New Insights into an Old Problem. Fluorescence Quenching of

Article pubs.acs.org/JPCA

New Insights into an Old Problem. Fluorescence Quenching of Sterically-Graded Pyrenes by Tertiary Aliphatic Amines Michael J. Bertocchi,† Alankriti Bajpai,‡ Jarugu N. Moorthy,‡ and Richard G. Weiss*,†,§ †

Department of Chemistry and §Institute for Soft Matter Synthesis and Metrology, Georgetown University, Washington, D.C. 20057-1227, United States ‡ Department of Chemistry, Indian Institute of Technology Kanpur, Kanpur 208 016, India S Supporting Information *

ABSTRACT: Although the quenching of singlet-excited states of aromatic molecules by amines has been studied for several decades, important aspects of the mechanism(s) remain nebulous. To address some of the unknowns, steric, and electronic factors associated with the quenching of the singlet-excited states of three electronically related aromatic molecules, pyrene, 1,3,6,8-tetraphenylpyrene (TPPy), and 1,3,6,8-tetrakis(4methoxy-2,6-dimethylphenyl)pyrene (PyOMe), by a wide range of tertiary aliphatic amines have been assessed quantitatively. Correlations among the steric and electronic properties of the amines and the pyrenes (e.g., sizes, shapes, conformational labilities, excitation energies, and oxidation or reduction potentials) have been used in conjunction with the steady-state and dynamic fluorescence quenching data and DFT calculations on the ground and excited state complexes to make quantitative assessments of the steric and electronic factors controlling the quenching processes. PyOMe is a rather rigid bowl-like molecule that, in its electronic ground state, does not make stable complexes with amines in solution. TPPy has a shallower bowl-like shape that is much more flexible. Experiments conducted with a crystalline ground-state complex of an amine and PyOMe demonstrate (as assumed in many other studies but not shown conclusively heretofore) that the geometry needed for quenching the excited singlet state of PyOMe must place the lone-pair of electrons of the amines over the π-system of the pyrenyl group. Furthermore, there is a significant dependence on the shape and size of the amine on its ability to quench PyOMe, but not on the less conformationally constrained TPPy. The conclusions obtained from these studies are clearly applicable to a wide variety of other systems in which fluorescence from an aromatic moiety is being quenched, and they provide insights into how weak host−guest pairs interact.

1. INTRODUCTION

Recently, some tetraarylpyrene derivatives have been employed as hosts to study different aspects of solid-state inclusion complexes with a variety of guest molecules.9−12 For example, 1,3,6,8-tetrakis(4-methoxy-2,6-dimethylphenyl)pyrene (PyOMe, Scheme 1) offers three distinct domains in which a guest molecule may reside (Figure 1).12 In the solid state, the one preferred is influenced by molecular packing considerations as well as the intrinsic shape of PyOMe in which the rings of

The orientation of and distance between the lone-pair of electrons of an amine and the π-system of an aromatic singletexcited state are crucial factors in determining the efficiency of quenching. The most favorable approach and distance of quenchers to the singlet-excited states of aromatic molecules are inferred usually from less than definitive data, although the degree of charge- or electron-transfer during the quenching process is dependent acutely on these factors. In this regard, fluorescence quenching of pyrene by amines has been investigated intensely during the last four decades.1−3 The quenching process can be characterized by the rates of quenching of the pyrene singlet-excited states and the nature and degree of exciplex formation.4−6 Although many aromatic tertiary amines quench pyrene singlet-excited states at or near the diffusion-controlled rate limit, aliphatic tertiary amines generally do at slower rates.7,8 These differences have been explained on the basis of structural rearrangements at the nitrogen center of the amine needed for partial or complete electron transfer that accompanies quenching. The nitrogen centers of aliphatic amines become more planar during the quenching process; the equilibrium geometries about the nitrogen centers of aromatic amines in their electronic ground states are already close to being planar. © XXXX American Chemical Society

Scheme 1. Structures of Tetraarylpyrenes: PyOMe and TPPy

Received: November 14, 2016 Revised: December 20, 2016

A

DOI: 10.1021/acs.jpca.6b11382 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A

Figure 1. Potential locations of guest molecules approaching PyOMe.

Scheme 2. Structures of Tertiary Amines Employed as Quenchers

the four dimethylanisyl moieties are at ca. 90° angles with respect to the plane defined by the pyrenyl basin.12 In the absence of ortho methyl groups (as in 1,3,6,8-tetraphenylpyrene (TPPy)), the calculated structure (vide inf ra) shows that the less sterically encumbered phenyl rings make an approximately 53° angle with respect to the pyrenyl plane and are more easily rotated than those in PyOMe. The increase in planarity results from added stabilization when the π-electron systems of the various aromatic moieties are able to be in conjugation.13 Thus, PyOMe is more discriminating than TPPy in how guest molecules can approach the pyrenyl π-system. From single crystal X-ray diffraction analyses of the structures of PyOMe inclusion compounds, it has been shown that small aliphatic guest molecules reside within the pyrenyl basin of PyOMe and structurally related hosts (i.e., those bearing ortho dimethyl groups on the aryl rings attached to the pyrene core), while larger guests are forced by steric factors to be located within the trough or concave regions (Figure 1).14 These solidstate observations provide important insights into orientational aspects of the mechanism(s) by which solution-phase quenching by tertiary amines of excited singlets of PyOMe (as well as those of TPPy and other pyrenyl molecules) occur. In solution, the approach of a quencher to an aromatic molecule must be inferred on the basis of less definitive data. Here, we examine the dynamics of solution-phase quenching of the singlet-excited states of PyOMe, TPPy, and pyrene by a series of tertiary aliphatic amines in which the sizes of the alkyl groups are varied to provide different overall shapes (Scheme 2). These data are analyzed with respect to the abilities of the lone-pairs of electrons on the nitrogen atoms of the amines to approach especially the basin of PyOMe. Furthermore, we have used a crystalline complex between PyOMe and transperhydroquinoline to demonstrate that the nitrogen lone-pair

must adopt an orthogonal (or near orthogonal) approach to the pyrenyl basin in order for quenching to occur.15 Although the latter has been an integral part of the accepted quenching model for many years, it has never been demonstrated unequivocally to the best of our knowledge.16 Additionally, the optimal distance and angle for approach of the nitrogen lone-pair of electrons to the pyrenyl surface in the ground-state of PyOMe have been examined in detail by DFT calculations. Some correlations between these parameters and the quenching rate constants have been explored. Overall, the experimental data and DFT calculations reported here provide quantitative information concerning the requirements for the efficient quenching of the excited singlet states of pyrene and related molecules by aliphatic amines.

2. EXPERIMENTAL SECTION 2.1. Materials. Tetrahydrofuran (Aldrich, anhydrous, inhibitor free, 99.9%), benzophenone (Aldrich, 99%), ethyl ether (Fisher, anhydrous), chloroform (Fisher, HPLC grade), acetonitrile (Sigma, HPLC grade), ethyl acetate (Sigma, HPLC grade), sodium perchlorate (Sigma, >99%), ferrocene (Acros, 98%), triethylamine (Alfa Aesar, 99%), tripropylamine (Aldrich, 99%), tributylamine (Sigma-Aldrich, >99.5%), dibutylmethylamine (Sigma, >96%), diethylmethylamine (Sigma, 99%), Nmethylpyrrolidine (Sigma, 99%), butyldimethylamine (SigmaAldrich, 98%), sodium (Aldrich, 99%), N-methylpiperidine (Sigma-Aldrich, 99%), N-ethylpiperidine (Sigma-Aldrich, 99%), 1,4-diazabicyclo[2.2.2]octane (Alfa, 98%), trans-decahydronaphthalene (Sigma, 99%), trans-decahydroquinoline (Maxtrix Sci, 97%). Tetrahydrofuran (THF) was freshly distilled under nitrogen from sodium benzophenone before use.17 Tertiary amines and trans-decalin used in quenching studies were distilled by fractional distillation using a low pressure (0.5−10 B

DOI: 10.1021/acs.jpca.6b11382 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A

compartment containing the analyte. The reference compartment was filled with 0.1 M NaClO4 in acetonitrile and was changed when each amine was examined to minimize water contamination in the sample electrode compartment. All potentials were calibrated against ferrocene as an internal standard taking E°(Fc/Fc+) + 0.424 V vs SCE.21 Measurements were obtained with 1−8 mM solutions of tertiary amines in HPLC-grade acetonitrile (dried over 3 Å molecular sieves) and containing 0.1 M NaClO4. The range chosen to encompass redox peak(s) of interest were 0−2 V for amines and 0 to −2.3 V for tetraarylpyrenes. Scans were recorded four times per sample at a scan rate of 200 mV/s. Due to the irreversibility of the redox processes of the compounds of interest, the first scan often differed from subsequent scans. 2.4. Quantum-Chemical Calculations. Density functional theory (DFT) at the M06/6-31G(d,p) level was used to calculate optimal geometries of PyOMe and TPPy, as well as approaches of amines to PyOMe. Geometries were considered optimized in the absence of imaginary frequencies. All calculations were carried out using the Gaussian 09 package in the “gas phase at 0 K”.22 Structures were visualized with Chemcraft.23 Potential energy surfaces (PES) were constructed starting with the optimized structure and then changing the dihedral angle of a single aryl-pyrenyl bond by 10° increments while reoptimizing the remainder of the structure by DFT methods with the M06/6-31G(d,p) program. The angles of the nitrogen lone-pairs with respect to the planes defined by the carbon atoms of the pyrenyl moiety were calculated by starting with amines to which a false H atom had been added to the nitrogen lone-pair and assuming that the orientation of the lone-pair and the N−H bond are the same. An angle of 0° was defined to be orthogonal to the pyrenyl surface. 2.5. Sample Preparation Procedures. A stock solution of a tetraarylpyrene (3 × 10−5 M PyOMe or 3 × 10−4 M TPPy) in THF was prepared and diluted by adding aliquots of separate solutions of tertiary amines in THF. The solutions were flamesealed in flattened glass capillaries (Vitro Dynamics) after being degassed until pressure remained constant by freeze−pump− thaw cycles (5−7 cycles) at 0.1 Torr on a mercury-free vacuum line. Crystals of inclusion compounds were prepared according to the published procedure.12 PyOMe and a guest, in a 3.3:1 w/w ratio, were dissolved in HPLC grade chloroform, and the mixture was allowed to evaporate over 3 d in darkness. White rhombic crystals were isolated. The crystals were then ground into a fine powder and placed into flattened glass capillaries that were closed with septa. The contents of the capillaries were purged with a stream of nitrogen gas for 30 min before measurements were made.

Torr) or house vacuum (250 Torr) prior to use, and the purity was ≥99% as indicated by gas chromatography. 1,4Diazabicyclo[2.2.2]octane was recrystallized twice from ethanol (mp 157−158 °C; Lit:18 155−157 °C). trans-Decahydroquinoline hydrochloride salt was made in and precipitated from diethyl ether. After recrystallization from ether, its melting point was found to be 223−224 °C (Lit:19 222−223 °C). The neutralized amine was obtained from the salt in a 3 M NaOH water/diethyl ether mixture and distilled. 1,3,6,8-Tetrakis(2,6dimethyl-4-methoxyphenyl)pyrene (PyOMe) and 1,3,6,8-tetraphenylpyrene (TPPy) were synthesized following previously reported procedures.20 Both tetraarylpyrenes were >99% pure according to HPLC analyses. 2.2. Instrumentation. 1H and 13C NMR spectra were obtained on a JEOL (400 MHz) spectrometer in CDCl3. Chemical shifts are reported relative to tetramethylsilane (TMS). Thermogravimetric analysis (TGA) (50−700 °C) and differential scanning calorimetry (DSC) (25−300 °C) data were obtained on a PerkinElmer DSC-7 instrument at a heating rate of 10 °C/min under a nitrogen atmosphere. UV/vis absorption spectra were recorded on a Varian UV−vis (Cary 300 Bio) spectrophotometer. Steady-state excitation and emission spectra were recorded on a Photon Technology International Fluorometer (SYS 2459). A 4 cm (width) × 7 cm (length) × 0.3 cm (i.d.) rectangular Pyrex capillary (Vitro Dynamics) was placed in a metallic cuvette holder in the sample container, which positioned the capillary, front-face, at angle of ∼45° with respect to the incident beam. Data were collected at approximately 23 °C unless otherwise indicated. Fluorescence decay histograms were obtained on an Edinburgh Analytical Instruments time-correlated single photon counter (model FL900) with H2 as the lamp gas. An instrument response function (IRF) was collected using a Ludox solution as the scatterer. Data were collected in 1023 channels. Deconvolution was performed by global analysis that minimizes χ2 using the FAST program supplied by Edinburgh. GC−MS analyses of the tertiary amines were performed on a Varian Saturn 2100 instrument using an electron ionization detector and a Zebron DB-5 (30 m × 0.25 mm) column. The heating rate was 15 °C/min starting from 50 °C and ending at 325 °C. The purities of the tetraarylpyrenes were determined by reverse-phase high-performance liquid chromatography (HPLC) using an Agilent Technologies (1220 LC) liquid chromatograph with a HP Zorbax Eclipse XDB-C8 column (150 mm × 4.6 mm, 5 μm particle size) and ethyl acetate as the eluent. Melting points were recorded on a Leitz 585 SM-LUXPOL microscope equipped with crossed polars by heating at a rate of 1 °C/min with a Leitz 350 hot stage. Temperature was measured by an Omega HH503 microprocessor thermometer connected to a J-K-T thermocouple. 2.3. Electrochemical Procedures. Cyclic voltammograms were recorded on a Radiometer 80 analyzer controlled by Voltamaster 4 software. The working electrode consisted of a BAS Pt disc electrode (1.6 mm diameter; MF-2013). The reference electrode was an Ag/AgCl electrode, 3 M NaCl fill/ soaking solution (BAS RE-5B; MF-2013). A platinum flag served as a counter electrode (approximate area: 3 cm2; fabricated in-house). Electrodes were polished between sample runs using a BAS polishing pad and 0.3 μm alumina, which was wetted with 18.3 MΩ cm water and then rinsed with the supporting electrolyte solution. An H-Type electrochemical cell was used with a glass frit between compartments to isolate the reference electrode from the working and counter electrode

3. RESULTS AND DISCUSSION 3.1. Rigidity of Tetraarylpyrene Cavities. Based on DFT calculations, the electron density of the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) of PyOMe, TPPy, and pyrene are distributed over the central pyrene core (Figure 2). In order to affect quenching of the singlet-excited state of PyOMe, TPPy, or pyrene a specific mutual orientation between the pyrenyl πsystem and the lone-pair of electrons on the amine quencher must be achieved.24,25 The calculations indicate that steric interactions imposed by the alkyl groups on the tertiary amine and the four dimethylanisyl rings of PyOMe affect the distance and angle of approach of the nitrogen lone-pair. C

DOI: 10.1021/acs.jpca.6b11382 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A

by 10° increments starting from the optimized ground-state structure shown in Figure 3, while all other bond parameters were allowed to relax. At each increment, the energy was calculated and the resultant potential energy surfaces are shown in Figure 4. Rotation of a PyOMe dimethylanisyl ring from the optimized ca. 88° to 70° angle caused a negligible increase in energy (∼0.6 kcal mol−1); rotation from 70° to 60° resulted in an energy increase of ca. 1.5 kcal mol−1. Figure 4 shows that the torsional freedom of the dimethylanisyl rings is limited and explains why the torsional angles in the crystal structure, 88.7°, 84.9°, 80.1°, and 71.7°,12 correspond to a narrow range of energy increases. By contrast, rotation of a TPPy phenyl ring within a 90−30° range of angles resulted in minimal (