Elucidating the Impact of Molecular Motors on Their Solvation Environment

Abstract

In response to external stimuli, molecular motors enable to control phenomena at the molecular scale with high precision. In order to utilize their unique properties and to gain designated functionalities, their molecular embedding is important. Despite the great progress in the development of corresponding functional materials, a detailed picture of how the structural and dynamic properties of these responsive molecular units are transferred to a macroscopic outcome is so-far missing. Here, we provide an atomistic insight into the solvation dynamics around a light-driven molecular motor. By performing molecular dynamic simulations based on an ab initio parametrized and validated force field, we elucidate in detail the intermolecular interactions depending on the state of the motor. Detailed analysis of the solvation shells revealed the impact on both the location of the primary interaction sites and the orientation of the solvent molecules with respect to the molecular motor. Furthermore, we studied the influence of structural modifications of the molecular motor on its local environment. By investigating the motor−solvent interaction, our results provide a strong foundation to decipher the ability of molecular machines to specifically alter molecular processes, which is fundamental to predict and tailor the resulting macroscopic functionality.

On-the-Fly Training of Atomistic Potentials for Flexible and Mechanically Interlocked Molecules

“…Over the last years, versatile algorithms have been utilized to develop FFs describing different molecular classes. − One popular technique is to employ genetic algorithms. − Such parametrization schemes are based on the minimization of an objective function, which includes deviations between FF and reference data. Frequently, equilibrium quantities including optimized structures and/or Hessian, normal modes (NMs), and eigenvalues (EVs) are used as reference data. − Corresponding QM data provide a good representation of local phase space properties and can be used as a training set (see Figure ). However, since often only information on one local minimum configuration is considered, predictions of properties out-of-the equilibrium or fluctuations between different minima can be error-prone.…”

“…The FF-calculated EVs are sorted according to the highest overlap of the corresponding NMs. , The subspace overlap parameter S i subspace in the fifth term is defined by , with the eigenvectors ( of the reference system and v⃗ j of the FF optimized system). This value S i subspace is a projection of the reference NMs on the FF calculated NMs and measures if the eigenmodes of the FF calculated system cover the same subspace of motions as the reference system. , All terms are normalized by the number of involved terms ( N bonds , N angles , N tors , N EV/NM ) and weighted by a constant ω, with ω ≥ 0. The number of considered NMs and EVs is calculated on the basis of the degrees of freedom of the N atoms of the molecules by 3 N – 6.…”

“…The FFcalculated EVs are sorted according to the highest overlap of the corresponding NMs. 42,58 The subspace overlap parameter S i s u b s p a c e i n t h e fi f t h t e r m i s d e fi n e d b y…”

“…This value S i subspace is a projection of the reference NMs on the FF calculated NMs and measures if the eigenmodes of the FF calculated system cover the same subspace of motions as the reference system. 42,58 All terms are normalized by the number of involved terms (N bonds , N angles , N tors , N EV/NM ) and weighted by a The MD simulations are conducted in gas phase using the Nose−Hoover thermostat. 62,63 The convergence criterium for the geometry optimizations at the FF level was 0.01 kcal/mol/Å.…”

On the Molecular Mechanism of a Photo‐Responsive Phase Change Memory

“…Therefore, a psGA is introduced for the parameterization of the force field. Based on the previously reported genetic algorithm schema, [37][38][39] this approach aims to improve the screening of the parameter space by utilizing local properties of the phase space with the non-local exchange of data. The structural and dynamic properties of both isomers, calculated at the quantum mechanical (QM) level, were thus included as reference data.…”