Abstract:Singlet fission is the spin-allowed conversion of a spin-singlet exciton into a pair of spintriplet excitons residing on neighbouring molecules. To rationalise this phenomenon, a multiexcitonic spin-zero triplet-pair state has been hypothesised as an intermediate in singlet fission. However, the nature of the intermediate states and the underlying mechanism of ultrafast fission have not been elucidated experimentally. Here, we study a series of pentacene derivatives using ultrafast 2D electronic spectroscopy and unravel the origin of the states involved in fission. Our data reveal the crucial role of vibrational degrees of freedom coupled to electronic excitations that facilitate the mixing of multiexcitonic states with singlet excitons.The resulting manifold of vibronic states drives sub-100-fs fission with unity efficiency. Our results provide a framework for understanding singlet fission and show how the formation of vibronic manifolds with a high density of states facilitates fast and efficient electronic processes in molecular systems. 2" "Singlet fission (SF) is an exciton multiplication process in organic semiconductors that allows one photogenerated spin-singlet excited state to be converted to two spin-triplet excitons.1 !The two generated spin-triplet excitons are initially correlated to form an overall spin-singlet state, making SF a spin-allowed process in contrast to intersystem crossing that involves a spin flip. For systems where the energy of the lowest lying singlet exciton (S) is close to double the energy of the triplet state (T), such as pentacene and its derivatives, SF can occur on a sub-100fs timescale with every singlet being converted to two triplets.2 SF has attracted great attention lately as it enables photovoltaic devices to overcome thermalisation losses by generating two electron-hole pairs per high-energy photon absorbed, potentially allowing single-junction devices that could beat the Shockley-Queisser limit on power conversion efficiency 3 . The first steps towards this goal have been taken with the demonstration of organic solar cells based on pentacene, that show external quantum efficiencies above 129%, the highest for any solar technology to date. 4,5 Despite advances in the experimental characterization of SF in several molecular systems 6-13 as well as extensive theoretical work, 1, 14-22 the fundamental mechanism of ultrafast SF remains unclear. In the kinetic model proposed by Merrifield and co-workers 23 ! the process can be represented as S!TT!T+TWhere: S is the lowest singlet excited singlet state, T is the molecular triplet state and T+T is a pair of fully independent T states. TT corresponds to a doubly excited pair of spin-correlated triplets, forming an overall spin singlet. The TT state, often referred to as the multiexciton state, is considered a dark state that cannot be optically populated from the ground state g, but serves as an intermediate to the formation of free independent triplets T+T.Current theoretical models for SF focus on characterising t...

Articles you may be interested inThree-centered model of ultrafast photoinduced charge transfer: Continuum dielectric approach A perturbation molecular orbital theory of electron-transfer ratesThe capability of multilevel Redfield theory to describe ultrafast photoinduced electron-transfer reactions is investigated. Adopting a standard model of photoinduced electron transfer in a condensed-phase environment, we consider electron-transfer reactions in the normal and inverted regimes, as well as for different values of the electron-transfer parameters, such as reorganization energy, electronic coupling, and temperature. Based on the comparison with numerically exact reference results, obtained using the self-consistent hybrid method, we discuss in some detail the advantages and shortcomings of two different versions of Redfield theory, which employ the time-dependent and stationary Redfield tensor, respectively. The results of the study demonstrate that multilevel Redfield theory, if applied in the appropriate parameter regime, is well suited to describe the ultrafast coherent dynamics of photoinduced electron-transfer reactions.

"Four-wave-mixing" is the generic name for a family of nonlinear electronic and vibrational spectroscopies. These techniques are widely used to explore dissipation, dephasing, solvation, and interstate coupling mechanisms in various material systems. Four-wave-mixing spectroscopy needs a firm theoretical support, because it delivers information on material systems indirectly, through certain transients, which are measured as functions of carrier frequencies, durations, and relative time delays of the laser pulses. The observed transients are uniquely determined by the three-pulse-induced third-order polarization. There exist two conceptually different approaches to the calculation of the nonlinear polarization. In the standard perturbative approach to nonlinear spectroscopy, the third-order polarization is expressed in terms of the nonlinear response functions. As the material systems become more complex, the evaluation of the response functions becomes cumbersome and the calculation of the signals necessitates a number of approximations. Herein, we review alternative methods for the calculation of four-wave-mixing signals, in which the relevant laser pulses are incorporated into the system Hamiltonian and the driven system dynamics is simulated numerically exactly. The emphasis is on the recently developed equation-of-motion phase-matching approach (EOM-PMA), which allows us to calculate the three-pulse-induced third-order polarization in any phase-matching direction by performing three (with the rotating wave approximation) or seven (without the rotating wave approximation) independent propagations of the density matrix. The EOM-PMA is limited to weak laser fields (its domain of validity is equivalent to the approach based on the third-order response functions) but allows for arbitrary pulse durations and automatically accounts for pulse-overlap effects. As an illustration, we apply the EOM-PMA to the calculation of optical three-pulse photon-echo two-dimensional (2D) signals for a generic model system, which represents a characteristic photophysical dynamics of large molecules or chromophores in condensed phases. The EOM-PMA is easy to implement and can straightforwardly be incorporated into any computational scheme, which provides the time-dependent density matrix or wave function of the material system of interest. In particular, EOM-PMA-based computer codes can efficiently be implemented on parallel computers. The EOM-PMA facilitates considerably the computation of four-wave-mixing signals and 2D spectra, in both vibrational and electronic spectroscopy. The EOM-PMA can be extended to higher order optical responses, e.g., heterodyned 3D IR, transient 2D IR, and other six-wave-mixing techniques.

An efficient method has been developed for the calculation of third-order time- and frequency-resolved optical signals. To obtain the general four-wave mixing signal, seven auxiliary density matrices have to be propagated in time. For the special cases of two-pulse photon-echo and transient-grating signals, two or three density matrices, respectively, are required. The method is limited to weak laser fields (it is thus valid within the third-order perturbation theory) but allows for any pulse durations and automatically accounts for pulse-overlap effects. To illustrate the method, we present the explicit derivation of the three-pulse photon-echo signal. Any other third-order optical signal can be calculated in the same manner. As an example, two- and three-pulse photon-echo and transient-grating signals for a weakly damped displaced harmonic oscillator have been calculated.

No abstract