Molecular Simulation

Research notes on molecular simulation methods, covering classical molecular dynamics, interatomic potentials, 3D conformer generation, and neural network potentials.

Classical Potentials and Simulation Methods

Year	Paper	Key Idea
1932	Lennard-Jones on Adsorption and Diffusion on Surfaces	Potential energy surface models unifying physical and chemical adsorption
1964	Correlations in the Motion of Atoms in Liquid Argon	First large-scale MD simulation of liquid argon with LJ potential
1977	Umbrella Sampling: Monte Carlo Free-Energy Estimation	Importance sampling technique for Monte Carlo free-energy calculations
1979	The Muller-Brown Potential: A 2D Benchmark Surface	Classic 2D benchmark for optimization and MD testing
1984	Embedded-Atom Method: Impurities and Defects in Metals	Foundational many-body potential for metal simulations
1985	Stillinger-Weber Potential for Silicon Simulation	3-body interaction model for tetrahedral semiconductors
1986	Evans 1986: Thermal Conductivity of Lennard-Jones Fluid	NEMD validation for heat flow, identifying long-time tail anomalies
1987	Second-Order Langevin Equation for Field Simulations	Reduces systematic errors to O(e2) in lattice field simulations
1989	Dynamical Corrections to TST for Surface Diffusion	Dynamical corrections revealing bounce-back recrossings at low T
1993	Embedded-Atom Method: Theory and Applications Review	Comprehensive review of EAM theory and parameterization
1994	Embedded-Atom Method User Guide: Voter’s 1994 Chapter	Practical guide for EAM potential fitting and applications

Surface Catalysis and Diffusion

Year	Paper	Key Idea
1984	Adatom Dimer Diffusion on fcc(111) Crystal Surfaces	Identifying simultaneous multiple jumps in adatom dimer diffusion
1985	Kinetic Oscillations in CO Oxidation on Pt(100): Theory	Coupled ODEs explaining CO oxidation via surface phase transitions
1992	Oscillatory CO Oxidation on Pt(110): Temporal Modeling	Kinetic model for temporal self-organization in catalytic oxidation
1994	In Situ XRD of Oxidation-Reduction Oscillations on Pt/SiO2	XRD validation of oxide model for kinetic rate oscillations
1994	MD Simulation of Self-Diffusion on Metal Surfaces (1994)	Iridium surface diffusion confirming atomic exchange mechanisms

Neural Network Potentials and Modern Methods

Year	Paper	Key Idea
2020	MAT: Graph-Augmented Transformer for Molecules (2020)	Transformer with inter-atomic distances and graph adjacency
2023	Ewald Message Passing for Molecular Graphs	Fourier-space long-range interactions improve GNN energy predictions
2025	Beyond Atoms: 3D Space Modeling for Molecular Pretraining	SpaceFormer models entire 3D molecular space for representations
2025	Dark Side of Forces: Non-Conservative ML Force Models	Critique of non-conservative forces in ML potentials
2025	DenoiseVAE: Adaptive Noise for Molecular Pre-training	Adaptive atom-specific noise distributions for better force fields
2025	Efficient DFT Hamiltonian Prediction via Adaptive Sparsity	SPHNet achieves up to 7x speedup in Hamiltonian prediction
2025	eSEN: Smooth Interatomic Potentials (ICML Spotlight)	Energy conservation as key MLIP diagnostic, introducing eSEN
2025	MOFFlow: Flow Matching for MOF Structure Prediction	Riemannian flow matching for Metal-Organic Framework generation
2025	PharMolixFM: Multi-Modal All-Atom Molecular Models	Unified diffusion, flow matching, and BFN for molecular modeling

Computational Chemistry

Diagram showing the Ewald decomposition of long-range interactions into short-range and Fourier-space components for molecular graph neural networks

Ewald Message Passing for Molecular Graphs

Proposes Ewald message passing, a Fourier-space scheme inspired by Ewald summation that captures long-range interactions in molecular graphs. The method is architecture-agnostic and improves energy MAEs by 10% on OC20 and 16% on OE62 across four baseline GNN models.

Computational Chemistry

Schematic overview of three multi-modal generative model variants for all-atom molecular denoising

PharMolixFM: Multi-Modal All-Atom Molecular Models

PharMolixFM proposes a unified framework for all-atom foundation models using three multi-modal generative approaches (diffusion, flow matching, BFN) and demonstrates competitive docking accuracy with fast inference.

Computational Chemistry

Bar chart comparing MAT average ROC-AUC against D-MPNN, GCN, and Weave baselines

MAT: Graph-Augmented Transformer for Molecules (2020)

Molecule Attention Transformer (MAT) augments Transformer self-attention with inter-atomic distances and graph adjacency, achieving strong property prediction across diverse molecular tasks with minimal hyperparameter tuning after self-supervised pretraining.

Computational Chemistry

MOFFlow assembles metal nodes and organic linkers into Metal-Organic Framework structures

MOFFlow: Flow Matching for MOF Structure Prediction

MOFFlow is the first deep generative model tailored for Metal-Organic Framework (MOF) structure prediction. It utilizes Riemannian flow matching on SE(3) to assemble rigid building blocks (metal nodes and organic linkers), achieving higher accuracy and scalability than atom-based methods on large systems.

Computational Chemistry

Graph of the Lennard-Jones 12-6 potential showing the characteristic attractive and repulsive forces

Dynamical Corrections to TST for Surface Diffusion

This paper bridges Molecular Dynamics and Transition State Theory by applying a dynamical corrections formalism to surface diffusion, identifying a low-temperature bounce-back mechanism causing non-Arrhenius behavior.

Computational Chemistry

Embedding energy and effective charge functions for Ni and Pd from the original EAM paper

Embedded-Atom Method User Guide: Voter's 1994 Chapter

This 1994 handbook chapter serves as a practical user guide for the Embedded-Atom Method (EAM). It details the theoretical derivation from density-functional theory, synthesizes related methods like the Glue Model, and provides a complete tutorial on fitting potentials, illustrated with a specific implementation for the Ni-Al-B system.

Computational Chemistry

Embedded-Atom Method: Theory and Applications Review

This 1993 review systematizes the Embedded-Atom Method (EAM) as a practical semi-empirical approach for metallic systems. It synthesizes theory, applications, and connections to related methods while addressing the limitations of pair potentials.

Computational Chemistry

Evans 1986: Thermal Conductivity of Lennard-Jones Fluid

This paper validates the homogeneous Evans method for calculating thermal conductivity against experimental Argon data. It demonstrates broad agreement across the phase diagram but identifies significant non-monotonic behavior and enhanced long-time tails near the critical point.

Computational Chemistry

Carbon monoxide molecule adsorbed on Pt(100) FCC surface in hollow site configuration

In Situ XRD of Oxidation-Reduction Oscillations on Pt/SiO2

This study provides the first direct experimental proof that rate oscillations in catalytic CO oxidation on supported Pt are driven by a periodic oxidation and reduction of the catalyst surface. By monitoring Bragg peak intensities in situ, the authors confirm the ‘oxide model’ over competing reconstruction or carbon models.

Computational Chemistry

Kinetic Oscillations in CO Oxidation on Pt(100): Theory

Imbihl et al. establish the first detailed microscopic model for CO oxidation oscillations on Pt(100), identifying the adsorbate-induced hex to 1x1 phase transition as the driving force. The study combines linear stability analysis with numerical reaction-diffusion simulations.

Computational Chemistry

MD Simulation of Self-Diffusion on Metal Surfaces (1994)

A molecular dynamics investigation using EAM and many-body potentials to elucidate atomic exchange mechanisms on Iridium surfaces, verifying Field Ion Microscope observations.

Computational Chemistry

Replication of Figure 7 showing stable oscillations in CO oxidation on Pt(110)

Oscillatory CO Oxidation on Pt(110): Temporal Modeling

This paper presents a 4-variable kinetic model coupling surface reaction dynamics with structural phase transitions to reproduce complex oscillatory behavior on Pt(110).

Classical Potentials and Simulation Methods#

Surface Catalysis and Diffusion#

Neural Network Potentials and Modern Methods#

Classical Potentials and Simulation Methods

Surface Catalysis and Diffusion

Neural Network Potentials and Modern Methods