Abstract
The excitation wavelength dependence of the steady-state and time-resolved emission spectra of ethyl 5-(4-aminophenyl)-3-amino-2,4-dicyanobenzoate (EAADCy) in tetrahydrofuran (THF) at room temperature has been examined. It is found that the ratio of the fluorescence intensity of the long-wavelength and short-wavelength fluorescence bands strongly depends on the excitation wavelength, whereas the wavelengths of the fluorescence excitation and fluorescence bands maxima are independent on the observation/excitation wavelengths. The dynamic Stokes shift of fluorophore in locally excited (LE) and intramolecular charge transfer (ICT) states has been studied with a time resolution about 30 ps. The difference between Stokes shift in the LE and ICT states was attributed to the solvent response to the large photoinduced dipole moment of EAADCy in the fluorescent charge transfer state. On this base we can state that, the relaxation of the polar solvent molecules around the fluorophore was observed.
Keywords: Ethyl 5-(4-aminophenyl)-3-amino-2,4-dicyanobenzoate; Donor-acceptor system; Excitation-wavelength dependent fluorescence
Introduction
Fluorescence spectroscopy represents an appropriate technique for investigating structural properties of fluorophores in liquid system [1]. The electronic absorption and emission spectra of molecules in solution, respectively, give reliable information about solvation effects in the ground and excited states. In general, fluorescence emission of molecules is governed by Kasha’s rule which states that fluorescence normally occurs from the lowest vibrational level of the first excited electronic state. It is obvious from this rule that fluorescence should be independent on the excitation wavelengths. On the other hand, it is known for some time, that the fluorescence bands maxima of some system (polar molecules) can shift towards longer wavelengths with an increase in the excitation wavelength [2–15]. This phenomenon is commonly known as red-edge effect [2]. Since this effect is observed only under conditions of restricted mobility, it has been used as a potential tool to estimate the fluorophore (both intrinsic and extrinsic) environment in the organized biological assemblies as membranes, micelles, and proteins [2–15].
It is generally accepted that wavelength-dependent fluorescence involves emission from multiple states or different conformers of the molecule. In a medium where inhomogeneous distribution of species appears, each solute molecule with its own surrounding possesses different absorption and fluorescence band. As a result of this the absorption and emission spectra possess a complex nature i.e., an inhomogeneous broadening of electronic spectra. These conformations of solute molecules interchange rapidly in a fluid medium whereas in a rigid medium the probe molecules are trapped in various geometric configurations.
Ethyl 5-(4-aminophenyl)-3-amino-2,4-dicyanobenzoate (EAADCy, see insert of the Fig. 1) belongs to the fluorophores which show significant red-edge effect [10]. The fluorescence intensity maximum position of the molecule under study in THF glass at 77 K shows considerable dependence on the excitation wavelengths. The origin of this behavior lies in the variety of solute-solvent interaction (fluorescence centers) in the ground and excited states. Such assembles of solute molecules excited in a narrow band causes a photoselection of different conformers of EAADCy, which emit own characteristic band.
Fig. 1.
a Ground-state rotational potential and the corresponding rotamer distribution function n(φ), b Dipole moments in the ground and first excited states as a function of the dihedral angle
In the present paper, we are continuing our previous investigation of the spectral inhomogenity of EAADCy. We present results of systematic studies concerning the excitation-wavelength dependent dual fluorescence of EAADCy. Steady-state and time-resolved measurements are made on molecule under study in THF at room temperature. It is found that the ratio of the fluorescence intensity of the long-wavelength and short-wavelength fluorescence bands (I exc (LW)/I exc (SW) and I em (LW)/I em (SW)) strongly depends on the observation/excitation wavelength, whereas the wavelengths of the fluorescence excitation and fluorescence bands maxima are independent on observation/excitation wavelengths.
Theoretical Background
Recently, several theoretical models of the solvation of molecule in liquid solvent have been developed [1]. Phenomenological description of inhomogeneous broadening of electronic spectra of organic molecules in solution has been given by Nemkovich et al. [11, 12]. They proposed a free energy diagram which describes the dependence of the free energy of the probe-environment centers (solvates) on the strength of the local electric field. According to Nemkovich et al., the ground-state solvate free energy can be written as:
 |
1 |
where 
is the electric field strength of the equilibrium solvate, f is the factor of the reaction field, which, in a linear approximation, equals to the coefficient of proportionality between the field and ground-state electric dipole moment (μ
g) 
. The free energy of the excited solvates is given by [11, 12]:
 |
2 |
where v
0 is the frequency of the pure (0–0) electronic transition of the free probe molecule, h the Planck constant, μ
e is the dipole moment in the excited state. 
is the local electric field in the solvent established by the solute molecules in the thermally equilibrated excited state.
From the above, it is evident that inhomogeneous broadening occurs because different set of solute molecule possesses different solvation shell and different strength of the reaction field R. The range of the inhomogeneous broadening, which dictates photoselection of the energetically different species, depends on the values of dipole moments of the dye molecules in the ground and excited states, and on the dielectric properties and structure of a solvent [11, 12].
As it was mentioned earlier, the excitation-wavelength dependence of fluorescence band intensity is a result of the existence of energetically different luminescence centers—solute molecules with different shell structure of the nearest solvent molecules (e.g., presence of various solute conformers having different geometries and absorbing at different wavelengths). In this case, it is necesary to use the rotamer distribution function n(φ) determined from Boltzmann’s law:
 |
3 |
Excited states processes are usually studied by measurements of the time-resolved emission spectra. In such experiments, light absorption promotes molecules to non-relaxed Franck-Condon state, which undergoes fast transition to the relaxed equilibrium singlet state. This dynamic process can be observed monitoring the time depending position of maximum of the fluorescence spectrum. The relaxation dynamic between solute and solvent molecules are generally expressed in terms of the correlation function c(t), defined as [16]:
 |
4 |
where 
, 
, and 
are the wavenumbers of the fluorescence band maximum at times t, 0, and ∞. The continuum model of the solution [16] predict an exponential decay of the correlation function, τ = exp(−t/τ
L), with a longitudinal relaxation τ
L time (
;τ
D denotes the dielectric relaxation time, ε
0 and 
the static and optical permittivity). It is important to note here that many experiments investigating solvation dynamics have shown that this model does not account fully for the observed changes of c(t).
Experimental Details
The solvent used in the present study (tetrahydrofuran—THF) was the spectroscopic grade and was used without any additional purification. Absorption, fluorescence excitation and emission spectra at room temperature were recorded using, respectively, a Shimadzu UV-2401 PC spectrophotometer and a Shimadzu RF-5301 spectrofluorometer with 5.0 nm band-widths in both excitation and emission. Fluorescence measurements were carried out using the triangular cuvette with frontal excitation to minimize the effect of reabsorption. The emission was observed perpendicular to the direction of the exciting beam. Thus the errors due to fluorescence reabsorption were reduced in a way that mathematical corrections were superfluous.
Time-resolved emission spectra of the sample were measured applying streak camera (C4334-01 Hamamatsu) and 2501 S spectrograph (Bruker Optics). Solid state Nd:YAG laser (PL 2143A/SS EKSPLA) and optical parametric generator (PG 401/SH EKSPLA) were used as an excitation light pulses source [17].
Result and Discussion
Steady-state Spectra of EAADCy
To test whether the changes in the absorption, fluorescence excitation and fluorescence spectra originate from the solute conformers possesing different angles between their interatomic planes of the donor and acceptor moieties, a series of conformational semiempirical quantum mechanical calculations were carried out. We found that for EAADCy molecule the ground-state rotational potential is essentially flat (in the range from 60° to 120°) [18, 19]. Figure 1a shows the calculated ground-state (S
0) rotational potential and the corresponding rotamer distribution function n(φ) determined from Boltzmann’s low at 298 K, for the molecule under study. According to the Boltzmann’s low, in the ground state about 33% of the all EAADCy molecules possess twist angle between two fragments (donor and acceptor) in the range from 60° to 120°. Thus, quantum-chemical calculations suggest the presence of various conformers in the ground state which have different geometry and absorb at different wavelengths. In addition, it should be noted that φ
eq (obtained using quantum-chemical calculations) does not differ from the mean twisting angle 
, which is convoluted with the distribution function n(φ), and equals 
.
As it was mentioned earlier, the range of inhomogeneous broadening depends on the values of dipole moments of the dye molecules in the ground and excited states. From the quantum-chemical calculations [18, 19] (see Fig. 1b) it is evident that dipole moment of EAADCy in the ground state shows a very week dependence on the dihedral angle, whereas in the first excited state planar molecule possesses five times greater dipole moment than that with perpendicular donor-acceptor arrangement (i.e., 15.6 D versus 3.0 D). The large differences between 
and 
suggest existence of molecules possessing different solvation shell and different strength of the reaction field R.
Figure 2a presents the observation-wavelength dependence of fluorescence excitation spectra (maxima position and shape of the spectra) of EAADCy in THF at room temperature. The major observation that can be made suggests the strong dependence of the fluorescence excitation spectrum on the observation wavelength. The long-wavelength fluorescence excitation band of EAADCy are composed of at least two bands, centered at about 360 nm and 430 nm. As it was shown in Fig. 2a, upon increasing the observation wavelength the intensity of the short-wavelength band at 360 nm decreases. This behavior is accompanied by an increase of the long-wavelength band centered at 430 nm. It is important to note here that each of the area under the fluorescence excitation and fluorescence (Fig. 2a and b) curves were normalized to unity. In this case, each of fluorescence spectrum corresponds to the emission of the same number of emitting EAADCy molecules.
Fig. 2.
Fluorescence excitation spectra (a) and fluorescence spectra of EAADCy in THF at room temperature as a function of λ obs (a) and λ exc (b)
We also carefully examined the excitation-wavelength dependence of fluorescence spectra of EAADCy in THF. We found that upon increasing the excitation wavelength the intensity of short-wavelength fluorescence band decreases. This behavior is accompanied by an increase of the long-wavelength emission band.
Analyzing the fluorescence excitation (Fig. 2a) and fluorescence (Fig. 2b) spectra of EAADCy in THF obtained at different observation and excitation wavelengths, it is found that the ratio of the fluorescence intensity of the long-wavelength and short-wavelength fluorescence bands (I em (LW)/I em (SW)) strongly depend on the excitation wavelength. A typical excitation- and observation-wavelength dependent behavior is highlighted in Fig. 3. As it can be seen (Fig. 3b), when the excitation wavelength is increased from 360 nm to 445 nm, the ratio I em (LW)/I em (SW) increases from ~0.5 to ~8. It is important to note here that the ratio I em (LW)/I em (SW) is of the solute concentration independent (with the limit of experimental error).
Fig. 3.
The ratio I exc(SW)/I exc(LW) (a) and I em(LW)/I em(SW) (b) as a function of λ obs and λ exc, respectively
The steady-state fluorescence excitation spectra are also strongly observation-wavelength dependent. As can be seen (Fig. 3a) the ratio I exc (SW)/I exc (LW) undergoes a significant decrease when increasing the observation wavelength (e.g., from ~6.5 (λ obs = 480 nm) to ~0.8 (λ obs = 605 nm). This observation/excitation wavelength dependence is similar to the typical red-edge effect observed in low-temperature glasses, polymer matrixes, and organized assemblies such as micelles, vesicles, proteins and membranes [2–15].
These results, as well as, our previous study [10, 18–20] indicate that for EAADCy a distribution of spatial conformers (lie in difference of rotational angles between donor and acceptor moieties) is always present (see Fig. 1). When the excitation is carried out at the maximum of the absorption band, the conformers having perpendicular or nearly perpendicular geometry (φ ~ 90°) are excited. As the excitation wavelength is shifted to the red, the rotamer with different angles (0 < φ < 90°) between their interatomic planes is selectively excited. It can be concluded that ICT state (long-wavelength fluorescence band) may be formed either by further geometrical changes of the normal conformer (φ ~ 90°) in the locally excited state or directly by excitation of flattened conformers.
Time-resolved Emission Spectra and Fluorescence Decay Times of EAADCy
In order to confirm the interpretation of the steady-state measurements data, the time-resolved emission spectra and from them calculated fluorescence decay times of EAADCy in THF were investigated. These measurements were carried out at 5 ns range of streak camera window, what gave us point-to point time resolution of 10.86 ps per point. Figure 4 shows the emission spectra of EAADCy in THF at various times after excitation (λ
exc = 400 nm). As can be seen, upon increasing the time after excitation the short-wavelength (LE) and long-wavelength (ICT) fluorescence maxima positions are red-shifted. Figure 4 clearly shows that a pronounced shift of the SW and LW bands occurs already at very short time after excitation. In order to make it more evident, the temporal behavior of 
and 
are shown in Fig. 5.
Fig. 4.
Time-resolved emission spectra of EAADCy in THF at room temperature at various times after excitation (λ exc = 400 nm)
Fig. 5.
Time-dependence of the frequency of the short-wavelength (SW) and long-wavelength (LW) emission band maximum of EAADCy in THF
In order to determine the Stokes-shift correlation function c(t), the time-dependent frequencies 
and 
are normalized according to Eq. 4. Figure 6 illustrates the resulting c(t) curves for short-wavelength and long-wavelength fluorescence bands. The decay times (obtained for different λ
exc) of c(t) are summarized in Table 1. Analyzing the experimental decay data, it follows that the obtained 
and 
values do not depend on λ
exc. From the 
and 
values obtained for both of the conformers (“perpendicular”—LE form, “flattened”—ICT form), it is evident that the longitudinal relaxation time is very different. The ratio of the corresponding longitudinal relaxation times 
is about 3. This large difference between 
and 
values is understandable taking into account different interactions of the conformers with the reactive electric field of the surrounding solvent. In our previous publication [18], it was shown that the “flattened” (ICT) form of EAADCy possesses greater dipole moment in the excited state than in the ground state, while “perpendicular” (LE) form of EAADCy possesses lower dipole moment value in the excited state in comparison to its value in the ground state. The following changes of dipole moment were obtained: 
, 
.
Fig. 6.
Decay of the solvent correlation function c(t) of LE (b) and ICT (a) form of EAADCy in THF at room temperature
Table 1.
The longitudinal relaxation times (
, 
), fluorescence decay times (τ
SW, τ
LW) and preexponential factors (A
SW, A
LW) of EAADCy in THF
|
|
τ SW (ps) | A SW (%) | τ LW (ps) | A LW (%) | χ2 |
|---|---|---|---|---|---|---|
| λ exc = 370 nm | ||||||
| 162 ± 20 | 53 ± 4 | 5.65 | 78.2 | 0.22 | 21.8 | 1.21 |
| λ exc = 390 nm | ||||||
| 170 ± 15 | 64 ± 8 | 5.83 | 66.4 | 0.21 | 33.6 | 1.09 |
| λ exc = 400 nm | ||||||
| 187 ± 20 | 59 ± 5 | 5.73 | 53.3 | 0.25 | 46.7 | 1.18 |
| λ exc = 420 nm | ||||||
| 182 ± 12 | 62 ± 8 | 5.55 | 43.2 | 0.20 | 56.8 | 1.22 |
The experimental decay curves were fitted to a multiexponential function 
, where A
i and τi are the preexponential coefficient (describing the contribution of the i-th fluorescence decay component to the total emission) and decay time of the i-th component, respectively. The average value of χ
2, describing the goodness of the fitting procedure for all decay curves, i.e., one- and two-exponential fitting, equals 
.
Table 1 collects the calculated fluorescence decay time data, τ, and the preexponential factors, A, describing the contribution of the i-th fluorescence decay component of the total emission. This results point that the luminescent solution of EAADCy in THF established a spectroscopic inhomogeneous system attributed to the emission from locally excited, S 1 (LE), and intramolecular charge transfer, S 1 (ICT), states. For all studied excitation wavelengths (370 nm, 390 nm, 400 nm, 420 nm) emission decay exhibits fast (τ LW ≅ 0.22 ns) and slow (τ SW ≅ 5.7 ns) decay components. The slow decay component originated from the emission of locally excited state (perpendicular conformer), whereas the fast decay component results from an ensemble of spatial conformers possessing the ICT structure (flattened conformer). It is important to note here that the relative amplitude of LE and ICT emissions strongly depend on the excitation wavelength. The preexponential factor A SW (describing the contribution of the LE fluorescence decay component of the total emission) decreases with increasing excitation wavelength from 78% for excitation at 370 nm to 43% for excitation at 420 nm. In contrast, as the excitation wavelength is increased from 370 nm to 420 nm, the A LW increases from 22% to 57%.
Conclusions
Analyzing the data obtained from the steady-state, time-resolved measurements and quantum-chemical calculations one can state that:
At room temperature, EAADCy exists as an equilibrium mixture of various conformers having different geometry (in the range from 60° to 120°) and absorb at different wavelengths. As the excitation wavelength is changed, a slightly different species are excited.
The ratio of the fluorescence intensity of the long-wavelength and short-wavelength fluorescence bands strongly depends on the observation/excitation wavelength, whereas the wavelengths of the fluorescence excitation and fluorescence bands maxima are independent on the observation/excitation wavelengths in accordance to the Kasha rule.
The difference between Stokes shift in the LE and ICT states was attributed to the solvent response to the large photoinduced dipole moment of EAADCy in the fluorescent charge transfer state.
The excitation wavelength dependent biexponential decay times (and preexponential factors A SW and A LW) of EAADCy in THF lead us to conclude that ICT state really may be formed either by further geometrical changes of the locally excited conformer or directly by excitation of flattened conformers.
Acknowledgements
The authors express their gratitude to Professor Józef Heldt, and Professor Jerzy Pączkowski for scientific discussions and valuable comments on the manuscript. This work was supported by The Ministry of Science and Higher Education of Poland research Grant N N204 054335 and by The University of Gdansk, Grant BW-5209-5-0348-0. The authors are grateful for these supports.
Open Access
This article is distributed under the terms of the Creative Commons Attribution Noncommercial License which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.
References
- 1.Lakowicz JR. Principle of fluorescence spectroscopy. New York: Kluwer Academic/Plenum Publishing Corporation; 2006. [Google Scholar]
- 2.Demchenko AP. Luminescence. 2002;17:19. doi: 10.1002/bio.671. [DOI] [PubMed] [Google Scholar]
- 3.Demchenko AP, Sytnik AI. Proc Natl Acad Sci USA. 1991;88:9311. doi: 10.1073/pnas.88.20.9311. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Sengupta B, Guharay J, Chakraborty A, Sengupta PK. Spectrochim Acta A. 2002;58:2005. doi: 10.1016/s1386-1425(01)00651-5. [DOI] [PubMed] [Google Scholar]
- 5.Kallir AJ, Suter GW, Wild UP. J. Phys. Chem. 1987;91:60. [Google Scholar]
- 6.Demchenko AP, Sytnik AS. J Phys Chem. 1991;95:10518. [Google Scholar]
- 7.Nemkovich NA, Rubinov AN, Tomin VI. Inhomogeneous broadening of electronic spectra of dye molecules in solutions, Chap. 8. In: Lakowicz JR, editor. Topics in fluorescence spectroscopy. New York: Plenum; 1991. [Google Scholar]
- 8.Chattopadhyay A, Mukherjee S. J Phys Chem B. 1999;103:8180. [Google Scholar]
- 9.Józefowicz M, Heldt JR, Heldt J. Z Naturforsch. 2002;57 a:787. [Google Scholar]
- 10.Józefowicz M, Heldt JR, Bajorek A, Pączkowski J. J Photochem Photobiol A. 2008;196:38. [Google Scholar]
- 11.Nemkovich NA, Rubinov AN. J Fluoresc. 1995;5:285. doi: 10.1007/BF00723900. [DOI] [PubMed] [Google Scholar]
- 12.Nemkovich NA, Kruchenok JV, Rubinov AN, Pivovarenko VG, Baumann W. J Photochem Photobiol A. 2001;139:53. [Google Scholar]
- 13.Guha S, Rawat SS, Chattopadhyay A, Bhattacharyya B. Biochemistry. 1996;35:13426. doi: 10.1021/bi961251g. [DOI] [PubMed] [Google Scholar]
- 14.Hof M, Lianos P, Laschewsky A. Langmuir. 1997;13:2181. [Google Scholar]
- 15.Hof M, Lianos P. Langmuir. 1997;13:290. [Google Scholar]
- 16.Maroncelli M, Fleming GR. J Chem Phys. 1987;86:6221. [Google Scholar]
- 17.Kubicki A, Bojarski P, Grinberg M, Sadownik M, Kukliński B. Opt Comm. 2006;263:275. [Google Scholar]
- 18.Józefowicz M, Aleksejew M, Heldt JR, Bajorek A, Pączkowski J, Heldt J. Chem Phys. 2007;338:53. [Google Scholar]
- 19.Józefowicz M, Heldt JR, Bajorek A, Pączkowski J. Chem Phys. 2009;363:88. [Google Scholar]
- 20.Józefowicz M, Heldt J, Heldt JR, Aleksiejew M, Iwulska AM, Pączkowski J. Optical Materials. 2009;31:1894. [Google Scholar]