Analysis of the vibronic structure of the trans-stilbene fluorescence and excitation spectra: the S0 and S1 PES along the Ce[double bond, length as m-dash]Ce and Ce-Cph torsions.

We analyze the highly resolved vibronic structure of the low energy (≤200 cm-1) region of the fluorescence and fluorescence excitation spectra of trans-stilbene in supersonic beams. In this spectral region the vibronic structure is associated mainly with vibrational levels of the Ce-Ce torsion (τ) and the au combination of the two Ce-Cph bond twisting (ϕ). We base this analysis on the well-established S0(τ, ϕ) two-dimensional potential energy surface (PES) and on a newly refined S1(τ, ϕ) PES. We obtain vibrational eigenvalues and eigenvectors of the anharmonic S0(τ, ϕ) and S1(τ, ϕ) PES using a numerical procedure based on the Meyer's flexible model [R. Meyer, J. Mol. Spectrosc., 1979, 76, 266]. Then we derive Franck-Condon factors and therefore intensities of the relevant vibronic bands for the S0 → S1 excitation and S1 → S0 fluorescence spectra. Furthermore, we assess the role of the bg combination of the two Ce-Cph bond twisting (ν48) in the structure of the S1 → S0 fluorescence spectra. By the use of these results we are able to assign most of the low energy vibrational levels of the S0 → S1 excitation spectra and of the fluorescence spectra of the emission from several low energy S1 vibronic levels. The good agreement between the observed and the computed vibrational structure of the S0 → S1 and S1 → S0 spectra suggests that the proposed picture of the E1(τ, ϕ) and E0(τ, ϕ) PES, in particular along the coordinate τ governing trans-cis photo-isomerization in S1, is accurate. In S0, the barriers for the Ce[double bond, length as m-dash]Ce torsion and for the au type Ce-Cph bond twisting are 16 080 cm-1 and 3125 cm-1, respectively, while in S1, where the bond orders of the Ce[double bond, length as m-dash]Ce and Ce-Cph bonds are reversed, the two barriers become 1350 cm-1 and 8780 cm-1, respectively.

Excited state evolution of DNA stacked adenines resolved at the CASPT2//CASSCF/Amber level: from the bright to the excimer state and back.

Deactivation routes of bright ππ* (La) and excimer charge transfer (CT) states have been mapped for two stacked quantum mechanical (CASPT2//CASSCF) adenines inside a solvated DNA double strand decamer (poly(dA)·poly(dT)) described at the molecular mechanics level. Calculations show that one carbon (C2) puckering is a common relaxation coordinate for both the La and CT paths. By mapping the lowest crossing regions between La and CT states, together with the paths connecting the two states, we conclude that at least one CT state can be easily accessible. The lowest-lying conical intersections between ground state (GS) and CT states have been fully characterized in a realistic DNA environment for the first time. We show that the path to reach this crossing region from the CT minima involves high barriers that are not consistent with experimental data lifetimes. Instead, the multiexponential decay recorded in DNA, including the longest (ca. 100 picoseconds) lifetime component detected in oligomeric single- and double-stranded systems, is compatible with both intra-monomer relaxation processes along the La deactivation path (involving small barriers) and the population of the excimer (CT) state that behaves as a trap. In the latter case, deactivation is feasible only going back to the La state by following its preferred decay coordinate.

Tracking the stilbene photoisomerization in the S(1) state using RASSCF.

In this work we compute the S1 potential energy curve responsible for stilbene cis-trans photoisomerisation employing the RASSCF approach, since the standard CASPT2//CASSCF protocol appears to be unsatisfactory in describing the stilbene S1 state. We find that RASSCF calculations, which are based on relatively few (but well chosen) configurations, produce qualitatively correct results and accurate relative excited state energies, both in the twisted and in the cis and trans regions of stilbene.

Product formation in rhodopsin by fast hydrogen motions.

The photochemical cis-trans isomerization of retinal in rhodopsin is investigated by structure sampling and excited state QM/MM trajectories with surface hopping. The calculations uncover the motions responsible for photoproduct formation and elucidate the reasons behind the efficient photoisomerization in the primary event of visual transduction.

Conical intersection dynamics of the primary photoisomerization event in vision.

Ever since the conversion of the 11-cis retinal chromophore to its all-trans form in rhodopsin was identified as the primary photochemical event in vision, experimentalists and theoreticians have tried to unravel the molecular details of this process. The high quantum yield of 0.65 (ref. 2), the production of the primary ground-state rhodopsin photoproduct within a mere 200 fs (refs 3-7), and the storage of considerable energy in the first stable bathorhodopsin intermediate all suggest an unusually fast and efficient photoactivated one-way reaction. Rhodopsin's unique reactivity is generally attributed to a conical intersection between the potential energy surfaces of the ground and excited electronic states enabling the efficient and ultrafast conversion of photon energy into chemical energy. But obtaining direct experimental evidence for the involvement of a conical intersection is challenging: the energy gap between the electronic states of the reacting molecule changes significantly over an ultrashort timescale, which calls for observational methods that combine high temporal resolution with a broad spectral observation window. Here we show that ultrafast optical spectroscopy with sub-20-fs time resolution and spectral coverage from the visible to the near-infrared allows us to follow the dynamics leading to the conical intersection in rhodopsin isomerization. We track coherent wave-packet motion from the photoexcited Franck-Condon region to the photoproduct by monitoring the loss of reactant emission and the subsequent appearance of photoproduct absorption, and find excellent agreement between the experimental observations and molecular dynamics calculations that involve a true electronic state crossing. Taken together, these findings constitute the most compelling evidence to date for the existence and importance of conical intersections in visual photochemistry.

Adenine deactivation in DNA resolved at the CASPT2//CASSCF/ AMBER level.

We have employed hybrid CASPT2//CASSCF/AMBER calculations to map the (1)L(a)(1pipi*) deactivation path of a single quantum mechanical adenine in a d(A)(10).d(T)(10) double strand in water that is treated at the molecular mechanics level. We find that (a) the L(a) relaxation route is flatter in DNA than in vacuo and (b) the L(a) relaxation energy in DNA is much larger than the stabilization energy of the corresponding L(a) excimer. An intra-monomer relaxation process is found to be compatible with the multiexponential decay recorded in DNA, possibly including the longer (4100 ps) lifetime component.

Significance of a zwitterionic state for fulgide photochromism: implications for the design of mimics.

Modeling and mimicry: an advanced computational model for the photocyclization of a furyl fulgide showed that a stable charge-transfer excited state, S(1), and the corresponding conical intersection with the ground state are responsible for the efficient photochromism observed in this system. This finding provides a rationale for the de novo design of related derivatives with similar (or even increased) efficiency.

Deciphering low energy deactivation channels in adenine.

The radiationless decay paths of 9H-adenine in its lowest excited states (1)npi*, (1)L(b)((1)pipi*), and (1)L(a)((1)pipi*) and in dissociative (1)pisigma* states have been mapped in vacuo at the CASPT2//CASSCF resolution. The minimum energy path (MEP) of the (1)L(a) state, which shows the strongest absorption below 5 eV, is found to decrease monotonically along the puckering coordinate from the vertical excitation to a S(0)/(1)L(a) conical intersection (CI). The vertically excited (1)npi* and (1)L(b) states are found to relax to the respective minima and to require some energy to reach CIs with S(0). This picture suggests that (1)L(a) alone is responsible of both components of the ultrafast biexponential decay (with tau(1) < 0.1 ps and tau(2) < 1 ps) recently observed in time-resolved pump-probe resonant ionization and fluorescence spectroscopy, and that the (1)npi* and (1)L(b) states do not act as important intermediates in the (1)L(a) decay process. We find that the (1)L(a)-->(1)pisigma(N9H)* internal conversion can be followed by N(9)-H photocleavage, albeit with tiny quantum yield. The amino N(10)-H bond photocleavage is hindered by the high barrier encountered along the N(10)-H bond-breaking path in the (1)pisigma(N10H)* state.

Revealing excited state interactions by quantum-chemical modeling of vibronic activities: the R2PI spectrum of adenine.

We present a computational study encompassing quantum-chemical calculations of the ground and low-lying excited states of 9H-adenine and modeling of vibronic activities associated with the S(0) --> L(b) and S(0) --> n pi* transitions. Minima on the ground and excited states and the saddle point on the n pi* potential energy surface are determined with CASSCF calculations. Vibrational frequencies are computed at the same level of theory on ground and excited states while transition dipole moments and oscillator strengths are estimated, at the optimized geometries, with CASPT2//CASSCF calculations. Modeling of vibronic activities includes both Franck-Condon and Herzberg-Teller induced contributions. While the adopted harmonic approximation is acceptable for the S(0) --> L(b) transition and allows the assignment of several observed bands in the R2PI spectrum of adenine, the computed anharmonic potential along the puckering coordinate in the n pi* state requires a different treatment. To this end the vibronic levels and intensities associated with vibronic transitions in the puckering coordinate are evaluated by numerical solution of the 2D potential including the anharmonic puckering coordinate. All the remaining vibrational coordinates are treated as harmonic. On the basis of the modeling, the four major bands in the R2PI spectrum of adenine are assigned, along with a number of minor bands in the spectra.

Deciphering intrinsic deactivation/isomerization routes in a phytochrome chromophore model.

High level ab initio correlated (CASPT2) computations have been used to elucidate the details of the photoinduced molecular motion and decay mechanisms of a realistic phytochrome chromophore model in vacuo and to explore the reasons underneath its photophysical/photochemical properties. Competitive deactivation routes emerge that unveil the primary photochemical event and the intrinsic photoisomerization ability of this system. The emerged in vacuo based static (i.e., nondynamical) reactivity model accounts for the formation of different excited state intermediates and suggests a qualitative rationale for the short (picosecond) excited state lifetime and ultrafast decay of the emission, its small quantum yield, and the multiexponential decay observed in both solvent and phytochromes. It is thus tentatively suggested that this is a more general deactivation scheme for photoexcited phytochrome chromophores that is independent of the surrounding environment. Spectroscopic properties have also been simulated in both isolated conditions and the protein that satisfactorily match experimental data. For this purpose, preliminary hybrid QM/MM computations at the correlated (CASPT2) level have been used in the protein and are reported here for the first time.

Resistive molecular memories: influence of molecular parameters on the electrical bistability.

The electrical bistability behavior of 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ) along with two additional benzoquinone derivatives (TCQ and TCN) and pentacene (PNT) is investigated by computing intra- and intermolecular charge transfer parameters and by comparing the efficiency of bulk charge transport and charge injection at the electrode/organic interface in the presence of neutral and charged molecular species. The bulk charge transport is modeled assuming a charge hopping regime and by computing hopping rates and mobilities. Molecular dynamics simulations are carried out to estimate the effect of thermal disorder on charge transfer integrals. The efficiency of the interface transport is estimated by comparing the electron affinities of benzoquinone derivatives and the ionization potential of pentacene with the work function of commonly employed electrodes. It is shown that the observed memory effect can be rationalized in terms of an interplay of the two transport mechanisms by showing that the OFF state is dominated by interface limited phenomena and the ON state may be determined also by bulk transport limited phenomena. While the contribution of collective effects cannot be ruled out for the macroscopic memory phenomenon, we show that, at a molecular level, sizable intramolecular reorganization energies are fundamental for the efficiency of the device, provided their magnitude does not hamper the charge transport across the device. It is suggested that control over molecular parameters might be exploited to design more efficient resistive molecular memories.

Electrostatic control of the photoisomerization efficiency and optical properties in visual pigments: on the role of counterion quenching.

Hybrid QM(CASPT2//CASSCF/6-31G*)/MM(Amber) computations have been used to map the photoisomerization path of the retinal chromophore in Rhodopsin and explore the reasons behind the photoactivity efficiency and spectral control in the visual pigments. It is shown that while the electrostatic environment plays a central role in properly tuning the optical properties of the chromophore, it is also critical in biasing the ultrafast photochemical event: it controls the slope of the photoisomerization channel as well as the accessibility of the S(1)/S(0) crossing space triggering the ultrafast decay. The roles of the E113 counterion, the E181 residue, and the other amino acids of the protein pocket are explicitly analyzed: it appears that counterion quenching by the protein environment plays a key role in setting up the chromophore's optical properties and its photochemical efficiency. A unified scenario is presented that discloses the relationship between spectroscopic and mechanistic properties in rhodopsins and allows us to draw a solid mechanism for spectral tuning in color vision pigments: a tunable counterion shielding appears as the elective mechanism for L<-->M spectral modulation, while a retinal conformational control must dictate S absorption. Finally, it is suggested that this model may contribute to shed new light into mutations-related vision deficiencies that opens innovative perspectives for experimental biomolecular investigations in this field.

Multistate photo-induced relaxation and photoisomerization ability of fumaramide threads: a computational and experimental study.

Fumaric and maleic amides are the photoactive units of an important and widely investigated class of photocontrollable rotaxanes as they trigger ring shuttling via a cis-trans photoisomerization. Here, ultrafast decay and photoinduced isomerization in isolated fumaramide and solvated nitrogen-substituted fumaramides (that are employed as threads in those rotaxanes) have been investigated by means of CASPT2//CASSCF computational and time-resolved spectroscopic techniques, respectively. A complex multistate network of competitive deactivation channels, involving both internal conversion and intersystem crossing (ISC) processes, has been detected and characterized that accounts for the picosecond decay and photochemical/photophysical properties observed in the singlet as well as triplet (photosensitized) photochemistry of fumaramides threads. Interestingly, singlet photochemistry appears to follow a non-Kasha rule model, where nonequilibrium dynamical factors control the outcome of the photochemical process: accessible high energy portions of extended crossing seams turn out to drive the deactivation process and ground-state recovery. Concurrently, extended singlet/triplet degenerate regions of twisted molecular structures with significant spin-orbit-coupling values account for ultrafast (picosecond time scale) ISC processes that lead to higher photoisomerization efficiencies. This model discloses the principles behind the intrinsic photochemical reactivity of fumaramide and its control.

The different photoisomerization efficiency of azobenzene in the lowest n pi* and pi pi* singlets: the role of a phantom state.

Azobenzene E<==>Z photoisomerization, following excitation to the bright S(pi pi*) state, is investigated by means of ab initio CASSCF optimizations and perturbative CASPT2 corrections. Specifically, by elucidating the S(pi pi*) deactivation paths, we explain the mechanism responsible for azobenzene photoisomerization, the lower isomerization quantum yields observed for the S(pi pi*) excitation than for the S1(n pi*) excitation in the isolated molecule, and the recovery of the Kasha rule observed in sterically hindered azobenzenes. We find that a doubly excited state is a photoreaction intermediate that plays a very important role in the decay of the bright S(pi pi*). We show that this doubly excited state, which is immediately populated by molecules excited to S(pi pi*), drives the photoisomerization along the torsion path and also induces a fast internal conversion to the S1(n pi*) at a variety of geometries, thus shaping (all the most important features of) the S(pi pi*) decay pathway and photoreactivity. We reach this conclusion by determining the critical structures, the minimum energy paths originating on the bright S(pi pi*) state and on other relevant excited states including S1(n pi*), and by characterizing the conical intersection seams that are important in deciding the photochemical outcome. The model is consistent with the most recent time-resolved spectroscopic and photochemical data.

Cyclohexenylphenyldiazene: a simple surrogate of the azobenzene photochromic unit.

We have carried out an experimental and computational study on the ground- and excited-state photochemical and photophysical properties of (1-cyclohexenyl)phenyldiazene (CPD), a species formally derived from azobenzene in which one of the phenyl rings is replaced by a 1-cyclohexene substituent. The results show that CPD does substantially behave like azobenzene, but with a higher (approximately 70%) Phi(Z-->E) (npi*) photoisomerization quantum yield, calling for CPD as an effective alternative of azobenzene itself with new functionalization possibilities. By use of state-of-the-art ab initio CASPT2//CASSCF minimum energy path computations, we have identified the most efficient decay and isomerization routes of the absorbing singlet (pipi*), S1 (npi*), T1, and S0 states of CPD. The resulting mechanistic scheme agrees with experimental findings and provides a rationale of the observed photoisomerization quantum yields. Furthermore, this study provides a deep insight on the photophysical and photochemical properties of compounds based on the -N=N- double bond which supplies a general model for the photoreactivity of azobenzene-type compounds in general. This is expected to be a useful guideline for the design of novel photoreactive azo compounds.

Effect of strain on the photoisomerization and stability of a congested azobenzenophane: a combined experimental and computational study.

The stability and trans-cis photoisomerization properties of a macrocycle constituted of two para-aminoazobenzene units connected by two methylene bridges have been investigated by a combination of experimental and computational techniques. Irradiation at 365 nm leads to a photostationary state in which only 50% of the azobenzene units have isomerized, in contrast with the behavior of para-aminoazobenzene, whose photoconversion is larger than 80%. In the case of the macrocycle, a faster cis --> trans thermal back-reaction is observed. To assist the interpretation of the experimental results, molecular mechanics and quantum chemical calculations have been carried out. Of the possible conformers, the most stable trans-trans geometric isomer has been identified along with the more plausible trans-cis and cis-cis isomers. Ground-state energy barriers along the NN torsional coordinates were also computed, along with excitation energies and intensities for the species that can contribute to the photostationary state. The calculations point to a sequential photoisomerization mechanism and support a predominance of the trans-cis photoproduct with minor contributions from the cis-cis species. The thermal and photochemical reactivity of the examined macrocycle is compared to that of previously investigated azobenzenophanes and explained in terms of strain and substituent effects both concurring to favor the thermal cis --> trans back-reaction.

Charge-transport regime of crystalline organic semiconductors: diffusion limited by thermal off-diagonal electronic disorder.

We propose that the electron transport in crystalline organic semiconductors at room temperature (RT) is neither polaronic nor a combination of thermally activated hopping and polaronic transport, as previously thought. Thermal molecular motions cause large fluctuations in the intermolecular transfer integrals that, in turn, localize the charge carrier. This effect destroys the translational symmetry of the electronic Hamiltonian and makes the band description inadequate for RT organic crystals. We used a one-dimensional semiclassical model to compute the (temperature dependent) charge carrier mobility in the presence of thermal fluctuations of the electronic Hamiltonian. This transport mechanism explains several contrasting experimental observations pointing sometimes to a delocalized "bandlike" transport and sometimes to the existence of strongly localized charge carriers.

Dynamics of the intermolecular transfer integral in crystalline organic semiconductors.

In organic crystalline semiconductor molecular components are held together by very weak interactions and the transfer integrals between neighboring molecular orbitals are extremely sensitive to small nuclear displacements. We used a mixed quantum chemical and molecular dynamic methodology to assess the effect of nuclear dynamics on the modulation of the transfer integrals between close molecules. We have found that the fluctuations of the transfer integrals are of the same order of magnitude of their average value for pentacene and anthracene. Under these conditions the usual perturbative treatment of the electron-phonon coupling is invalid, the band description of the crystal breaks down and the charge carriers become localized. Organic crystals of pentacene and anthracene, even in the absence of defects, can be regarded as disordered media with respect to their charge transport properties. These results suggest that the dynamic electronic disorder can be the factor limiting the charge mobility in crystalline organic semiconductors.