Aneesur Rahman
Aneesur Rahman (24 August 1927 – 6 June 1987) was an Indian-born American physicist whose research established several foundational methods in computational statistical mechanics. His numerical study of liquid argon provided one of the earliest systematic demonstrations that the time-dependent microscopic behavior of a dense liquid could be calculated by integrating the classical equations of motion for a finite collection of interacting atoms. Subsequent work extended this approach to molecular liquids, phase transitions, and solids with dynamically variable simulation cells.
Rahman’s principal contributions concerned molecular dynamics, the calculation of time-correlation functions, and the relationship between atomistic trajectories and measurable thermodynamic properties. His work formed part of the broader development of computer simulation as a method positioned between analytical theory and laboratory experiment.
Education and scientific context
Rahman was born in Hyderabad, then part of the princely Hyderabad State. He studied physics at Osmania University and continued his education at the Indian Institute of Science. He later completed doctoral research at the Catholic University of Leuven, receiving his doctorate in physics in 1953.
His training coincided with the emergence of electronic computation as a practical instrument for theoretical physics. Earlier approaches to liquids had depended primarily on approximate integral equations, perturbative calculations, and interpretations of experimental structure data. Electronic computers made it possible to represent a many-particle system directly and to calculate its evolution through successive numerical time steps.
Rahman joined Argonne National Laboratory, where access to large scientific computers enabled him to investigate collective atomic motion. The laboratory environment brought condensed-matter theory into sustained contact with numerical analysis and high-performance computing. This institutional setting was central to the transition of molecular dynamics from a small collection of preliminary calculations into a reproducible research methodology.
Liquid-argon simulation
Rahman’s 1964 article, “Correlations in the Motion of Atoms in Liquid Argon,” examined a model containing 864 argon atoms. The atoms interacted through a Lennard-Jones potential, which approximated the balance between short-range repulsion and longer-range attraction. Their classical trajectories were obtained by repeatedly integrating Newton's equations of motion under periodic boundary conditions.
The calculation used a CDC 3600 computer and generated a continuous numerical history of particle positions and velocities. Rahman developed the computational formulation, while You Watanabe participated in the organization of the trajectory calculations and the numerical analysis of their correlation data. The resulting study treated the simulated trajectory as a statistical sample from which both equilibrium structure and time-dependent behavior could be extracted.
The computed radial distribution function described the probability of finding pairs of atoms at specified separations. Its oscillatory form represented the short-range organization characteristic of a dense liquid, in which local coordination persists without the long-range periodicity of a crystal. Agreement between the calculated distribution and available scattering measurements demonstrated that a relatively simple pair potential could reproduce major structural properties of liquid argon.
Rahman also calculated the velocity autocorrelation function, which measures the persistence of an atom’s velocity over time. The function became negative at intermediate times because an atom moving through a dense liquid was frequently redirected by the temporary cage formed by its neighbors. This result connected a microscopic dynamical pattern with diffusion and other macroscopic transport processes.
The study differed from a static calculation because it preserved the temporal ordering of atomic configurations. A numerical trajectory could therefore be used to investigate relaxation, collective motion, and transport coefficients in addition to equilibrium averages. This combination of structural and dynamical analysis became a standard feature of later molecular simulations.
Molecular dynamics as a research method
Early molecular-dynamics calculations were constrained by processor speed, memory capacity, and numerical precision. These limitations required models with modest particle numbers and relatively short simulated durations. Rahman nevertheless showed that such finite calculations could produce physically interpretable correlation functions when the system size, boundary conditions, interaction model, and integration procedure were treated consistently.
His work helped establish the computational experiment as a distinct form of physical inquiry. Unlike a closed analytical derivation, a simulation produced numerical trajectories whose interpretation required statistical reduction. Unlike a laboratory experiment, it operated on a specified mathematical model in which every interparticle force was defined in advance. Comparison with measured properties could consequently test both the model potential and the numerical method used to evolve it.
The liquid-argon calculation also clarified the importance of fluctuations. Thermodynamic quantities represented averages over a trajectory, while instantaneous configurations exhibited substantial local variation. The trajectory itself therefore contained information not visible in a single equilibrium structure. Later simulation research generalized this distinction through systematic use of ensemble averages, correlation functions, and Green–Kubo relations.
Simulation of molecular liquids
Rahman later collaborated with Frank H. Stillinger on the molecular simulation of liquid water. Their calculations represented each water molecule through a structured interaction model rather than as a single spherical particle. The electrostatic arrangement and molecular geometry generated orientational correlations associated with hydrogen bonding.
This research demonstrated that molecular dynamics could treat liquids whose behavior depended on directional interactions. Water presented a substantially different modeling problem from liquid argon because translational motion was coupled to molecular orientation and to a fluctuating intermolecular network. The work contributed to the development of numerical water models used in physical chemistry, materials science, and later biomolecular simulation.
The water studies also illustrated the dependence of simulation results on the interaction potential. A trajectory exactly integrates the chosen model only within the limits of numerical discretization; it does not remove approximations embedded in the representation of intermolecular forces. Rahman’s research consequently linked computational technique with the continuing problem of constructing potentials that reproduce relevant physical observables.
Variable-cell dynamics
A later development arose from Rahman’s collaboration with Michele Parrinello. The resulting Parrinello–Rahman method treated the dimensions and shape of the simulation cell as dynamical variables rather than fixed external parameters. This formulation allowed a simulated solid to respond collectively to pressure and internal stress.
In a conventional fixed-cell calculation, atoms move inside a box whose geometry remains unchanged. Such a constraint can prevent the representation of transformations that require a change in lattice symmetry or cell shape. Parrinello–Rahman dynamics enlarged the system’s mathematical phase space by associating generalized coordinates and an effective inertia with the cell matrix.
The method enabled direct investigation of pressure-induced structural transformations and elastic responses. It became especially relevant to simulations of crystalline phases, where a transition can involve coordinated changes in both atomic positions and lattice vectors. Later constant-pressure algorithms modified its numerical implementation while retaining the underlying concept of a dynamically evolving periodic cell.
Variable-cell dynamics also influenced the development of first-principles molecular dynamics. In that context, interatomic forces are calculated from an electronic-structure model during the trajectory rather than supplied entirely by a predetermined classical potential. The combination of electronic forces with an adaptable simulation cell provided a framework for studying materials under changing mechanical conditions.
Scientific significance
Rahman’s research contributed to the establishment of molecular simulation as a quantitative component of condensed-matter physics. The liquid-argon work demonstrated that microscopic trajectories could reproduce recognizable structural data and reveal dynamical correlations that were difficult to obtain from static theory. His studies of water extended this methodology to directional molecular interactions, while variable-cell dynamics incorporated collective deformation into the equations governing a simulated system.
These developments share a common methodological structure. Each replaced a macroscopic constraint with explicitly evolving microscopic or collective variables, after which statistical analysis connected the calculated motion to observable behavior. The resulting framework is now used in research on liquids, crystalline solids, polymers, surfaces, and molecular materials.
The American Physical Society commemorates this area of research through the Aneesur Rahman Prize for Computational Physics. The prize recognizes work involving computational methods and their application to physical problems, reflecting the disciplinary role acquired by the methods that Rahman helped develop.