Conserving Energy

From wiki.gromacs.org

In principle Molecular Dynamics simulation should conserve the total energy, the total momentum and (in a non-periodic system) the total angular momentum. A number of algorithmic and numerical issues make that this is not always the case.

• Cut-off treatment and/or long-range electrostatics treatment (Van der Spoel & Van Maaren, J. Chem. Theor. Comp. 2, 1-11 (2006))
• Treatment of neighborlists
• Constraint algorithms (see e.g. Hess, J. Chem. Theor. Comp. 2008)
• The integration timestep
• Temperature coupling and pressure coupling
• Round-off error (in particular in single precision). Example could be subtracting large numbers (Lippert at al., J. Chem. Phys. 126, 046101 (2007))
• The choice of the integration algorithm (In GROMACS this is leap-frog)
• Removal of center of mass motion: when doing this in more than one group the conservation of energy will be violated