Skip to content

A unified formulation of the constant temperature molecular dynamics methods

Why this mattered

Nosé’s 1984 paper mattered because it put constant-temperature molecular dynamics on a statistical-mechanical footing rather than treating thermostats as merely numerical devices for controlling kinetic energy. By deriving equilibrium distribution functions and comparing them directly with the canonical ensemble, the paper clarified which proposed methods actually sampled the intended ensemble and under what constraints. Its central result was that the Nosé extended-system formulation could generate the rigorous canonical distribution, apart from conserved total momentum and angular momentum effects. That turned thermostat design into a problem of ensemble correctness, not just temperature regulation.

The practical shift was large: molecular dynamics could now be used more confidently as a route to equilibrium thermodynamic averages at fixed temperature, not only as a simulation of microcanonical trajectories. The paper also showed that the Hoover-style approach could be understood as a constrained form of Nosé’s method, helping unify what had appeared to be separate techniques. This connection directly enabled the Nosé-Hoover thermostat, which became one of the standard tools of atomistic simulation because it offered deterministic, time-reversible dynamics compatible with canonical sampling under appropriate conditions.

Its influence extends through later molecular simulation methods that depend on controlled ensemble sampling: constant-pressure dynamics, free-energy calculations, biomolecular simulation, materials modeling, and nonequilibrium molecular dynamics. The paper did not solve every ergodicity problem associated with deterministic thermostats, but it supplied the conceptual framework in which those problems could be diagnosed and improved. In that sense, Nosé’s formulation helped transform molecular dynamics from trajectory generation into a general-purpose computational statistical mechanics method.

Abstract

Three recently proposed constant temperature molecular dynamics methods by: (i) Nosé (Mol. Phys., to be published); (ii) Hoover et al. [Phys. Rev. Lett. 48, 1818 (1982)], and Evans and Morriss [Chem. Phys. 77, 63 (1983)]; and (iii) Haile and Gupta [J. Chem. Phys. 79, 3067 (1983)] are examined analytically via calculating the equilibrium distribution functions and comparing them with that of the canonical ensemble. Except for effects due to momentum and angular momentum conservation, method (1) yields the rigorous canonical distribution in both momentum and coordinate space. Method (2) can be made rigorous in coordinate space, and can be derived from method (1) by imposing a specific constraint. Method (3) is not rigorous and gives a deviation of order N−1/2 from the canonical distribution (N the number of particles). The results for the constant temperature–constant pressure ensemble are similar to the canonical ensemble case.

Sources