The fluorescence quenching of 3-cyanoperylene upon electron transfer from N,N-dimethylaniline in three room-temperature ionic liquids (RTILs) and in binary solvent mixtures of identical viscosity has been investigated using steady-state and time-resolved fluorescence spectroscopy. This study was stimulated by previous reports of bimolecular electron transfer reactions faster by one or several orders of magnitude in RTILs than in conventional polar solvents. These conclusions were usually based on a comparison with data obtained in low-viscous organic solvents and extrapolated to higher viscosities and not by performing experiments at similar viscosities as those of the RTILs, which we show to be essential. Our results reveal that (i) the diffusive motion of solutes in both types of solvents is comparable, (ii) the intrinsic electron transfer step is controlled by the solvent dynamics in both cases, being slower in the RTILs than in the conventional organic solvent of similar viscosity, and (iii) the previously reported reaction rates much larger than the diffusion limit at low quencher concentration in RTILs originate from a neglect of the static and transient stages of the quenching, which are dominant in solvents as viscous as RTILs.

To access the intrinsic, diffusion free, rate constant of bimolecular photoinduced electron transfer reactions, fluorescence quenching experiments have been performed with 14 donor/acceptor pairs, covering a driving-force range going from 0.6 to 2.4 eV, using steady-state and femtosecond time-resolved emission, and applying a diffusion-reaction model that accounts for the static and transient stages of the quenching for the analysis. The intrinsic electron transfer rate constants are up to 2 orders of magnitude larger than the diffusion rate constant in acetonitrile. Above ∼1.5 eV, a slight decrease of the rate constant is observed, pointing to a much weaker Marcus inverted region than those reported for other types of electron transfer reactions, such as charge recombination. Despite this, the driving force dependence can be rationalized in terms of Marcus theory.

The dynamic Stokes shift of coumarin 153 has been measured in two room-temperature ionic liquids, 1-(3-cyanopropyl)-3-methylimidazolium bis(trifluoromethylsulfonyl)imide and 1-propyl-3-methylimidazolium tetrafluoroborate, using the fluorescence up-conversion technique with a 230 fs instrumental response function. A component of about 10-15% of the total solvation shift is found to take place on an ultrafast time scale < 10 ps. The amplitude of this component is substantially less than assumed previously by other authors. The origin of the difference in findings could be partly due to chromophore-internal conformational changes on the ultrafast time scale, superimposed to solvation-relaxation, or due to conformational changes of the chromophore ground state in polar and apolar environments. First three-pulse photon-echo peak-shift experiments on indocyanine green in room-temperature ionic liquids and in ethanol indicate a difference in the inertial component of the early solvent relaxation of <100 fs.

No abstract

The effect of viscosity on the bimolecular electron transfer quenching of a series of coumarins by N,N-dimethylaniline was investigated using steady-state and time-resolved fluorescence spectroscopy. The data reveal that the static and transient stages of the quenching become dominant as viscosity increases. When extracting the quenching rate constants using a simple Stern-Volmer analysis, a decrease of the rate constant with increasing driving force is observed above ~2 cP. However, this apparent Marcus inverted region, already reported several times with the same system in micelles and room temperature ionic liquids, totally vanishes when analyzing the data with a model accounting for the static and transient stages of the quenching. It appears that the apparent Marcus inverted region arises from the neglect of these quenching regimes together with the use of fluorophores with different excited-state lifetimes.

The driving-force dependence of bimolecular fluorescence quenching by electron transfer in solution, the Rehm-Weller experiment, is revisited. One of the three long-standing unsolved questions about the features of this experiment is carefully analysed here, that is, is there a diffusional plateau? New experimental quenching rates are compiled for a single electron donor, 2,5-bis(dimethylamino)-1,3-benzenedicarbonitrile, and eighteen electron acceptors in acetonitrile. The data are analysed in the framework of differential encounter theory by using an extended version of the Marcus theory to model the intrinsic electron-transfer step. Only by including the hydrodynamic effect and the solvent structure can the experimental findings be well modelled. The diffusional control region, the "plateau", reveals the inherent distance dependence of the reaction, which is shown to be a general feature of electron transfer in solution.

The fluorescence dynamics of perylene in the presence of tetracyanoethylene in acetonitrile was studied experimentally and theoretically, taking into consideration that the quenching is carried out by remote electron transfer in the Marcus inverted region. The initial stage was understood as a convolution of the pumping pulse with the system response accounting for the fastest (kinetic) electron transfer accompanied by vibrational relaxation. The subsequent development of the process was analyzed with differential encounter theory using different models of transfer rates distinguished by their mean square values. The single channel transfer having a bell-shaped rate with a maximum shifted far from the contact produces the ground state ion pair. It was recognized as inappropriate for fitting the quenching kinetics at moderate and long times equally well. A good fit was reached when an additional near contact quenching is switched on, to account for the parallel electron transfer to the electronically excited state of the same pair. The concentration dependence of the fluorescence quantum yield is well fitted using the same rates of distant transfer as for quenching kinetics while the contact approximation applied to the same data was shown to be inadequate

No abstract