What the study found
The study found that thermostat algorithms in constant-temperature molecular dynamics simulations do not perform identically. The Nosé-Hoover chain and Bussi velocity rescaling methods gave reliable temperature control, while the Grønbech-Jensen-Farago Langevin scheme was the most consistent for sampling both temperature and potential energy.
Why the authors say this matters
The authors conclude that the comparison offers practical guidance for choosing thermostats in classical molecular dynamics simulations. They also say the findings provide useful insights for applications including glass transition, phase separation, and nucleation.
What the researchers tested
The researchers carried out a systematic comparison of representative thermostat methods in constant-temperature molecular dynamics. They examined the Nosé-Hoover thermostat, its chain generalization, the Bussi velocity rescaling method, and several Langevin dynamics implementations using a binary Lennard-Jones liquid as a model glass former.
What worked and what didn't
The Nosé-Hoover chain and Bussi thermostats provided reliable temperature control, but potential energy showed a pronounced dependence on time step. Among the Langevin methods, the Grønbech-Jensen-Farago scheme gave the most consistent sampling of temperature and potential energy. The abstract also says Langevin dynamics typically costs about twice as much computationally because of random number generation overhead, and that diffusion coefficients decrease systematically as friction increases.
What to keep in mind
The summary describes a comparison on a binary Lennard-Jones liquid model glass former, so the results are limited to that setting. The abstract does not describe additional limitations beyond the time-step dependence, computational cost, and friction-related diffusion changes it reports.
Key points
- Nosé-Hoover chain and Bussi thermostats provided reliable temperature control.
- Potential energy showed a pronounced dependence on time step for some thermostats.
- The Grønbech-Jensen-Farago Langevin scheme was the most consistent for temperature and potential energy sampling.
- Langevin dynamics typically required about twice the computational cost.
- Diffusion coefficients decreased systematically as friction increased.
Disclosure
- Research title:
- Thermostat methods differ in temperature control and energy sampling
- Authors:
- Kumpei Shiraishi, Emi Minamitani, Kang Kim
- Institutions:
- Osaka University of Economics, The University of Osaka, The University of Osaka, The University of Osaka
- Publication date:
- 2026-04-22
- DOI:
- 10.1063/5.0327041
- OpenAlex record:
- View
Get the weekly research newsletter
Stay current with scholarly research without reading academic papers — one filtered digest, every Friday.