Adiabatic nuclear potential energy surfaces (PESs) defined via the Born-Oppenheimer (BO) approximation are a fundamental concept underlying chemical reactivity theory. For a wide range of excited-state phenomena such as radiationless decay, energy and charge transfer, and photochemical reactions, the BO approximation breaks down due to strong couplings between two or more BO PESs. Non-adiabatic molecular dynamics (NAMD) is the method of choice to model these processes. We review new developments in quantum-classical dynamics, analytical derivative methods, and time-dependent density functional theory (TDDFT) which have lead to a dramatic expansion of the scope of ab initio NAMD simulations for molecular systems in recent years. We focus on atom-centered Gaussian basis sets allowing highly efficient simulations for molecules and clusters, especially in conjunction with hybrid density functionals. Using analytical derivative techniques, forces and derivative couplings can be obtained with machine precision in a given basis set, which is crucial for accurate and stable dynamics. We illustrate the performance of surface-hopping TDDFT for photochemical reactions of the lowest singlet excited states of cyclohexadiene, several vitamin D derivatives, and a bicyclic cyclobutene. With few exceptions, the calculated quantum yields and excited state lifetimes agree qualitatively with experiment. For systems with ∼50 atoms, the present Turbomole implementation allows NAMD simulations with 0.2-0.4 ns total simulation time using hybrid density functionals and polarized double zeta valence basis sets on medium-size compute clusters. We close by discussing open problems and future directions.
Affinities of alkali cations and halide anions for the peptide group were quantified using molecular dynamics simulations of aqueous solutions of N-methylacetamide using both nonpolarizable and polarizable force fields. Potassium and, more strongly, sodium exhibit an affinity for the carbonyl oxygen of the amide group, while none of the halide anions shows any appreciable attraction for the amide hydrogen. Heavier halides, however, interact with the hydrophobic methyl groups of N-methylacetamide. Using the present results for a model of the peptide bond we predict that the destabilizing effect of weakly hydrated Hofmeister ions, such as bromide or iodide, is not due to direct interactions with the backbone but rather due to attraction to hydrophobic regions of the protein.
The effect of non-adiabatic transitions on branching ratios, kinetic and internal energy distribution of fragments, and reaction mechanisms observed in acetadehyde photodissociation is investigated by non-adiabatic molecular dynamics (NAMD) simulations using time-dependent hybrid density functional theory and Tully surface hopping. Homolytic bond breaking is approximately captured by allowing spin symmetry to break. The NAMD simulations reveal that non-adiabatic transitions selectively enhance the kinetic energy of certain internal degrees of freedom within approximately 50 fs. Branching ratios from NAMD and conventional ``hot'' Born-Oppenheimer molecular dynamics (BOMD) are similar and qualitatively agree with experiment. However, as opposed to the BOMD simulations, NAMD captures the high-energy tail of the experimental kinetic energy distribution. The extra "kick'' of the nuclei in the direction of the non-adiabatic coupling vector results from non-adiabatic transitions close to conical intersections. From a mechanistic perspective, the non-adiabatic effects favor asynchronous over synchronous fragmentation and tend to suppress roaming.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.