Ultrafast photo-induced processes in complex environments: The role of accuracy in excited-state energy potentials and initial conditions

Light induces non-equilibrium time evolving molecular phenomena. The computational modeling of photo-induced processes in large systems, embedded in complex environments (i.e., solutions, proteins, materials), demands for a quantum and statistical mechanic treatment to achieve the required accuracy in the description of both the excited-state energy potentials and the choice of the initial conditions for dynamical simulations. On the other hand, the theoretical investigation on the atomistic scale of times and sizes of the ultrafast photo-induced reactivity and non-equilibrium relaxation dynamics right upon excitation requests tailored computational protocols. These methods often exploit hierarchic computation schemes, where a large part of the degrees of freedom are required to be treated explicitly to achieve the right accuracy. Additionally, part of the explicit system needs to be treated at ab initio level, where density functional theory, using hybrid functionals, represents a good compromise between accuracy and computational cost, when proton transfers, non-covalent interactions, and hydrogen bond dynamics play important roles. Thus, the modeling strategies presented in this review stress the importance of hierarchical quantum/molecular mechanics with effective non-periodic boundary conditions and efficient phase-sampling schemes to achieve chemical accuracy in ultrafast time-resolved spectroscopy and photo-induced phenomena. These approaches can allow explicit and accurate treatment of molecule/environment interactions, including also the electrostatic and dispersion forces of the bulk. At the same time, the specificities of the different case studies of photo-induced phenomena in solutions and biological environments are highlighted and discussed, with special attention to the computational and modeling challenges.