state absorption (ESA) overlapping the emission band in BOPHYs than in BODIPYs, as it is known that ESA reduces the laser efficiency [39] and modifies the shape of the amplifying band [40]. This, in turn, could explain as well the unexpected faint, or even lack of, laser emission in compounds 1a, 4a and 5a, in spite of their good photophysical signatures. For π-extended BOPHYs, similar to the herein tested 4a and 5a, a higher probability of photobleaching has been reported [41].

Therefore, despite BOPHYs showing slightly better fluorescence performance under weak irradiation conditions (where ESA is negligible), BODIPYs outperform BOPHYs under intense irradiation regime (where ESA is noticeable), the former being more robust and efficient dyes.

3.2. BINOLated chiroptical BOPHYs

One of the most successful approaches to endow BODIPY dyes with chiroptical activity (i.e. to chirally perturb the inherently achiral BODIPY chromophore) is by constructing C₂-symmetric BINOL-based spiranic O-BODIPYs [42,43]. This can be easily done by chemically replacing the boron fluorine atoms of the starting BODIPY by the BINOL oxygens by means of a nucleophilic substitution reaction activated with aluminum trichloride. Besides, this chemical transformation allows the modulations of photophysical properties in BODIPYs through the promotion of intramolecular charge transfer (ICT) [44]. In BOPHYs, such a synthetic avenue is feasible, too. Thus, Cheng and col. have grafted two configurationally identical BINOL moieties to a BOPHY scaffold, featuring a molecular D-π-A system with tetraphenylethylene (TPA) as the energy-transfer donor and BOPHY as the acceptor. This system is the first enantiopure bis(BINOL)ated dispiranic O,O′-BOPHY (i.e. each boron of which is linked to the two oxygens of a BINOL moiety), which displays visible CD (absorption dissymmetry factor, g_abs, up to 7.4·10⁻⁴ for the (R,R)-enantiomer in THF solution) and visible CPL (luminescence dissymmetry factor, g_lum, up to 5.4·10⁻⁴ for the (R,R)-enantiomer in THF solution) by exciting the TPA moiety (energy transfer) [25]. These dissymmetry factors are lower, and sign-opposite, than those found for a related BINOL-based BODIPY reported by the same group (g_abs up to 1·10⁻³ and g_lum up to 2·10⁻³ for the (R)-enantiomer in THF solution [45]). Such a lower chiral perturbation for BINOL-based BOPHYs, when compared to related BINOL-based BODIPYs, is unexpected given the higher ratio between perturbing moieties (BINOL) and perturbed chromophore (2/1 for BOPHYs vs. 1/1 for BODIPYs). In this context, we found interesting to study this behavior by analyzing additional BODIPY/BOPHY couples that involve closer structures and, besides, by exciting directly the BOPHY chromophore.

Thus, we decided to synthesize a set of BINOL-based O,O′-BOPHYs (see 1b-3b and 1c-3c in Fig. 2) to be compared, chemically, photophysically and chiroptically, with closely related BODIPYs. Treatment of BOPHYs 1a-3a with commercial (R)-BINOL, under the standard conditions used to BINOLate related BODIPYs [44], straightforwardly yields the corresponding bis(BINOL)ated dispiranic O,O′-BOPHYs 1b-3b (see Fig. 2 and Experimental section). Analogously, the use of commercial (R)-3,3′-dibromoBINOL instead of (R)-BINOL, led to the preparation of unprecedented bis(3,3′-dibromoBINOL)ated BOPHYs 1c-3c (see Fig. 2 and Experimental section). The reached chemical yields were similar, but slightly lower than the obtained when BINOLating (or 3,3′-dibromoBINOLating) related BODIPYs [44], probably due to the involved BOPHY double functionalization.

The bis(BINOL)ation of the simplest BOPHY 1a has scarce impact in the absorption spectral properties, as it was expected, since the linking boron atoms do not take part in the delocalized BOPHY chromophore. However, it dramatically alters the fluorescence signatures, entailing almost complete suppression of the fluorescence response (cf. 1b in Table S1 in Supporting Information vs. 1a in Table 1). Conversely, bis (BINOL)ation of tetralkylated BOPHY 2a slightly ameliorates the fluorescence efficiency in apolar media (cf. 2b in Table S1 in Supporting Information vs. 2a in Table 1). However, as the solvent polarity increases, the fluorescence quantum yield of 2b drops, and its fluorescence decay curve becomes bi-exponential and dominated by short lifetimes (Table S1 in Supporting Information and Fig. 6). In fact, the fluorescence signal switches off almost entirely in highly polar methanol (Fig. 6).

Fig. 6.

Evolution of the fluorescence efficiency (fluorescence quantum yields) with the solvent polarity for tetralkylated bis(BINOL)ated and bis(3,3′-dibromoBINOL)ated O,O′-BOPHYs 2b and 2c, respectively, and hexalkylated bis(BINOL)ated and bis(3,3′-dibromoBINOL)ated O,O′-BOPHYs 3b and 3c, respectively.

Finally, bis(BINOL)ation of hexalkylated BOPHY 3a involves less impact on the fluorescence emission (cf. 3b in Table S1 in Supporting Information vs. 3a in Table 1), as well as a lower dependency on the solvent polarity (Table S1 in Supporting Information and Fig. 6). All these trends pinpoint to the promotion of a non-emissive ICT excited state (involving charge transfer from electron-rich BINOL moiety to electron-poor BODIPY core), being such a state the main non-radiative deactivation channel from the locally excited (LE) state especially in polar media. Thus, the stability of said ICT state is expected to be higher for non-alkylated (more electron-poor) BOPHY 1b, being such a state the low-lying excited state regardless of the solvent and, therefore, suppressing entirely the emission from the LE. Indeed, the ICT emission can be detected in cyclohexane, albeit it is very weak and red-shifted (Fig. 7). Further increasing the solvent polarity favors ICT state population from the LE state, but it also increases charge separation and, therefore, the probability of non-radiative deactivation from the ICT, avoiding the detection of its own emission. On the other hand, as the alkyl substitution increases (2b and 3b), the BOPHY becomes a worse electron acceptor from BINOL upon the excitation, which results in more intense fluorescence from the LE state (Fig. 7). In such alkylated BOPHYs, the quenching effect of the BINOL-to-BOPHY ICT is only evident in polar media, where the polar ICT state is further stabilized, enabling its population from the LE. In other words, the simple alkylation of the BOPHY core, together with selection of the media polarity, allows the modulation of the ICT probability, which triggers the fluorescence response in bis(BINOL)ated O,O′-BOPHYs, as it also occurs in BINOLated O-BODIPY dyes [44].

Fig. 7.

Fluorescence spectra in cyclohexane of bis(BINOL)ated O,O′-BOPHYs 1b, 2b and 3b. Inset: amplified emission of 1b to account for the simultaneous emission from the LE and the ICT states (each contribution deconvoluted as dotted-line spectra).

Another strategy to modulate ICT viability and its impact on the fluorescence signatures could be to manipulate the electron-donating capability of the BINOL unit. In fact, bromination of the BINOL moiety is a suitable approach to reduce its electron-donating capability owing to the inductive electron-withdrawing effect exerted by the halogen atoms. This strategy has been successfully applied in BINOL-based BODIPYs to enhance fluorescence response while keeping the chiral perturbation of the BODIPY chromophore [44]. However, in the BOPHY case, substitution of the BINOL moieties of non-alkylated 1b by 3,3′dibromoBINOLs, to generate 1c, is not enough to vanish the ICT, and very low, albeit slightly higher, fluorescence efficiencies are registered (cf. 1b and 1c in Table S2 in Supporting Information). Once again, a long-wavelength, broad and weak emission tail (ICT emission) is detected for 1c but, in this case, a polar medium like methanol is required to record it. Nevertheless, BINOL bromination combined with BOPHY alkylation (2c and 3c) seems to be a valid approach to avoid ICT population, as shown in Fig. 6. Indeed, the sensitivity of the fluorescence efficiency to the solvent polarity is much lower in these cases (as an example, cf. 2b vs. 2c in Fig. 6). Such a lack of solvatochromic effect in the fluorescence efficiency is a proof of hindered ICT population. However, the values of the fluorescence quantum yields are lower than the expected ones (up to 60%; see Table S2 in Supporting Information and Fig. 6). This could be caused by a higher probability of intersystem crossing (ISC) due to heavy-atom effect (four bromines are involved in the molecular structure of bis(3,3′-dibromoBINOL)ated O,O′-BOPHYs, against two in related O-BODIPYs). In fact, bis(3,3′-BINOL)ated O,O′-BOPHYs 1c-3c are poorer fluorophores than the corresponding parent non-BINOLated BOPHYs 1a-3a, and no laser emission was recorded from them under laser pumping, but just amplified spontaneous emission. The same occurs for bis(BINOL)ated O,O′-BOPHYs 1b-3b. Such lack of lasing behavior is not strange based on the aforementioned photophysical phenomena induced by the involved BINOL-based moieties: fluorescence quenching by ICT and, mainly, ISC in the case of the 3,3′-dibromoBINOL derivatives. In fact, ISC allows triplet population and, therefore, triplet-triplet absorption, which is one of the most common losses in the resonator cavity to record laser action [46].

Regarding the BINOL-induced chiral perturbation, two negative visible dichroic signals could be clearly detected in the CD spectra of BINOL-based O,O′-BOPHYs 1b, 1c, 2c and 3c ((R,R) isomers) recorded in solution (see Experimental section, and Fig. S5 in Supporting Information). This dichroic double signalization matches well with the BOPHY-chromophore visible absorption bands (e.g., see Fig. 8 for 3c), the corresponding g_abs values varying from −0.4·10⁻³ for 3c, to −1.3·10⁻³ for 1b (see Table S1 in Supporting Information). However, such a double dichroic signalization could not be detected for 2b and 3b, likely due to interference with the strong dichroic band of the corresponding BINOL moiety for these compounds (see Fig. S5 in Supporting Information). Interestingly, computation of the CD spectra for these (R,R) BOPHYs by means of TD-DFT predicts, in all the cases, a negative signalization for the long-wavelength electronic transition associated to the BOPHY visible absorption (Fig. S4 in Supporting Information).

Fig. 8.

CD (left, blue) and UV–vis absorption (left, red), CPL (right, blue) and total luminescence (right, red) spectra for (R)-3,3′-dibromoBINOL-based O,O′-BOPHY 3c in chloroform solution. For details, see Experimental section.

Therefore, it is fully proven that the (R)-BINOL moiety induces preferentially the absorption of right-handed visible circularly polarized light by the BOPHY chromophore, as it also occurs in related BODIPYs [47]. However, the presence of two BINOL moieties linked to the BOPHY core, instead of a single one in the case of the BODIPYs, do not give place to a significant increase of the chiral perturbation, and the obtained |g_abs| values (ca. 1·10⁻³) are similar to those obtained for related BINOL-based BODIPYs [47].

Regarding the emission of circularly polarized light, negative CPL spectra were recorded for BINOL-based O,O′-BOPHYs 2b, 2c, 3b and 3c in solution upon visible irradiation; i.e., by unprecedented direct excitation of the BOPHY chromophore (see Fig. 8 for 3b, and Fig. S6 in Supporting Information). The CPL spectra of non-alkylated BOPHYs 1b and 1c could not be registered due to dye decomposition under the required experimental conditions (see Experimental section). The registered CPL spectra demonstrate that the chiral perturbation exerted by the BINOL-based moieties on the BOPHY chromophore is working also at the excited state. The corresponding g_lum values calculated from the recorded CPL spectra range from −0.4·10⁻³ for 2b and 2c, to −1.4·10⁻³ for 3c (see Table S3 in Supporting Information), being similar to those obtained for related BINOL-based BODIPYs, and standing into the typical interval for most CPL-enabling small organic molecules (SOMs) [48–50]. Therefore, it is demonstrated that increasing the ratio between perturbing moieties (BINOL) and perturbed chromophore (2/1 for BOPHYs vs. 1/1 for BODIPYs) does not improve the chiral perturbation. However, contrary to the case of BODIPYs [47], the sign of the recorded visible CPL agrees with that of the visible CD for ICT-enabling BINOL-based 1b-3b (the same sign was obtained for g_abs and for g_lum from a specific enantiomer). The different sign between g_abs and g_lum in BINOL-based O-BODIPYs was explained by different handedness in the emission from the ICT or from the LE states [47]. Therefore, the results obtained from BINOL-based O,O′-BOPHYs 1b-3b seem to indicate that, in the BOPHY case, the preferential handedness of the corresponding ICT and LE circularly emissions is the same.

Therefore, the obtained results show than BOPHY chromophores can be chirally perturbed by enantiopure BINOL-based moieties as efficiently as in the case of the BODIPY chromophores, to develop valuable chiroptical dyes. However, such a perturbation is achieved by involving a higher ratio of perturbing chiral moieties, and it is not possible to reverse the CPL sign by modulating ICT emission.

3.3. BOPHYs as singlet oxygen photosensitizers

Direct bromination of the BOPHY chromophore has been reported as a successful strategy to endow BOPHY chromophores with the capability to photogenerate singlet oxygen [51]. As with BODIPY [14] and other highly fluorescent chromophores [52], it relies on the activation of ISC processes induced by the heavy atom effect. It must be noted that singlet-oxygen photosensitizers can be used in a myriad of applications, being photodynamic therapy (PDT), particularly cancer PDT, the most interesting one [53]. In this sense, the aforementioned possibility of ISC working in 3,3′-dibromoBINOL-based O,O′-BOPHYs envisages their activity as singlet oxygen photosensitizers. The synthetic approach here reported could set the foundations for the establishment of a new design route to develop smarter agents for PDT, displaying additionally a valuable advantage: chemical easiness for introducing the key halogens. To add to the previous, the herein studied 3,3′-dibromoBINOL-based O,O′-BOPHYs 2c and 3c retain a notable fluorescence response, which, in principle, would make it possible to apply them not only in PDT, but also in the valuable previous diagnosis of the pathology to be treated by fluorescence bioimaging (i.e. theranostic application). Encouraged by this possibility, we decided to test the performance of BOPHY 2c and 3c as singlet oxygen photosensitizers. Satisfactorily, they showed a significant capability to photogenerate singlet oxygen (singlet-oxygen quantum yield, ϕ_Δ, ca. 50% and 44% for 2c and 3c in chloroform, respectively; see details in Experimental section), while retaining a notable fluorescence efficiency (fluorescence quantum yield, ϕ, ca. 44% and 48% in chloroform, respectively). This outstanding result places these BOPHYs, to the best of our knowledge, among the best balanced organic dyes regarding fluorescence efficiency and single-oxygen photogeneration capability, which, in turn, qualifies the bis(3,3′-dibromoBINOL)ated dispiranic O,O′-BOPHY structure as a highly interesting design for the development of smarter agents for PDT and fluorescence-PDT theranostic applications.

4. Conclusions

In the last years, BOPHYs have arisen as competitive rivals for the famous BODIPYs as versatile organic dyes. Excellent photophysical signatures and good (photo)stability, as well as chemical versatility to finely modulate key photonic properties towards specific applications characterize both dye families, including the design of multi-functional dyes. BOPHY chromophores display a more intense fluorescence signal than related BODIPY chromophores across the green-red region of the visible electromagnetic spectrum, but the former shows less efficient lasing performance under the required strong irradiation regime. Both BODIPYs and BOPHYs can be endowed with the valuable capability to exhibit visible CD and visible CPL by simply tethering enantiopure BINOL-based moieties to their molecular structures through the involved boron atoms by means of straightforward nucleophilic substitutions. Such a transformation also allows tuning their photophysics by modulating the BINOL-induced ICT process, which can be easily done by properly balancing electronic demand in both the BODIPY/BOPHY and the BINOL moieties. However, one of the main advantages of BOPHYs against BODIPYs is the possibility of activating significant ISC by linking two 3,3′-dibromoBINOL moieties to the chromophoric scaffold. This possibility, joined to the said ICT modulation, allows achieving a perfect balance between efficient singlet-oxygen generation and efficient fluorescence response. This outstanding property makes 3,3′-dibromoBINOLated BOPHYs be valuable candidates for the development of organic theranostic agents with both PDT and fluorescence-bioimaging capabilities. Therefore, judging by the herein reported results, BODIPY still prevails as a multifunctional and modern fluorophore, albeit BOPHY is an excellent scaffold with tunable photonics.