Experimental and theoretical investigation of fluorescence solvatochromism of dialkoxyphenyl-pyrene molecules

Fengniu Lu a, Naoki Kitamura b, Tomohisa Takaya c, Koichi Iwata c, Takashi Nakanishi *a and Yuki Kurashige *b
aFrontier Molecules Group, International Center for Materials Nanoarchitectonics (WPI-MANA), National Institute for Materials Science (NIMS), 1-1 Namiki, Tsukuba 305-0044, Japan. E-mail: nakanishi.takashi@nims.go.jp
bDepartment of Chemistry, Graduate School of Science, Kyoto University, Kitashirakawa Oiwake-cho, Sakyo-ku, Kyoto 606-8502, Japan. E-mail: kura@kuchem.kyoto-u.ac.jp
cDepartment of Chemistry, Faculty of Science, Gakushuin University, 1-5-1 Mejiro, Toshima-ku, Tokyo 171-8588, Japan

Received 6th October 2017 , Accepted 1st November 2017

First published on 1st November 2017

We investigated the fluorescence properties of dialkoxyphenyl-pyrene molecules experimentally as well as theoretically. Our experiments confirmed fluorescence solvatochromism in 2,5-dimethoxyphenyl-pyrene and, in contrast there was no significant solvent-effect on the emission properties of the isomers, 3,5- and 2,6-dimethoxyphenyl-pyrene. This clear difference in the solvent-dependence would reflect the difference in character of the excited-state between the isomers, which differ only in the substitution positions of the two methoxy groups. The positional effects of the di-substituted molecules are successfully explained theoretically by the topologies of the highest occupied molecular orbital of the phenyl group that are governed by the relative positions of the two substituents, though it is somewhat contradictory to the meta-effect for the mono-substituted molecules. Theoretical calculations were also used to analyze the character of the excited states; 2,5-dimethoxyphenyl-pyrene alone exhibited an intramolecular charge transfer character for the excited state, which was responsible for the solvatochromism effect. The dynamics of the excited states were analyzed using time-resolved fluorescence measurements, in which a characteristic increase of the fluorescence intensity was observed for 2,5-dialkoxyphenyl-pyrene; this observation was supported by the theoretical calculations as well.


Fluorescence (FL) properties of chromophores are very sensitive to their surrounding environmental conditions, such as pH,1,2 viscosity,3 and polarity.4 Of these, the polarity-induced change in the photophysical properties, often denoted as FL solvatochromism,5 is widely utilized to quantitatively characterize the properties of solvents or to identify various polarity-dependent chemical or biological events.6–8 Solvatochromic molecules generally undergo a charge transfer (CT) owing to intramolecular donor–acceptor interplay upon excitation. The CT process results in changes in the electronic structures, which are very sensitive to solvent polarity.6 Therefore, solvent switching induces changes in the emission intensity, colour, and lifetime. Predicting the FL solvatochromic feature of a chromophore from the molecular structure is of great significance in advancing the novel design and full use of functional dyes.

Biaryl chromophores are an important family of FL solvatochromic dyes; they contain two aryl chromophores linked by a single bond.9 Biaryl chromophores have attracted increasing attention due to their facile synthesis and the multifunctional nature of the aryl ring. However, the structural versatility of such chromophores makes the FL solvatochromic feature unpredictable. Unlike conventional ‘electron donor–π–acceptor’ type dyes whose CT nature can be easily anticipated from their molecular structure, FL solvatochromism of biaryl chromophores occurs only when the donor/acceptor substituents reside at specific positions of the aryl ring. In spite of the numerous reports on biaryl chromophores,10–19 a clear correlation between the molecular structure and FL solvatochromic behaviour is still missing. For instance, it was claimed that only a meta-substituent on an anthracenyl ring can stabilize the excited state-CT of a 9-phenyl-anthraene derivative.16 On the contrary, both meta- and para-substituents on a pyrenyl ring could cause bathochromic shift of the FL spectrum of 1-phenyl-pyrene from nonpolar to polar solvents.12,19 Therefore, clarifying the effect of substitution on the FL solvatochromism of biaryl derivatives is vital to understand their solvatochromic behaviour and expand their practical applications.

In this study, we investigated the molecular parameters influencing FL solvatochromism and demonstrated a reliable methodology to clarify the solvatochromic effect of biaryl derivatives using a simple 1-phenyl-pyrene model. A series of 1-phenyl-pyrene isomers, 1–3 (Scheme 1), was prepared. All the compounds were substituted with two methoxy groups (1a,102a,20 and 3), or two 2-hexyldecyl chains (1b20 and 2b20) on the phenyl ring, but the substitution positions were either (2,5)-, (3,5)-, or (2,6)-motifs. Photophysical studies showed that only 1a of the (2,5)-motif exhibited a solvatochromic effect, which was retained even after replacing the methoxy groups with bulky alkyl chains (1b). Picosecond time-resolved fluorescence (TRFL) spectra of 1 after excitation allowed us to capture the effects of solvent on the time evolution of the fluorescence peak position and intensity. We then employed the algebraic diagrammatic construction (ADC) scheme to perform electronic-structure calculations for the excited molecules in non-polar and polar solvents. The exclusive solvatochromic feature of 1 could be well understood by the unique topology of its molecular orbitals (MOs), based on which the excited-state CT character was found only in the (2,5)-motif. The calculation results and experimental data are in excellent agreement with each other.

image file: c7cp06811e-s1.tif
Scheme 1 Chemical structures of the dialkoxyphenyl-pyrene derivatives, 1–3.

Results and discussion

Synthesis and photophysical properties

The derivatives 1b, 2a, and 2b were prepared according to previously reported methods.201a and 3 were synthesized via the Suzuki coupling reaction from 1-bromopyrene and 2,5-dimethoxyphenylboronic acid21 or 2,6-dimethoxyphenylboronic acid22 (see the ESI, for details). Both the compounds were characterized by 1H nuclear magnetic resonance (1H NMR), 13C NMR, and matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) (Fig. S1–S6, ESI).

The photophysical properties of 1–3 were studied in solvents of different dielectric constants (1.89–46.7) to analyze the effect of solvent polarity on their optical features. Clear colourless solutions of 1–3 were obtained when they were dissolved at 10 μM concentration in common organic solvents, such as n-hexane, toluene, chloroform (CHCl3), tetrahydrofuran (THF), dichloromethane (CH2Cl2), methanol (CH3OH), acetonitrile (CH3CN), and dimethyl sulphoxide (DMSO) (see selected examples in Fig. 1a). No aggregates could be detected by dynamic light scattering in any of the solutions.

image file: c7cp06811e-f1.tif
Fig. 1 (a) Photographs of 1a, 2a, and 3 in solvents of different polarities (as denoted inside the figure) in daylight (top row) and under a UV lamp of 365 nm irradiation wavelength (bottom row) at 298 K. UV-vis absorption (left) and normalized fluorescence (right) spectra of (b) 1a, (c) 2a, and (d) 3 recorded in various solvents (UV-vis: 10 μM; FL: 1 μM) at 298 K. The solvents (dielectric constant) used were n-hexane (1.89), toluene (2.38), CHCl3 (4.81), THF (7.58), CH2Cl2 (8.93), CH3OH (32.7), CH3CN (37.5), and DMSO (46.7).

The ultraviolet-visible (UV-vis) absorption spectra of all the compounds exhibit features resembling those of the parent, 1-phenyl-pyrene23 (Fig. 1b–d and Fig. S7, ESI). The absorption bands became more vibronic and sharper from Fig. 1c (2a) to Fig. 1b (1a) to Fig. 1d (3). These spectral changes can be attributed to an increase in the steric hindrance upon increasing the number of ortho-ether groups when the substitute position is changed from (3,5)- to (2,5)- to (2,6)-. No obvious spectral shifts were observed in any of the compounds when the polarity of the solvent changed.

In contrast to the similar absorption characteristics of 1a in different solvents, the colours emitted by the 1a solutions exposed to UV irradiation (365 nm) in the dark changed from deep blue to light blue and to greenish blue in solvents of greater polarity (Fig. 1a). Therefore, the steady-state FL spectra of 1a were significantly influenced when the solvents were switched. As shown in Fig. 1b, different emission spectral profiles were observed for the 1a solutions when they were excited at the respective longest absorption maximum (λmax) in each solvent. The FL spectrum of 1a in n-hexane exhibits a sharp and vibrionic band (350–500 nm) with three peaks at 381, 387, and 397 nm. However, the emission band becomes more and more structure-less and broader on increasing the polarity of the solvent. For example, the spectrum in DMSO spans from 365 to 700 nm and contains a single peak at 468 nm. Meanwhile, the emission maximum undergoes a remarkable red-shift. As summarized in Table 1, the maximum FL peaks of 1a were found at 381 nm in n-hexane, 399 nm in toluene, 404 nm in CHCl3, 415 nm in THF, 422 nm in CH2Cl2, 435 nm in CH3OH, 459 nm in CH3CN, and 468 nm in DMSO. The Stokes shift was only 40 nm in the n-hexane solution but as large as 123 nm in the DMSO solution. These different emission features led to different emission colours.

Table 1 Photophysical parameters (λ: wavelength and ΦFL: absolute fluorescence quantum yield) and fitting parameters (τ: decay time and χ2: chi-squared value) of the fluorescence decays of 1a in different solvents
Solvent Absorption featurea Fluorescenceb Fluorescence decayb
λ abs, nm (ε, 104 dm3 mol−1 cm−1) λ max, nm (ΦFL) τ , ns χ 2
a Concentration: 10 μM. b Concentration: 1 μM. λex: 341 nm for n-hexane, CH3OH and CH3CN; 345 nm for toluene, CHCl3 and DMSO; 343 nm for THF and 344 nm for CH2Cl2. c Monitored at their respective maximum emission wavelengths; λex: 279 nm.
n-Hexane 243 (4.51)

276 (3.21)

341 (2.64)



7.4 1.15
Toluene 278 (0.30)

330 (1.95)

345 (2.70)



10.0 1.09
CHCl3 245 (3.88)

279 (2.91)

345 (2.50)



7.9 1.05
THF 243 (5.01)

277 (3.44)

343 (2.70)



6.5 1.07
CH2Cl2 244 (4.57)

278 (3.22)

344 (2.59)



6.0 1.07
CH3OH 242 (4.79)

276 (3.27)

341 (2.71)



4.5 1.02
CH3CN 242 (4.69)

276 (3.21)

341 (2.60)



2.8 1.04
DMSO 279 (3.28)

345 (2.62)



3.2 1.03

To understand the origin of the differences in the emission features, FL decays were investigated. As summarized in Table 1, all the solutions exhibit single-exponential decays in the nanosecond time scale when monitored at their respective maximum emission wavelengths. The results indicate that all the solutions contain a single emission species at least after a few nanoseconds from the photoexcitation. Notably, the decays in all the solutions are in the range of a few nanoseconds. The similarities in the time scales indicate the monomeric character of the fluorescence and exclude the formation of pyrene-excimers in highly polar solvents, such as CH3CN and DMSO, though their broad and red-shifted emission bands are analogous to those of a pyrene-excimer.24 In agreement with this hypothesis, the excitation spectra of 1a, monitored at the monomer emission wavelength (381 nm) and the respective emission maximum of each solvent, possess almost identical spectral features (Fig. S8, ESI).

The derivative 1b, produced by changing the methyl group of 1a to bulky 2-hexyldecyl chains, also shows a fluorescence solvatochromic effect. Similar to those of 1a, the steady-state FL spectra of 1b exhibit a substantial dependence on the solvent. By increasing the polarity of the solvent stepwise, the emission band gradually broadens, loses the vibronic structures, and shows red-shifts (Fig. S7a and Table S1, ESI). Therefore, the substitution of bulky alkyl chains does not influence the FL solvatochromic behaviour.

In contrast, the FL spectra of 2a exhibit a sharp and vibronic band (350–500 nm) with two peaks at 380 nm and 399 nm in all the tested solvents (Fig. 1c and Table S2, ESI). In addition, single-exponential fluorescence decays were observed in all the solutions. The polarity of the solvent did not exert any influence on the FL spectral shape and peak positions. As shown in Fig. 1a, the solutions of 2a exhibit a deep-blue emission colour in different solvents. Compound 2b, substituted with bulky alkyl chains, shows an optical behaviour similar to that of 2a. The UV-vis spectra and FL spectra of 2b are analogous in all the tested solvents (Fig. S7b and Table S3, ESI).

Furthermore, the FL colour of 3 is not affected by the solvent (Fig. 1a). In the FL spectra, a single emission band (350–500 nm) with vibronic peaks at 377, 388, and 397 nm was observed in all the tested solvents (Fig. 1d) along with single-exponential decay (Table S4, ESI). Therefore, the solvent polarity exerts no influence on the FL spectral shape and peak positions of 2 and 3.

Excited-state dynamics

To gain a deeper insight into the FL solvatochromism of the (2,5)-dialkoxyphenyl-pyrene compounds, picosecond TRFL spectroscopy was conducted on 1b and 2b in four solvents, namely n-hexane, CH2Cl2, ethanol (C2H5OH), and 1-butanol (1-C4H9OH). The obtained TRFL spectra are shown in Fig. 2 and Fig. S9 (ESI) for 1b and 2b, respectively. The results corresponding to 1b show a clear dependence of the excited-state dynamics on the solvent, whereas those for 2b do not show a substantial solvent dependence. A slow increase in the FL intensity is observed for 1b in CH2Cl2, C2H5OH, and 1-C4H9OH in the 0–2 ns region, whereas no spectral change is observed in the n-hexane solution of 1b in the 0–720 ps region. The FL band of 1b is substantially down-shifted with time in C2H5OH and 1-C4H9OH, whereas the shift is negligible in n-hexane and CH2Cl2. The peak position after the downshift depends significantly on the solvent. Such solvent dependence is consistent with the solvatochromism observed in steady-state FL measurements (Table S1 and Fig. S7, ESI).
image file: c7cp06811e-f2.tif
Fig. 2 Picosecond TRFL spectra of 1b in (a) n-hexane, (b) CH2Cl2, (c) C2H5OH, and (d) 1-C4H9OH with photoexcitation at 345 nm. * denotes Raman scattering signals from the solvents, which are confirmed by the measurements of the blank solutions without the solute.

The time dependence of the fluorescence area intensity and peak positions is plotted against the time delay and fitted with exponential functions for estimating the time constants of the spectral changes. The results are shown in Table 2 and Fig. 3. Both the area intensity and peak positions change with smaller time constants in the order of CH2Cl2, C2H5OH, and 1-C4H9OH. These time constants are positively correlated with solvent viscosity, as indicated in Table 2. The peak position of the 0 ps spectrum, λ0, and the position when the downshift completes, λ, also depend significantly on the solvent (Table 2). A CH2Cl2 solution exhibits similar values of λ0 and λ, whereas the C2H5OH and 1-C4H9OH solutions exhibit substantial differences. The magnitude of the downshift, Δλ, increases in the order of CH2Cl2, C2H5OH, and 1-C4H9OH, which is positively correlated with the solvent viscosity. The differences in the Δλ values suggest that the spectral changes proceed faster as the viscosity decreases and are almost complete within the instrument response time of our spectrometer in the case of the CH2Cl2 solution.

image file: c7cp06811e-f3.tif
Fig. 3 Time dependence of the (a) area intensity and (b) peak position of the fluorescence band of 1b in CH2Cl2, C2H5OH, and 1-C4H9OH with photoexcitation at 345 nm. The solid traces represent the best-fitted curves obtained by the least-squares fitting analysis with exponential functions.
Table 2 Parameters obtained by the analysis of time dependence of the fluorescence band of 1b and the solvent properties
Solvent T rise, ns T shift, ns λ 0, nm λ , nm Δλ, nm ε η , mPa s
a Ref. 25 b At 298 K. c Most probably completes within 40 ps. d Estimated at 40 ps.
CH2Cl2 0.08 (<0.04)c 448d 449 1 8.93 0.413
C2H5OH 0.18 0.09 445 457 12 24.9 1.074
1-C4H9OH 0.25 0.19 430 448 18 17.4 2.54

Theoretical analysis of excited state properties

To understand the underlying mechanism behind the solvatochromism observed only for 1a (among 1a, 2a, and 3), we investigated the electronic properties of the excited states responsible for the emission using quantum chemical calculations. Geometry optimizations of the ground and excited states in the solution were performed using DFT and TD-DFT at the CAM-B3LYP/6-31+G*26,27 level of theory with the Gaussian1628 program package. The effect of the solvent was modelled by the polarizable continuum model (PCM).29 There are two characteristic low-lying excited states in 1a, 2a, and 3. One is a local π–π* excited (LE) state on the pyrene group and the other is an intramolecular charge transfer (ICT) state from the phenyl group to the pyrene group. Because the energy levels of the two states can be altered depending on the structure and the environment, their structures were individually optimized by energy minimization.

Table 3 shows the dihedral angle θ around the C–C bond between the pyrene and phenyl groups of the optimized structures for the LE and ICT states. The dihedral angles decreased, i.e. the structures became more planar, in the order of 3 > 1a > 2a, regardless of the solvent and character of the excited states. This observation can be explained by the steric hindrance of the methoxy groups on the phenyl group. The effect of solvents on the optimized structural parameters seems negligible; the optimized dihedral angles differ only by 1° to 3° in both the excited states. It is interesting that the trend of the optimized dihedral angles between the ICT and LE states is opposite in the cases of 1a and 2a; the ICT state exhibits a large θ with the LE state in the case of 1a while its value is small in the case of 2a.

Table 3 Dihedral angle θ [°] between the planes of the pyrene and phenyl groups of the optimized structures for LE and ICT states
1a 2a 3
LE state n-Hexane 48.8 37.2 51.9
CH2Cl2 48.9 36.7 52.9
ICT state n-Hexane 52.2 32.6 51.9
CH2Cl2 52.3 32.2 54.7
S0 state n-Hexane 68.6 59.9 90.1
CH2Cl2 69.1 59.8 88.9

As observed in our experiments, the emission wavelength of 1a was gradually red-shifted with an increase in the solvent polarity. To calculate the emission wavelength with high accuracy, ADC(2)30,31 calculations were performed with def2-SVPD basis sets32 and the solvent effect was modelled using the conductor-like screening model (COSMO)33 with the Turbomole34,35 software package. Excited-state calculations were performed at the energy minimum structures of each excited state by TD-DFT calculations; henceforth, we denote each structure as min (‘target state’), e.g. min(ICT) for the energy minimum structure of the ICT state. The reaction fields of the solvents were equilibrated for the lowest excited states in the COSMO model.

Table 4 shows the relative energies of the lowest excited state of the min(ICT) structures and the min(LE) structures. The min(LE) structures are more stable than the min(ICT) structures in almost all cases except for 1a in CH2Cl2, for which the relative energy is negative, i.e. the min(ICT) structure is more stable than the min(LE) structure. The more stable of the min(LE) or min(ICT) structures were assumed to be the energy minimum structures of the lowest excited states and are responsible for the emissions observed in the experiments.

Table 4 Relative energies [kcal mol−1] of the lowest excited state of the min(ICT) structures and the min(LE) structures
1a 2a 3
n-Hexane 1.2 6.5 5.7
CH2Cl2 −3.6 6.5 2.9

The calculated emission wavelengths and dipole moments of the excited states are shown in Table 5. In the case of 2a and 3, the emission wavelengths are in the range of 374–381 nm, which are in good agreement with the experimental observations of 377–387 nm; the shifts caused by the solvents n-hexane and CH2Cl2 are negligible. In the case of 1a, while the emission wavelength in hexane is 376 nm (3.29 eV), which is not very different from that of 2a and 3, the emission wavelength in CH2Cl2 is 452 nm (2.74 eV), which is red-shifted by 76 nm (0.55 eV) from that in n-hexane. The fact that solvatochromism is observed only in 1a but not in 2a and 3 is consistent with the experimental observations.

Table 5 Fluorescence wavelengths [nm] calculated by ADC(2) with the COSMO solvent model. The dipole moments [debye] are shown between parentheses
1a 2a 3
n-Hexane 376 (5.1) 376 (2.8) 374 (3.5)
CH2Cl2 452 (24.1) 381 (2.8) 378 (5.0)

To elucidate the origin of the FL solvatochromism observed in 1a and the reason behind its absence in other molecules, the electronic structures of the lowest excited states were further analyzed. Fig. 4 shows the natural transition orbitals36 of the lowest excited states. Apparently, the character of the lowest excited states of 1a is different from those of 2a and 3. While the characters of the lowest excited states of 2a and 3 are characterized by the transitions from pyrene-HOMO to pyrene-LUMO, i.e. the LE states of the pyrene groups, 1a is characterized by transitions from phenyl-HOMO to pyrene-LUMO, i.e. the ICT from the phenyl group to the pyrene group; in fact, the former exhibits a moderate dipole moment while the latter exhibits a large dipole moment. In general, solvatochromism is observed when the excited states have large dipole moments because they are stabilized by the polar solvent environment. The ICT character found in the excited state of 1a only is the main source of the FL solvatochromism observed in the experiments and ADC(2) calculations.

image file: c7cp06811e-f4.tif
Fig. 4 Surface plots of the natural transition orbitals (NTOs) of the lowest excited states in CH2Cl2. The weights obtained by NTO analysis are shown between parentheses.

The effect of the substitution of electron-donating methoxyl groups with phenyl groups is commonly understood by analyzing its position. In ref. 16, the characters of the excited states of methoxy-phenyl anthracenes were explained by the positional effect; the meta-methoxy group of the anthracenyl ring was considered to stabilize internal charge transfer on the excited state and thus resulted in a red-shift with increasing solvent polarity; this effect was not observed with the ortho- or para-methoxy groups. The meta-effect on the mono-substituted compounds, however, cannot fully explain the properties of the excited states of the di-substituted compounds 1a, 2a, and 3. For example, the 3,5-substituted compound, 2a where two meta positions (both 3- and 5-) are substituted, does not exhibit a charge transfer character in the excited state whereas the 2,5-substituted compound, 1a, exhibits a red-shift with increasing solvent polarity. This implies that the positional effect on the mono-substituted compounds is not additive to that on the di-substituted compounds.

The positional effect on the di-substituted compounds, 1a, 2a, and 3, can be explained by the topology of the molecular orbitals of the di-substituted phenyl group. Fig. 5 and 6 shows the surface plots of the MO of 1a, 2a, 3, and the di-substituted methoxy-benzene molecules. The doubly degenerated HOMO orbitals of the non-substituted benzene molecules are split by the substitution of methoxy groups and the orbitals that are antisymmetric to the mirror plane between the methoxy groups have higher energies (ph:HOMO) than the orbitals that are symmetric to it (ph:HOMO−1). In the case of 1a, 2a, and 3, the HOMO of the pyrene group (pyr:HOMO) interacts with either of the orbitals of the phenyl groups which involve the 2p lobe of the carbon atom directly linked with the pyrene group, i.e. ph:HOMO in the case of 1a and ph:HOMO−1 in the case of 2a and 3, as shown in Fig. 5. In addition, the contribution of the orbitals on the phenyl group to the HOMO of 1a is larger than the contribution of 2a and 3. This is because the energy gap between pyr:HOMO and ph:HOMO is smaller than that between pyr:HOMO and ph:HOMO−1 and the orbital mixing of the former is greater than that of the latter. The charge transfer character of the HOMO to LUMO excitation is therefore larger in 1a than in 2a and 3.

image file: c7cp06811e-f5.tif
Fig. 5 Surface plots of the highest occupied molecular orbitals of 1a, 2a, and 3.

image file: c7cp06811e-f6.tif
Fig. 6 Surface plots of the molecular orbitals of di-substituted methoxyl-benzene molecules. The orbital energies (Eh) are shown in between parentheses.

The theoretical prediction that the lowest excited state of 1a in CH2Cl2 is dominated by the ICT character while that in n-hexane and that of 2a and 3 in CH2Cl2 and n-hexane are dominated by the LE character is consistent with the results of the TRF measurements. After excitation by the pump pulse, the molecular structure should be relaxed towards the energy minimum of the excited state. In Table 3, it can be seen that the dihedral angle θ significantly varies along the structural relaxation on the excited state potential surface; for 1a in CH2Cl2, the dihedral angle varied from θ = 69.1° of min(S0) to θ = 52.3° of the min(ICT). If the excited state has a LE character, i.e. the transitions between the π orbitals are localized on the pyrene groups, it is reasonable to expect that the excited state properties are insensitive to the interactions between the phenyl and pyrene groups because of the local character. If the excited state has an ICT character, properties such as the wavelength and intensity of emission are sensitive to the interaction between the π orbitals on the phenyl and pyrene groups, which would be significantly changed with the dihedral angle θ. Table 6 shows the calculated emission wavelengths and oscillator strength for 1a in CH2Cl2 for various structures with different θ. The structures of the ICT state were optimized with the TD-DFT at the CAM-B3LYP/6-31+G* level of theory and the dihedral angle θ was kept frozen during the geometry optimization. While the emission wavelength is changed by 15 nm between the structures with θ = 70° and θ = 50°, the oscillator strength increases by about two times. This is consistent with the experimental observation that the fluorescence intensity increased with time for 1b in CH2Cl2 while decreasing monotonically in hexane. In addition, because the structural relaxation of the ICT state involves the rotation of the bulky phenyl group, the kinetics can reflect the viscosity of the solvent as observed in the TRF spectra.

Table 6 Emission wavelengths and oscillator strength for 1a in CH2Cl2 solution calculated by ADC(2) for various structures with different dihedral angles θ between the phenyl and pyrene groups
θ = 50° θ = 60° θ = 70°
nm 452 457 467
osci. 0.081 0.064 0.044


We have investigated the fluorescence properties of dialkoxyphenyl-pyrene molecules by experiments and quantum chemical calculations. A characteristic FL solvatochromic shift was observed in the fluorescence spectra of 2,5-dimethoxyphenyl-pyrene (1a), while no significant changes were observed for 3,5- and 2,6-dimethoxyphenyl-pyrene (2a and 3) derivatives in different solvents. Such a drastic change in the photophysical properties, depending on the positions of substitution, is one of the characteristics of biaryl chromophores, which consist of two aryl chromophore groups, and the excited state properties are sensitive to the interactions between the aryl chromophores. In our quantum chemical calculations, it was confirmed that only 1a possesses the ICT excited state as the lowest excited state while the other isomers, i.e.2a and 3, possess the LE excited state of the pyrene group as the lowest excited state. Because the ICT state is sensitive to the polarity of the solvent while the LE state is not, only 1a exhibits clear FL solvatochromism in calculations with the ADC(2) scheme and the COSMO solvent model. This agrees well with the experimental observation on the solvatochromism of 1a. In addition, based on the topology of the molecular orbitals, we have proposed a new positional effect for the two methoxy substitution groups, which can reasonably explain why only 1a has the ICT state as the lowest excited state.

The dynamics of the excited states were further analyzed using picosecond TRFL spectra after excitation; a characteristic increase in the fluorescence intensity was observed for 1b but not for 2b, which can be explained by the difference in the character of the excited states between the (2,5)-motif and the (3,5)-motif, which are confirmed in the NTO analysis for the dimethoxyphenyl-pyrene molecules (1a, 2a, and 3). In fact, the oscillator strength of the ICT increases along with the structural change from the vertical excitation towards the minimum point of the ICT state from the ADC(2) calculations for 1a. The calculation results and experimental data are in excellent agreement with each other. We believe that the current calculation method offers a good tool to predict the FL solvatochromic behaviour of 1-aryl-pyrene chromophores, which is crucial for the reasonable molecular design and practical applications of solvatochromic dyes.

Conflicts of interest

There are no conflicts to declare.


This work was partially supported by Grants-in-Aid for Scientific Research (JSPS KAKENHI Grant Number JP25104011, JP15H03801, JP16H00850, JP16H00855), the NIMS Molecule & Material Synthesis Platform in the “Nanotechnology Platform Project”, and the MEXT-Supported Program for the Strategic Research Foundation at Private Universities, 2015–2019. The computations were performed using Research Center for Computational Science, Okazaki, Japan.


  1. J. R. Casey, S. Grinstein and J. Orlowski, Nat. Rev. Mol. Cell Biol., 2010, 11, 50–61 CrossRef CAS PubMed .
  2. D. Wencel, T. Abel and C. McDonagh, Anal. Chem., 2014, 86, 15–29 CrossRef CAS PubMed .
  3. X. Peng, Z. Yang, J. Wang, J. Fan, Y. He, F. Song, B. Wang, S. Sun, J. Qu, J. Qi and M. Yan, J. Am. Chem. Soc., 2011, 133, 6626–6635 CrossRef CAS PubMed .
  4. H. Sunahara, Y. Urano, H. Kojima and T. Nagano, J. Am. Chem. Soc., 2007, 129, 5597–5604 CrossRef CAS PubMed .
  5. W. Liptay, Angew. Chem., Int. Ed., 1969, 8, 177–188 CrossRef CAS .
  6. S. Nigam and S. Rutan, Appl. Spectrosc., 2001, 55, 362A–370A CrossRef CAS .
  7. S. Sasaki, G. P. C. Drummen and G.-i. Konishi, J. Mater. Chem. C, 2016, 4, 2731–2743 RSC .
  8. G. S. Loving, M. Sainlos and B. Imperiali, Trends Biotechnol., 2010, 28, 73–83 CrossRef CAS PubMed .
  9. Z. R. Grabowski, K. Rotkiewicz and W. Rettig, Chem. Rev., 2003, 103, 3899–4032 CrossRef PubMed .
  10. I. A. Z. Al-ansari, J. Photochem. Photobiol., A, 1993, 72, 15–21 CrossRef CAS .
  11. T. Fiebig, K. Stock, S. Lochbrunner and E. Riedle, Chem. Phys. Lett., 2001, 345, 81–88 CrossRef CAS .
  12. E. Pandurski and T. Fiebig, Chem. Phys. Lett., 2002, 357, 272–278 CrossRef CAS .
  13. W. Weigel, W. Rettig, M. Dekhtyar, C. Modrakowski, M. Beinhoff and A. D. Schlüter, J. Phys. Chem. A, 2003, 107, 5941–5947 CrossRef CAS .
  14. M. Dekhtyar, W. Rettig and W. Weigel, Chem. Phys., 2008, 344, 237–250 CrossRef CAS .
  15. A. Wiessner, G. Huettmann, W. Kuehnle and H. Staerk, J. Phys. Chem., 1995, 99, 14923–14930 CrossRef CAS .
  16. Y.-H. Wang and P. Wan, Photochem. Photobiol. Sci., 2011, 10, 1934–1944 CAS .
  17. N. Basarić, C. Clementi, B. Carlotti, M. Alešković and F. Elisei, J. Photochem. Photobiol., A, 2015, 299, 94–102 CrossRef .
  18. P. M. Pithan, D. Decker, M. S. Sardo, G. Viola and H. Ihmels, Beilstein J. Org. Chem., 2016, 12, 854–862 CrossRef CAS PubMed .
  19. M. Beinhoff, W. Weigel, M. Jurczok, W. Rettig, C. Modrakowski, I. Brüdgam, H. Hartl and A. D. Schlüter, Eur. J. Org. Chem., 2001, 3819–3829 CrossRef CAS .
  20. F. Lu, T. Takaya, K. Iwata, I. Kawamura, A. Saeki, M. Ishii, K. Nagura and T. Nakanishi, Sci. Rep., 2017, 7, 3416 CrossRef PubMed .
  21. S. Sagar and I. R. Green, Synthesis, 2009, 935–940 CAS .
  22. T. Tu, Z. Sun, W. Fang, M. Xu and Y. Zhou, Org. Lett., 2012, 14, 4250–4253 CrossRef CAS PubMed .
  23. S. Schneider, J. Kurzawa, A. Stockmann, R. Engl, J. Daub, P. Matousek and M. Towrie, Chem. Phys. Lett., 2001, 348, 277–284 CrossRef CAS .
  24. F. M. Winnik, Chem. Rev., 1993, 93, 587–614 CrossRef CAS .
  25. CRC Handbook of Chemistry and Physics, CRC Press, 97th edn, 2016 .
  26. T. Yanai, D. P. Tew and N. C. Handy, Chem. Phys. Lett., 2004, 393, 51–57 CrossRef CAS .
  27. R. Krishnan, J. S. Binkley, R. Seeger and J. A. Pople, J. Chem. Phys., 1980, 72, 650–654 CrossRef CAS .
  28. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, G. A. Petersson, H. Nakatsuji, X. Li, M. Caricato, A. V. Marenich, J. Bloino, B. G. Janesko, R. Gomperts, B. Mennucci, H. P. Hratchian, J. V. Ortiz, A. F. Izmaylov, J. L. Sonnenberg, D. Williams-Young, F. Ding, F. Lipparini, F. Egidi, J. Goings, B. Peng, A. Petrone, T. Henderson, D. Ranasinghe, V. G. Zakrzewski, J. Gao, N. Rega, G. Zheng, W. Liang, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, K. Throssell, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. J. Bearpark, J. J. Heyd, E. N. Brothers, K. N. Kudin, V. N. Staroverov, T. A. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. P. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, J. M. Millam, M. Klene, C. Adamo, R. Cammi, J. W. Ochterski, R. L. Martin, K. Morokuma, O. Farkas, J. B. Foresman and D. J. Fox, Gaussian 16, Revision A.03, Gaussian, Inc., Wallingford CT, 2016 Search PubMed .
  29. J. Tomasi and M. Persico, Chem. Rev., 1994, 94, 2027–2094 CrossRef CAS .
  30. J. Schirmer, Phys. Rev. A: At., Mol., Opt. Phys., 1982, 26, 2395 CrossRef CAS .
  31. A. B. Trofimov and J. Schirmer, J. Phys. B: At., Mol. Opt. Phys., 1995, 28, 2299–2324 CrossRef CAS .
  32. D. Rappoport and F. Furche, J. Chem. Phys., 2010, 133, 134105 CrossRef PubMed .
  33. A. Klamt and G. Schüürman, J. Chem. Soc., Perkin Trans. 2, 1993, 799–805 RSC .
  34. R. Ahlrichs, M. Bar, M. Ha[s with combining umlaut]er, H. Horn and C. Kolmel, Chem. Phys. Lett., 1989, 162, 165–169 CrossRef CAS .
  35. C. Hättig, Adv. Quantum Chem., 2005, 50, 37–60 CrossRef .
  36. R. L. Martin, J. Chem. Phys., 2003, 118, 4775 CrossRef CAS .


Electronic supplementary information (ESI) available: Experimental details including synthesis, UV-vis, steady state emission spectra, excitation spectra, picosecond TRFL spectra and data set for photophysical parameters. See DOI: 10.1039/c7cp06811e

This journal is © the Owner Societies 2018