Skip to content

Geometry Optimization

Geometry Optimization is one of the most commonly performed calculations in computational chemistry. Its objective is to determine the most stable molecular structure by adjusting the positions of the atoms until the total electronic energy reaches a minimum.

Experimental molecular structures obtained from crystallographic databases or molecular builders are often not located exactly at the minimum of the potential energy surface. Small deviations in bond lengths, bond angles, or dihedral angles may remain due to experimental uncertainty or because the structure was manually constructed. Before calculating molecular properties such as vibrational frequencies, NMR chemical shifts, or reaction energies, it is therefore essential to optimize the molecular geometry.

During an optimization, Gaussian repeatedly performs a series of Single Point Energy calculations while systematically modifying the atomic coordinates. After each energy evaluation, the forces acting on the atoms are calculated, and the molecular geometry is updated in the direction that lowers the total energy. This iterative process continues until predefined convergence criteria for the energy, atomic forces, and atomic displacements are simultaneously satisfied.

Unlike a Single Point Energy calculation, where the atomic coordinates remain fixed throughout the calculation, a Geometry Optimization allows the molecular structure to change. The final optimized geometry corresponds to a stationary point on the potential energy surface, where the first derivative of the energy with respect to every nuclear coordinate becomes approximately zero.

For most molecular studies, a Geometry Optimization is the first computational step. Once an optimized structure has been obtained, it can be used for subsequent calculations such as Frequency Analysis, Intrinsic Reaction Coordinate (IRC) calculations, NMR shielding constants, excited-state calculations, and many other computational investigations.

Learning Objectives

After completing this section, you will be able to

  • Understand the purpose of a Geometry Optimization calculation.
  • Recognize the differences between a Single Point Energy calculation and a Geometry Optimization.
  • Prepare a Gaussian Geometry Optimization input file.
  • Understand the purpose of the Opt keyword.
  • Interpret the important sections of a Geometry Optimization output file.
  • Understand how Gaussian searches for an energy minimum.
  • Learn the mathematical principles behind geometry optimization algorithms.
  • Verify whether an optimization has successfully converged.
  • Use the optimized molecular structure for subsequent calculations such as Frequency Analysis and NMR calculations.

Topics Covered

This section is organized into the following chapters:

  1. Introduction to Geometry Optimization
  2. Geometry Optimization Input File
  3. Route Section and Keywords
  4. Explanation of the Input File
  5. Understanding the Output File
  6. How Geometry Optimization Works
  7. Mathematics Behind Geometry Optimization
  8. Complete Geometry Optimization Algorithm

Prerequisites

Before beginning this section, you should be familiar with

  • Molecular geometries and Cartesian coordinates.
  • Single Point Energy calculations.
  • Basis sets and Density Functional Theory.
  • The Self-Consistent Field (SCF) procedure.
  • Gaussian input and output files.

Example Used Throughout This Section

Throughout this documentation, the Geometry Optimization procedure will be demonstrated using a formic acid dimer calculated at the B3LYP/6-31G level of theory. The same example will be used consistently to explain the input file, optimization procedure, output file, convergence criteria, and the mathematical foundations of geometry optimization.