Computational chemistry

Computational chemistry is a branch of chemistry that uses computer simulation to assist in solving complex chemical problems. It employs methods of theoretical chemistry, incorporated into efficient computer programs, to calculate the structures and properties of molecules and solids. Its necessity arises from the fact that, apart from relatively recent results concerning the hydrogen molecular ion, the Schrödinger equation for systems of more than one electron cannot be solved analytically. While computational results normally complement the information obtained by chemical experiments, they can in some cases predict hitherto unobserved chemical phenomena.

Overview

The field is a direct product of the digital age, leveraging the power of modern supercomputers and sophisticated algorithms to explore chemical space. It intersects heavily with physics, mathematics, and computer science, forming a core discipline within the broader realm of molecular modelling. Pioneering work by scientists like John Pople, who developed the Gaussian (software) program and won the Nobel Prize in Chemistry in 1998, and Walter Kohn, a co-recipient that year for his work on density functional theory, established its foundational credibility. Major research is conducted at institutions like the University of California, Berkeley, Massachusetts Institute of Technology, and the Max Planck Society.

Theoretical foundations

The discipline is built upon the principles of quantum mechanics and statistical mechanics. The central equation is the Schrödinger equation, which describes how the quantum state of a physical system changes over time. For many-electron systems, approximations are essential, leading to the development of the Born–Oppenheimer approximation. Key theoretical frameworks include ab initio quantum chemistry methods, which attempt to solve the Schrödinger equation from first principles, and semi-empirical quantum chemistry methods, which incorporate experimental data. The work of Robert S. Mulliken on molecular orbital theory and Linus Pauling on valence bond theory provided critical conceptual tools.

Methods and approaches

A wide spectrum of methodologies exists, each with a specific balance of accuracy and computational cost. Molecular mechanics uses classical Newtonian mechanics and force field (chemistry) parameters to model large systems like proteins and polymers, as seen in software like AMBER and CHARMM. For electronic structure, Hartree–Fock method provides a mean-field starting point, while post-Hartree–Fock methods like coupled cluster and Møller–Plesset perturbation theory introduce electron correlation. Density functional theory, popularized by Walter Kohn and Pierre Hohenberg, has become a workhorse for materials and inorganic chemistry. Molecular dynamics simulations, pioneered by researchers like Aneesur Rahman, track atomic motions over time.

Applications

Its applications are vast and transformative across scientific and industrial domains. In pharmaceutical research, it is crucial for drug design, predicting how potential drug candidates bind to targets like the HIV protease. In materials science, it aids in designing novel catalysts, battery components, and semiconductors. It is used to study atmospheric chemistry, such as the breakdown of ozone by chlorofluorocarbons, and in petroleum engineering to model hydrocarbon reservoirs. Organizations like the National Institutes of Health and the Department of Energy heavily fund such research for its potential in addressing challenges in medicine and energy.

Software and tools

The field is enabled by a wide array of specialized software packages and programming environments. Prominent commercial packages include Gaussian (software), Materials Studio, and Schrödinger. Open-source projects like GAMESS (US), NWChem, and CP2K are also widely used in academia. For molecular visualization and analysis, tools like VMD (software), PyMOL, and Chimera (software) are standard. High-performance computing is facilitated by libraries such as MPI and OpenMP, often run on systems at national facilities like Argonne National Laboratory and Oak Ridge National Laboratory.

Challenges and limitations

Despite its power, the discipline faces significant hurdles. The high computational cost of accurate methods for large systems, often referred to as the scalability problem, remains a primary constraint. The accuracy of results depends heavily on the chosen method and basis set, and all approximations introduce some error. The development of accurate and transferable force field (chemistry) parameters for novel materials is a persistent challenge. Furthermore, the interpretation of massive datasets from simulations, such as those generated by the Folding@home project, requires sophisticated data analysis techniques. Ongoing research at institutions like the Simons Foundation aims to push these boundaries through new algorithmic and hardware advances.

Category:Chemistry