| [page 89↓] |
Abstract
Pyrrolobenzenes with different substitution patterns, MP2BN and MP2-B25CN, are investigated by using steady-state and time-resolved optical spectroscopy. The absorption and fluorescence spectra of these compounds are red shifted with respect to the parent compound p-PBN, indicating a stabilisation of the Franck-Condon (FC) excited state by mesomeric interaction. Both the position and strength of the electron acceptor moiety influence the emission characteristics of these molecules. The large radiative rate constant of MP2BN indicates an allowed emission due to mesomeric interaction between the donor and acceptor moieties, (MICT), whereas in the case of p-PBN and MP2-B25CN, the reduced radiative rate constant indicates a forbidden emission from a twisted intramolecular charge transfer (TICT) state.
The photophysics of donor-acceptor substituted benzenes are of great interest in the study of intramolecular charge transfer (ICT) states. ICT states are commonly observed in N-phenylpyrrole (see Fig. 8.1) and several sterically hindered derivatives [29, 66, 82]and are identified by their large fluorescence red shift in medium and strongly polar solvents. ICT compounds either have an allowed emission (high transition dipole moment, M f ) or a forbidden emission (small M f ). The difference between the two types of CT states is the mesomeric interaction between the molecular subsystems [40].ICT states with allowed emission are generally found in nearly coplanar compounds with high mesomeric interaction (mesomeric ICT state, MICT). For planar systems, this corresponds to the so-called PICT (Planar Intramolecular Charge Transfer) state as introduced by Zachariasse et al [69], but is more general and can also be used for compounds which cannot become planar but nevertheless show dual fluorescence [87]. ICT states with forbidden emission in nearly perpendicular compounds with small mesomeric interaction are commonly called TICT (Twisted Intramolecular Charge Transfer) states [1, 9].
| [page 90↓] |
Figure 8.1: Molecular structures of the investigated compounds and their abbreviated formulas. | ||
Phenyl pyrroles contain an sp2 hybridized nitrogen, thus the inter-moiety twist angle α is better defined than in dimethylanilino derivatives with a pyramidal nitrogen, and these compounds can in principle populate an ICT state in the twisted geometry [29, 66, 82]. In other cases, the population of nonfluorescent TICT-like states has been identified as the cause for intramolecular fluorescence quenching in many commonly used dyes [88, 89] and even laser dyes [90, 91, 92]. According to the TICT model, the fluorescence can originate from the primarily excited ‘locally excited’ (LE) state as well as from the charge transfer (CT) state accessible only by an adiabatic photoreaction from the LE state, which includes a torsional motion around the bond linking the donor and acceptor moieties. The relative amount of TICT fluorescence depends on the height of the barrier separating the LE and CT states and on their energy difference. If there is no barrier between these two excited states and their energy difference is large enough, the LE state population is rapidly converted into the TICT state and only the long wavelength emission from the charge transfer state is observable. This TICT emission probability is usually small, i.e. with a reduced value of the transition moment, due to the small π overlap in the strongly twisted arrangement of chromophores [16, 19, 93].
Recently, theoretical studies on N-pyrrolobenzene (PB), N-pyrrolobenzonitrile (PBN) and 4-N,N-dimethylaminobenzonitrile (DMABN) for the ground and excited states were done by Parusel [78] using a DFT/MRCI approach. It was found that the TICT state of PB is stabilized only in polar solvents whereas in the case of p-PBN, the TICT state is more stable than the LE state even in non polar solvents.
| [page 91↓] |
Zilberg et al. [73] calculated the energies, dipole moments and molecular structures of the locally excited (LE) and charge transfer (CT) states of DMABN, PP and p-PBN by DFT (used only for optimization), CASSCF and CASPT2 methods (for calculating energies). Their calculations support the existence of two distinct structures for the CT states. One possesses a quinoid structure that has a coplanar arrangement of the chromophores at the energy minimum, corresponding to the MICT [40] or PICT model [69, 71] (in both cases a planar intramolecular charge-transfer state with large interchromophoric coupling). The other one has an antiquinoid bond length distribution in the benzene ring and an energy minimum for the twisted structure connected with a larger dipole moment consistent with the TICT model [9].
Recently, Yoshihara et al. [67] experimentally determined the excited state dipole moments of both ICT and LE states for dual fluorescent molecules such as N-phenylpyrrole, N-(4-methylphenyl)pyrrole, N-(4-cyanophenyl)pyrrole and N-(3-cyanophenyl)pyrrole from solvatochromic and thermochromic measurements. They also compared the experimental ICT state dipole moments with theoretical values from the literature calculated for both coplanar (PICT) and twisted (TICT) conformations of the phenyl pyrrole and cyanophenyl pyrroles. It was found that for N-phenylpyrrole and N-(4-methylphenyl)pyrrole the main fluorescence is of LE character, while N-(4-cyanophenyl)pyrrole and N-(3-cyanophenyl)pyrrole exhibit major emission from the ICT state.
In the PICT and TICT states of pyrrolobenzenes, the pyrrole unit acts as the donor moiety and benzonitrile as the acceptor, and electron transfer (ET) takes place from the pyrrolo group to the center of the phenyl ring, or still farther toward the acceptor substituent. The aim of the present study is to investigate whether the CT nature (either allowed (MICT/PICT or forbidden (TICT)) can be influenced by changing the position of the acceptor moiety on the pyrrole unit, and also how it depends on the acceptor strength. For this purpose, the pyrrolo benzonitrile derivatives 2-(4-cyanophenyl)pyrrole (MP2BN) and 2-(2,5-cyanophenyl)pyrrole (MP2-B25CN) have been synthesized with a linkage in 2-position of pyrrole, and their photophysical characteristics have been compared with those of p-PBN.
The experimental details about the absorption, fluorescence, lifetime and quantum yield measurements are described in chapter 3.
| [page 92↓] |
The absorption and fluorescence spectra of p-PBN, MP2BN and MP2-B25CN in solvents of varying polarity are shown in Fig. 8.2. The corresponding photophysical parameters are collected in Table 8.1. The absorption spectrum of p-PBN is a single and broad band whereas in the case of MP2BN and MP2-B25CN, a slightly structured band can be observed. For all three compounds, the absorption maximum changes very little with the polarity of the solvent. In nonpolar hexane, the absorption and fluorescence maxima of MP2BN and MP2-B25CN are significantly red-shifted with respect to the parent compound (p-PBN), indicating the stabilization of both their excited and ground states by the larger mesomeric interaction.
Figure 8.2: Normalised Absorption and fluorescence spectra of p-PBN, MP2BN and MP2-B25CN in solvents of different polarity (Hex – n-hexane, BOB – dibutylether, EOE – diethylether, THF – tetrahydrofuran, ACN – acetonitrile. | ||
|
Table 8.1: Spectral and photophysical data of p-PBN, MP2BN and MP2-B25CN in various solvents at room temperature.
On the other hand, the solvatochromic shift of the fluorescence of these compounds is much smaller (3400 and 4700cm-1) than for p-PBN (8500 cm-1) indicating smaller changes in their dipole moment from ground to excited state. The fluorescence quantum yield decreases from solvents of low to high-polarity for all three compounds investigated in this work. However one notes that p-PBN is weakly fluorescent with a yield smaller than 8%, MP2-B25CN exhibits a yield smaller than 40%, whereas MP2BN, has a yield close to 1 in all solvents except in acetonitrile where it drops to 50%. As pointed out below, these differences can mainly be traced back to changes in the radiative rate constant k rad .
| [page 94↓] |
The schematic excited-state hypersurfaces for p-PBN, MP2BN and MP2-B25CN as a function of the twist angle are shown in Fig. 8.3. It explains the relative energetic position of the excited states both in the planar (0°) and perpendicular (90°) geometries. In going from 0° to 90° there is a stabilization of the TICT state (twisted geometry) for p-PBN and MP2-B25CN.
Figure 8.3:. Schematic representation of the excited state hypersurfaces for p-PBN, MP2BN and MP2-B25CN in the planar and perpendicular geometry. (LE0 - locally excited state; CT0 - charge transfer state, both without mesomeric interaction; Emes- Mesomeric energy – zero for the perpendicular, of varying size for the planar geometry resulting in the TICT (perpendicular) and the MICT states (planar geometry; ΔEM-T - energy difference between MICT and TICT states; ΔE0- energy difference LE0 and CT0 states) | ||
On the other hand, for MP2BN as compared to p-PBN, the MICT state (planar form) is more stable than the TICT state due to the increased mesomeric interaction between the two submoieties. This is a consequence of the different size of the coefficients of the donor orbital on the pyrrole, with a large value on carbon atom 2 (MP2BN) and a node on the nitrogen (p-PBN). In MP2-B25CN, the TICT state is lower than the MICT state in spite of the larger mesomeric interaction, due to the increased energy gap ΔE0 between the zero order LE and TICT states.
| [page 95↓] |
In Table 8.2, the ground and excited state dipole moments of all three compounds are collected. The excited state dipole moments, μ
e are calculated from the plot of the solvatochromic shift of the emission maxima (νf) versus the solvent polarity function (see fig. 8.4a and fig. 8.4b), and are calculated from the following equation [53, 54]:
where ε is the dielectric constant and n the refractive index of the solvent
with ρ the density of the compound, M its molecular weight and NA Avogadro's number.
Figure 8.4a: Fluorescence maxima of p-PBN, MP2BN and MP2-B25CN at room temperature versus the solvent polarity parameter Δf’ (see text). | ||
| [page 96↓] |
Figure 8.4b: Fluorescence maxima of p-PBN in diethylether versus Δf’at different temperatures, indicated in Kelvin on the curve. | ||
The Onsager radii a given in Table 8.2 for MP2BN and MP2-B25CN were calculated relativeto p-PBN from the mass-density formula (eq. 8.2) by assuming equal densities ρ. The ground-state dipole moments, μ g, are calculated by using the AM1 (Austin model 1) semiempirical method [42] of the AMPAC software package [44]. As can be seen from Table 8.2, the μ g values for p-PBN and MP2-B25CN are found to be similar and about half that of MP2BN. The equilibrium twist angle is found to increase from p-PBN (α = 23.3°), to MP2BN (α = 37.5°) and to MP2-B25CN (α = 46.1°), which normally leads to a decrease of the dipole moment. The increased ground-state dipole moment for MP2BN is therefore due to a better electronic coupling between the two moieties, because the pyrrole group is linked in 2-position where the orbital coefficient is large. In MP2-B25CN, the dipole moment is decreased with respect to MP2BN by two sources: the increased twist angle and the CN substituents in positions where they will keep any transferred charge near the center of the benzene ring whereas in MP2BN this charge is pulled towards the CN substituent and hence moved further away from the donor pyrrole group.
Similar arguments hold for the excited state. In contrast to the ground state dipole moments, however, the μ e values of MP2BN and MP2-B25CN are smaller than for the reference compound p-PBN. This can be rationalized by the assumption, supported by the experimental k rad values, that two different CT states are populated. As detailed above, MP2BN can be assumed to populate a state with maximal mesomeric interaction, close to [page 97↓]planarity. In this MICT state, the dipole moment must be lower than for the extreme situation of the 90° twisted TICT state [9, 19],because mixing between the pure CT and nonpolar states occurs. As supported by the very low k rad values (see below), p-PBN and MP2-B25CN both populate a TICT state, which differs however by the center of charge distribution in the LUMO of the acceptor moiety. For p-PBN, the center of negative charge of the acceptor orbital is located further away from the donor pyrrole group, whereas for MP2-B25CN, since the cyano groups are located in the ortho and meta positions with respect to the donor group, the center of negative charge is positioned in the middle of the acceptor orbital. This must lead to a smaller value of μ e for the TICT state in MP2-B25CN. Thus, the MICT character of MP2BN is characterized by two observables: the somewhat reduced value of the dipole moment when compared to the other two compounds in the excited state, and the increased radiative rate constant.
The fluorescence decay curves are monoexponential, which allows the evaluation of radiative and nonradiative rate constants, k rad and knr respectively, according to equations 8.3 and 8.4. In eqn. 8.4, k nr corresponds to the sum of all nonradiative processes including triplet formation. The measured data and calculated photophysical values for p-PBN, MP2BN and MP2-B25CN are collected in Tables 8.1.
The k rad values for all three compounds decrease from non-polar to polar solvents, indicating a less allowed emission in solvents of higher polarity. For both p-PBN and MP2-B25CN, k rad values in highly polar solvents are about one order of magnitude smaller as compared to MP2BN. Particularly, the k rad values (see Table 8.1) in acetonitrile are below107 s-1 in p-PBN and MP2-B25CN indicating a forbidden emission from a TICT state whereas in the case of MP2BN the CT emission is tenfold more allowed. We assign this emission to a CT state with near planar geometry, i.e. partial twisting and mesomeric stabilization (MICT state).
| [page 98↓] |
By changing the position of the acceptor part, and also by increasing its acceptor strength by introducing two cyano substituents from MP2BN to MP2-B25CN, interesting changes in the photophysical properties are observed. As discussed above, the nature of the emissive state can be judged from the magnitude of the radiative rate constant (Tables 8.1). The spectra are also affected: as can seen from fig. 8.2, for MP2BN, where the acceptor has been substituted in 4-position with respect to the donor part of methyl-pyrrole, a blue shift of the emission in both hexane and acetonitrile is observed with respect to the emission of MP2-B25CN. On the other hand, the absorption and emission maxima of MP2-B25CN are both red-shifted with respect to those of p-PBN and MP2BN. This can be attributed to the increased acceptor strength in MP2-B25CN because CT transitions vary with the donor acceptor properties of the constituents.
The transient absorption spectra measured for MP2BN in medium (THF) and strongly polar (ACN) solvents by subpicosecond transient absorption spectroscopy show a dominant absorption band below 400 nm and a residual gain band in the red-wing of the fluorescence spectrum (see fig.8.5). The absorption band is similar to that of the benzonitrile radical anion [94]but is somewhat blue shifted due to the overlap with the gain band. The observation of a gain band indicates that the stimulated-emission cross section is large enough, which is consistent with the allowed MICT-nature of the excited-state. Such a gain band is indeed not observed for DMABN [94] although one cannot exclude that for this latter compound the excited state absorption is only slightly dominant so that the sum of gain and absorption yields a relatively weak transient absorption signal.
| [page 99↓] |
Figure 8.5 : Transient absorption spectra of MP2BN in acetonitrile (red) and in THF (black) measured 50 ps after excitation with a subpicosecond laser pulse. The stimulated emission of the MICT state is observed in the red-wing of the fluorescence spectra, (the maxima of which are indicated by the vertical arrows) due to the overlap with the transient absorption band. | ||
Steady-state and time-resolved spectroscopy yielded evidence that the excited state CT character in the phenyl pyrrole derivatives p-PBN, MP2BN and MP2-B25CN is different. The MICT character (large mesomeric interaction for near-planar geometries) of MP2BN as compared to the TICT character of p-PBN and MP2-B25CN is supported by its reduced excited-state dipole moment, the enhanced radiative rate constant k rad values and by the observation of stimulated emission in pump-probe experiments. The reason for the increased mesomeric interaction in MP2BN can be traced back to a more efficient orbital interaction for the phenylpyrroles linked in the 2-position. Because the HOMO of the pyrrole has a node on the nitrogen, the HOMO of planar p-PBN is localised on the pyrrole, but delocalised for near-planar MP2BN where the linkage of the molecular moieties involves two carbon atoms with large orbital coefficients. In MP2-B25CN we also expect a MICT state, but the TICT state is energetically lowered much more strongly due to the increased acceptor strength (lower reduction potential) of dicyano-benzene in comparison with mono-cyano benzene, so that it may be the only minimum in polar solvents.
© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme. | ||
DiML DTD Version 4.0 | Zertifizierter Dokumentenserver der Humboldt-Universität zu Berlin | HTML generated: 07.07.2005 |