Second generation Car–Parrinello molecular dynamics

Abstract

Computer simulation methods, such as Monte Carlo or Molecular Dynamics, are very powerful computational techniques that provide detailed and essentially exact information on classical many-body problems. With the advent of ab-initio molecular dynamics, where the forces are computed on-the-fly by accurate electronic structure calculations, the scope of either method has been greatly extended. This new approach, which unifies Newton's and Schrödinger's equations, allows for complex simulations without relying on any adjustable parameter. This review is intended to outline the basic principles as well as a survey of the field. Beginning with the derivation of BornOppenheimer molecular dynamics, the Car-Parrinello method and the recently devised efficient and accurate Car-Parrinello-like approach to Born-Oppenheimer molecular dynamics, which unifies best of both schemes are discussed. The predictive power of this novel second-generation Car-Parrinello approach is demonstrated by a series of applications ranging from liquid metals, to semiconductors and water. This development allows for ab-initio molecular dynamics simulations on much larger length and time scales than previously thought feasible.

Recent Advances in Metallic Glass Nanostructures: Synthesis Strategies and Electrocatalytic Applications

“…Although very powerful, quantum calculations for the most part are still limited to the behavior of small clusters of few interacting atoms, and thus the extrapolations of those results to macro-scale experiments over the usual time domain of the experimental realm are difficult. [146,152] This feature allowed researchers to study medium-range interactions and to build systems closer to the experimental ones, seen by many as computational experiments. Another powerful tool on computational chemistry is classical Molecular Dynamics (MD), which aims to solve trajectories in a many-body problem by integrating the classical equations of motion for the interacting particles using effective potentials that consider the quantum nature of molecular interactions in an implicit manner, and therefore can be used to simulate larger systems with up to millions of atoms.…”

“…[146,151] Medium-range and long-range interactions on a solid interface plus complex nonuniform surfaces in 3D materials also play a role on the final measurable property which is difficult, if not unfeasible, to be addressed by those tools. [152] Computational simulations have been used in the past to predict the glass forming-ability of alloys by means of computational experiments (virtual fast-quenching) [153] and more recently to study the nanometer size structure of metallic glasses, investigating their correlation with the material properties, revealing a more complex system than purely random due to short range ordering (SRO). [146,152] This feature allowed researchers to study medium-range interactions and to build systems closer to the experimental ones, seen by many as computational experiments.…”

Inverse simulated annealing: Improvements and application to amorphous InSb

“…However, due to the large number of degrees of freedom of disordered systems, the annealing has to be conducted as slowly as possible and is therefore computationally exceedingly expensive. This is even more pronounced in conjunction with ab initio electronic structure calculations, in spite of recent progress 15,16 . As a consequence, the attainable quench-rates are typically several orders of magnitude faster than in experiment.…”

On the role of interfacial hydrogen bonds in “on-water” catalysis

“…As outlined in Appendix A, this method unifies the efficiency of the CPMD approach with the large integration time steps of Born-Oppenheimer molecular dynamics (BOMD). 75,76 At variance to the recently proposed SCC-DFTB based BOMD scheme of Karplus et al, 77 not even a single diagonalization step, but just an electronic force calculation is required, similar to the SCC-DFTB based CPMD technique of Seifert and coworkers. 78 However, contrary to the latter, integration time steps up to the ionic resonance limit can be taken that are typically one order of magnitude larger.…”

“…90 The second-generation CPMD method combines the best of both schemes by retaining the large integration time steps of BOMD and, at the same time, preserving the efficiency of CPMD. 72,76,87 To that extent, the original fictitious Newtonian dynamics of CPMD is substituted by an improved coupled electron-ion dynamics that keeps the electrons very close to the instantaneous electronic ground state and does not require an additional fictitious mass parameter. The superior efficiency of this new approach, which, depending on the system, varies between one to two orders of magnitude, is due to the fact that not only the SCF cycle, but also the iterative wave function optimization is fully bypassed.…”