The Non‐Ergodic Nature of Internal Conversion

Abstract

The absorption of light by molecules can induce ultrafast dynamics and coupling of electronic and nuclear vibrational motion. The ultrafast nature in many cases rests on the importance of several potential energy surfaces in guiding the nuclear motion-a concept of central importance in many aspects of chemical reaction dynamics. This Minireview focuses on the non-ergodic nature of internal conversion, that is, on the concept that the nuclear dynamics only sample a reduced phase space, potentially resulting in localization of the dynamics in real space. A series of results that highlight the nonstatistical nature of the excited-state deactivation process is presented. The examples are categorized into four groups. 1) Localization of the energy in one degree of freedom in S2 →S1 transitions, in which the transition is either determined by the time spent in the S2 →S1 coupling region or by the time it takes to reach it. 2) Localization of energy into a single reactive mode, which is dictated by the internal conversion process. 3) Initiation of the internal conversion by activation of a single complex motion, which then specifically couples to a reactive mode. 4) Nonstatistical internal conversion as a tool to accomplish biomolecular stability. Herein, the discussion on nonstatistical internal conversion in DNA as a mechanism to eliminate electronic excitation energy is extended to include molecules with an S-S bond as a model of the disulfide bridge in peptides. All of these examples are summed up in Kasha's rule. For systems with multiple degrees of freedom it will be possible to locate an appropriate motion somewhere in phase space that will take the wavepacket to the coupling region and facilitate an ultrafast transition to S1. Once at S1, the momentum of the wavepacket is lost and the only options left are the statistical processes of reaction or light emission.

Distortion dependent intersystem crossing: A femtosecond time-resolved photoelectron spectroscopy study of benzene, toluene, and p-xylene

“…The observation of an S 2 → S 1 transition faster than 150 fs implies a highly non-statistical process mainly activating a few reaction coordinates largely determined by Franck-Condon factors and the topology of the potential energy surface near the Franck-Condon region 44 . The current and previous 64 calculations predict the S 2 equilibrium structure to be boat-shaped and distorted out-of-plane relative to the Franck-Condon geometry (Figure 3), and thus early activation of prefulvene-like or boat-shaped modes is expected 66 .…”

“…However, in comparison to IC, the importance and behavior of ISC in the ultrafast time regime are much less established. For IC, it has been shown that the ultrafast nature of the process rests on large degrees of non-ergodicity; 44,45 in other words, only a few vibrational modes are active during the transition. If the vibrations that are activated upon photon absorption couple different electronic surfaces, the possibility of an ultrafast transition is available and can occur within the timescale of the respective vibrational period.…”

Conserving Coherence and Storing Energy during Internal Conversion: Photoinduced Dynamics of cis- and trans-Azobenzene Radical Cations

“…Femtosecond time-resolved mass spectrometry (FTRMS) has brought about many discoveries pertaining to the electronic structure and initiated dynamics of polyatomic molecules, as well as to the understanding of (non)-ergodicity in chemical reactions. [1][2][3][4][5][6][7] Recently, this technique has been increasingly applied to studying the post-ionization dynamics of radical cations. [8][9][10][11][12][13][14][15][16][17] A major challenge in these experiments stems from the fact that ionization via multiphoton absorption typically produces vibrationally hot ions.…”

Hexamethylcyclopentadiene: time-resolved photoelectron spectroscopy and ab initio multiple spawning simulations

“…This is especially important because a profound understanding of photoinduced dynamics almost always necessitates a detailed comparison of experiment with theory. Many model systems have been investigated by time-resolved experiments [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19] or dynamical simulations. [20][21][22][23][24][25][26] We believe that fundamental questions could be answered more easily by joint studies connecting experimental approaches with such simulations.…”