Tandrima Chaudhuri‡
*a,
Neelam Shivranb,
Soumyaditya Mulab,
Animesh Karmakar‡a,
Soumi Chattopadhyaya,
Subrata Chattopadhyayb and
Debasis Bandyopadhyayc
aDepartment of Chemistry, Dr Bhupendranath Dutta Smriti Mahavidyalaya, Burdwan – 713407, India. E-mail: tanchem_bu@yahoo.co.in
bBio-Organic Division, Bhabha Atomic Research Centre, Trombay, Mumbai – 400085, India
cDepartment of Chemistry, Bankura Christian College, Bankura – 722101, India
First published on 15th June 2016
Polycyclic aromatic hydrocarbons (PAHs) demonstrated unusual weak C–H⋯F type H-bonding interaction with meso-substituted Bodipy dyes (1–3) in ethanol medium. Isosbestic absorptions and enhancement of emission of Bodipy were spectroscopically detected with three chosen PAHs (naphthalene, anthracene and phenanthrene). IR, 1H and 19F NMR, cyclic voltammetry, powder X-ray and Monte Carlo simulation of adducts confirmed interaction of PAHs via C–H⋯F type H-bonding with the chosen dyes.
The BF2-dipyrromethene (Bodipy) molecules (Fig. 1) are highly fluorescent dyes and have diverse applications as biolabels,8a artificial light harvesters,8b solar cells sensitizers,8c fluorescent sensors,8d–k molecular photonic wires,8l,m electron-transfer reagents,8n and laser dyes.9a For the past few years, we are actively pursuing research in chemistry and applications of the Bodipys.9a–f Hence, it was of interest to examine the EDA interactions between some Bodipy group of compounds and the PAHs. For this, the meso-aryl Bodipy dyes 1–3 (Fig. 1) and three PAHs viz. naphthalene, anthracene and phenanthrene (designated as D1, D2 and D3 respectively) were chosen,§ and their EDA interactions in ethanol medium studied, using photophysical (absorption and emission) as well as other spectroscopic (IR, 1H and 19F NMR), electrochemical and theoretical studies.¶ Our studies revealed an isosbestic absorbing adduct of all three PAHs (D1–D3) with the Bodipys in ethanol. The interaction cause enhancement of fluorescence quantum yield of Bodipys on interacting with PAHs. Moreover, the IR, NMR, electrochemical, powder X-ray and Monte Carlo simulation studies rationalized the selective recognition on the basis of preferential unusual C–H⋯F type weak H-bonding interaction between the axial C–H proton of PAHs with the F atoms of the Bodipy dyes.10 However, to the best of our knowledge, this is the first report of a H-bonding-mediated detection of PAHs by the Bodipys, which may have far-reaching implications in designing new sensors.
Molecular sensors often rely upon host–guest interaction promoted by hydrogen bonding, electrostatic force, metal–ligand coordination, hydrophobic and van der Waals interaction for species-selective recognition.11
The Bodipy dye 1 is commercially available, while the dyes 2 and 3 were synthesized following our reported procedures.9a,d Absorption spectroscopy studies of D1–D3 (fixed concentration) with increasing concentrations of the dyes revealed formation of well-defined absorption isosbestics in ethanol. Typically, with D2 and dye 2, with incremental addition of 2, the absorbance (at 375.6 nm) of D2 decreased, while that of dye 2 (at 398 nm) increased to reach a maximum (Fig. 2a), giving rise to sets of isosbestics. Thus, all the three PAHs form stable equilibrium with the Bodipy dyes 1–3 in ethanol in the ground states (Fig. S1†). The ground state Jobs plots indicated the formation of dye:
PAH adducts in 1
:
2 mole ratio. A representative plot for dye 3 and D1 is shown in Fig. 2b. The emission spectroscopic study, however, showed steady enhancement of fluorescence intensity of all three dyes with addition of any PAH (Fig. 3 and S2†). This revealed that the adduct formed in the ground state also exists in excited state and showing intensity based sensing for PAHs. The enhancement is around 15% for dye 1 irrespective of PAHs. Excited state limit of detection of D2 in ethanol was 3.55 mM.
The IR finger print region (400–1700 cm−1) (Fig. S3†) of the dye 1/D2 adduct showed peaks, associated with both D2 and dye 1, providing evidence formation of a non-crystalline solid adduct. Also, the B–F asymmetric stretching vibration (1057.88 cm−1)12 of the pure PM567 decreased to 1054.99 cm−1 in the dye 1/D2 adduct (Fig. S3†). This indicated weakening of the B–F bond due to weak H-bonding interaction. Then the 1H NMR studies were carried out to investigate further the interaction of dye 2 with D2 (1:
2) in d4-MeOH. The 1H NMR spectrum of D2 (Fig. S4a†) showed two multiplates at δ 7.457–7.490 and at δ 8.022–8.055 and a singlet at δ 8.469 for the C-9 axial protons. The 1H NMR spectrum of dye 2 (Fig. S4a†) showed well separated peaks viz. doublets at δ 7.105 and 6.961 ppm (aromatic protons), singlets at δ 2.482 and 1.434 ppm (aromatic methyl) and the resonances for the ethyl group at δ 2.364 ppm (q, CH2) and at δ 1.015 ppm (t, CH3). Due to the anthracene ring current effect, the peaks of dye protons shifted upfield in the 1H NMR spectrum of the dye 2/D2 adduct (Fig. S4b†). With regard to the anthracene proton resonances, its axial proton signal at δ 8.469 ppm was slightly upfield shifted to δ 8.467 ppm (Fig. S4c†), without much change in the other peaks. These suggested some weak interaction between the axial proton of anthracene and the dye 2.
Comparison of the 19F NMR spectra (Fig. 4) of the dye 1 and dye 1/D2 adduct gave credence the proposed weak interaction. The F quartets of dye 1 at δ −146.708 ppm were significantly downfield shifted in presence of D2 due to the adduct formation. This also indicated an unusual weak H-bonding interaction between the axial proton of anthracene and the dye F atoms which was also found in IR experiments and discussed before. This was subsequently confirmed by electrochemical and computational study.
The current–potential curves for the dye 1, D1, D2 and (1 + D1) (1 + D2), mixtures in ethanol using supporting electrolyte Bu4NPF6 in Ag/Ag+ ethanol electrode were shown in Fig. 5. The voltammogram showed one irreversible reduction peak at −1.44 V and one reversible oxidation peak at 0.68 V (black trace) within the accessible potential window of the solvent was observed for dye 1 (Fig. 5, black trace). Both D1 and D2 are electron rich systems still exhibited two reduction peaks in cathodic excursion. D2 exhibited two irreversible reductions at −0.22 V and −1.25 V while D1 showed one irreversible (at −0.26 V) and another reversible (at −1.43 V) reduction peaks in its cathodic excursion (Fig. 5, blue and magenta trace). For (dye 1 + D2) mixture reduction peak current increased along with slight shift of reduction potentials towards more negative side compared to free D2, suggested increment of electron charge density on D2 in presence of dye 1. This supported the proton abstraction tendency of dye 1 from D2. Just similar was the observation for D1 and its mixture with dye 1.** The only difference was that all three reductions observed in the current–potential plot of dye 1 + D1 mixture were reversible while for dye 1 + D2 mixture only one reduction peak was reversible, and other two were irreversible or quasi-reversible.13
In order to elucidate the structure of co-crystal of dye 1–D2 adduct, X-ray powder diffraction (XRD) of pure dye 1 and pure D2 were done as well. The XRD results of the D2 and dye 1 exhibited sharp and intense reflections (Fig. 6), indicating the existence of regular structures. It was expected that different arrangements of the individual molecules in the cocrystal would result in different diffraction peaks. However, the XRD spectrum of the dye 1–D2 co-crystal comprised of few peaks, characteristics of both dye 1 and D2 only. This may be due to the highly overlapped Bragg peaks and interference from preferred orientation.14
Nevertheless, the XRD pattern provided clear evidence of co-crystal formation. Moreover, the XRD also exhibited sharp and intense reflections (Fig. 6), indicating existence of regular bonded adducts without any sort of π–π stacking between dye 1 and D2 in the adduct.
The result of the gas phase conformational search of dye 2/D2 adduct gave better interpretation of the 1H NMR as well as 19F NMR spectral findings. The global minimum conformer of the adduct, shown in Fig. 7 clearly reveals that unusually weak C–H⋯F type H-bonding is present in the adduct, and the most acidic C-9 proton of anthracene (D2) participates in the shortest length H-bond formation (Fig. 7a).|| It also showed that the two anthracene moieties in the adduct are inclined at ∼45° with the dye. Hence the dye protons would experience prominent ring current effects of anthracene, resulting in the upfield shift of the Bodipy 1H NMR resonances in the 1/D2 adduct. The FMO features were more interesting. Here, when the HOMO resides on D2, LUMO was on the dye, but when HOMO−2 was on dye then LUMO+2 was on D2 unit. This clearly indicated back donation of electron from electron-deficient dye to electron-rich D2.
![]() | ||
Fig. 7 (a) Wireframe structure of global minimum conformers of dye 2/D2 (1![]() ![]() |
In conclusion, the Bodipy dye chromophore was found to be highly suitable for recognising polyaromatic hydrocarbons in polar medium. The detection of PAHs by the dyes 1–3 is unique to this class of dye, as it involved weak H-bonding type interaction between its F-atoms and the axial acidic proton of PAHs with LOD 3.5 mM. The observed isosbestic dyes/PAH (1:
2) adducts formation was also precisely explained by enhancement of fluorescence quantum yield, IR, NMR, CV, powder X-ray and conformational searching studies of adducts. We have earlier7 shown that electron-deficient (dibenzoylmethanato)boron difluoride having a similar –BF2 unit can efficiently sense electron-rich PAHs, D1–D3 in ethanol medium by simple π–π stacking, without any H-bonding type interaction. Thus, the present observation on the interaction in Bodipy–PAH system is unique, warranting further extension of the work for sensor applications.
Footnotes |
† Electronic supplementary information (ESI) available: IR and NMR. See DOI: 10.1039/c5ra27748e |
‡ Author TC acknowledges the grant received from UGC funded Major project, F.42-390/2013(SR)/dated 25/03/2013, AK is thankful to UGC for fellowship. TC thankful to Dr A. Hazra, Visva Bharati University, WB, India for infrastructural facility. |
§ Materials and methods: HPLC grade ethanol was used for the spectroscopic studies. All the PAHs were from Aldrich and used as such. The concentration of PAHs were in the range (10−7 to 10−6 M) in all the spectral measurements and the dyes (10−6 to 10−5 M). |
¶ The absorption (UV-Vis) spectral measurements were performed with a Shimadzu UV 1800 spectrophotometer fitted with an electronic temperature controller unit (TCC – 240 A). The steady state fluorescence emission and excitation spectra were recorded with a Hitachi F-4500 spectrofluorometer equipped with a temperature controlled cell holder. Temperature was controlled within ±0.1 K by circulating water from a constant temperature bath (Heto Holten, Denmark). The 1H and 19F NMR spectra were recorded with a Bruker 500 spectrometer at 298 K in d4-MeOH. Powder X-ray diffraction measurements were performed on a Bruker D8 diffractometer with Cu Kα radiator. |
|| Computational calculations of conformational searching were performed using Spartan'14 molecular modelling software from Wavefunction Inc. (Irvine, CA, USA). Using Monte Carlo simulation15a in Spartan package, global minima search for all the optimized complexes neglecting solvent were performed using Merck molecular force-field calculations (MMFF).15b FMO calculation using DFT/MPWPW91/6-31G* single point were performed using the same software. |
** 2,6-Diethyl-4,4-difluoro-8-p-hydroxystyryl-1,3,5,7-tetramethyl-4-bora-3a,4a-diaza-s-indecene (3). A mixture of dye 1 (1.0 mmol), p-hydroxybenzaldehyde (1.1 mmol), glacial acetic acid (0.5 mL) and piperidine (3.5 mL) was refluxed in toluene (35 mL) with simultaneous azeotropic removal of water formed during the reaction. After the consumption of 1, H2O (100 mL) was added into the reaction mixture, which was extracted with CHCl3 (3 × 20 mL). The organic layer was dried and concentrated in vacuo to give a residue, which on column chromatography (silica gel, hexane–EtOAc) furnished dye 3. Yield: 0.120 g (28.4%); red solid; 1H NMR: δ 1.05 (t, J = 8 Hz, 6H), 2.19 (s, 6H), 2.41 (q, J = 7.5 Hz, 4H), 2.48 (s, 6H), 6.70 (d, J = 16 Hz, 1H), 6.91 (d, J = 9 Hz, 2H), 7.15 (d, J = 16.5 Hz, 1H), 7.55 (d, J = 8.5 Hz, 2H); 13C NMR: δ 12.4, 14.4, 14.9, 17.5, 116.5, 120.4, 128.8, 129.3, 131.4, 133.0, 138.2, 138.8, 141.1, 152.9; EI-MS (m/z): 422.3 [M]+. Anal. calcd for C25H29BF2N2O: C, 71.10; H, 6.92; N, 6.63%; Found: C, 71.23; H, 6.89; N, 6.78%. |
This journal is © The Royal Society of Chemistry 2016 |