First Principles Molecular Dynamics for Reactive Systems

You are reading version 1 of this direction, frozen 2026-10-04. The latest version is version 2.

Version 1 of 225 papers26 conceptsfrozen 2026-10-04

In traditional molecular simulations, atoms are often treated like billiard balls connected by springs. These "springs"—known as empirical force fields—are mathematical shortcuts that describe how atoms push and pull on each other. While efficient, these shortcuts have a fatal flaw: they cannot easily describe what happens when a spring breaks or a new one forms. They struggle to model chemistry itself.

To solve this, researchers have turned to Ab Initio Molecular Dynamics (AIMD), also known as first-principles molecular dynamics. Instead of relying on pre-defined rules for atomic interactions, AIMD calculates the forces acting on atoms "on the fly" by solving quantum mechanical equations at every single step of the simulation.

Beyond Balls and Springs

The core intuition behind AIMD, as explored by Michele Parrinello and Michiel Sprik in Readive liquids from first principles, is the integration of Density Functional Theory (DFT) with thermal motion.

In a standard simulation, if you want to know how a water molecule behaves, you give the computer a fixed formula for the bond between Oxygen and Hydrogen. But in a real chemical reaction, that bond might stretch, weaken, and eventually snap. AIMD handles this by looking at the electrons. By calculating the electronic ground state for a specific arrangement of nuclei, the simulation determines the exact quantum mechanical force exerted on every atom. The atoms move a tiny fraction of a femtosecond, the electronic cloud reshapes itself to the new positions, and the process repeats.

This allows researchers to observe reactive processes—like the dehydroxylation of minerals studied by Esther Molina‐Montes and Alfonso Hernández‐Laguna—where bonds are constantly breaking and forming Water Release from Pyrophyllite during the Dehydroxylation Process Explored by Quantum Mechanical Simulations.

Simulating the Unreachable

One of the most powerful applications of this "first principles" approach is in environments where experiments are impossible to conduct. When matter is squeezed to millions of atmospheres, the very nature of chemical bonding changes.

G. Chiarotti and Marco Bernasconi used these techniques to peer inside giant planets like Neptune, uncovering how simple molecules polymerize under extreme pressure Simple Molecular Systems at Very High Pressures: Computer simulation studies.. Similarly, Weijie Li and Zi Li have applied AIMD to model the Earth's liquid iron-nickel core, calculating viscosity and diffusion coefficients that help us understand the geodynamo Ab initio determination on diffusion coefficient and viscosity of FeNi fluid under Earth’s core condition.

Because these simulations don't rely on empirical data (which doesn't exist for the center of the Earth), they provide a "ground truth" based purely on the laws of physics. This is equally critical in fusion research, where Mohan Chen and Xinguo Ren use AIMD to track how deuterium isotopes diffuse through liquid metal walls, a process that determines the lifespan of a nuclear reactor First-principles molecular dynamics study of deuterium diffusion in liquid tin.

Capturing Complexity and Chaos

Beyond extreme pressures, AIMD excels at finding "defects" in complex networks that empirical models often smooth over. In liquid water, for instance, the standard view is a neat tetrahedral network. However, Piero Gasparotto and Ali Hassanali used ab initio methods to show that water is full of highly correlated "weak bond" defects that drive its unique properties Probing Defects and Correlations in the Hydrogen-Bond Network of ab Initio Water.

This ability to capture non-obvious behavior extends to mass spectrometry. Shunyang Wang and Dean J. Tantillo demonstrated that by using Born-Oppenheimer molecular dynamics—a flavor of AIMD—they could predict how molecules fragment after being hit by high-energy electrons. Unlike rule-based software, this first-principles approach can discover complex molecular rearrangements that no human had programmed into the system Predicting in-silico electron ionization mass spectra using quantum chemistry.

The Thermodynamic Bridge

Finally, AIMD serves as a bridge between the microscopic and the macroscopic. By combining these simulations with "enhanced sampling" techniques like metadynamics, researchers like Vassiliki‐Alexandra Glezakou and Roger Rousseau can map out entire free-energy landscapes.

This allows them to explain why $CO_2$ reacts differently in water versus supercritical fluids How Collective Phenomena Impact CO2 Reactivity and Speciation in Different Media, or how the confinement of water inside a zeolite pore fundamentally alters its acidity Confinement Effects and Acid Strength in Zeolites. In these cases, the simulation doesn't just show where the atoms are; it explains the thermodynamic "why" behind the chemistry.

Papers in this version

History & citationVersion 2 · 2026-10-04

Current · Permanent link

  1. Version 2 (latest) · 2026-10-04 · 18 papers
  2. Version 1 · 2026-10-04 · 25 papers
Cite this direction

Latest version (a living page):

@misc{lacuna_ab_initio_molecular_dynamics_ec8f783b,
  title = {ab initio molecular dynamics},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/ab-initio-molecular-dynamics/rdr_ec8f783b23fb40e6a70ede9ecc618596},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-20},
}

This exact version (version 1, frozen text):

@misc{lacuna_first_principles_molecular_dynamics_for__ec8f783b_v1,
  title = {First Principles Molecular Dynamics for Reactive Systems},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/first-principles-molecular-dynamics-for-reactive-systems/rdr_ec8f783b23fb40e6a70ede9ecc618596/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-20},
}