drag value = drag factor added to thermostat (0.0 = no drag) +drag value = drag factor added to thermostat (0.0 = no drag) + chain value = yes or noExamples:
fix 1 all nvt 300.0 300.0 100.0 -fix 1 all nvt 300.0 300.0 100.0 drag 0.2 +fix 1 all nvt 300.0 300.0 100.0 drag 0.2 chain noDescription:
@@ -55,15 +56,29 @@ example, a value of 100.0 means to relax the temperature in a timespan of (roughly) 100 time units (tau or fmsec or psec - see the units command). +The chain keyword determines whether Nose/Hoover chains are used or +not. If chain is specified as no, then the original Nose/Hoover +formulation is used. If chain is specified as yes, which is the +default, then chains as described in (Martyna) are used +which include extra non-physical variables which couple to the +thermostat. Nose/Hoover chains provide a more robust NVT integrator, +overcoming non-ergodic sampling issues and energy oscillations found +with ordinary Nose/Hoover dynamics. Our implementation uses one chain +and integrates the equations of motion via a Trotter expansion good to +2nd order accuracy in the timestep size. +
In some cases (e.g. for solids) the temperature of the system can -oscillate undesirably when a Nose/Hoover thermostat is applied. The -optional drag keyword will damp these oscillations, although it -alters the Nose/Hoover equations. A value of 0.0 (no drag) leaves the -Nose/Hoover formalism unchanged. A non-zero value adds a drag term; -the larger the value specified, the greater the damping effect. -Performing a short run and monitoring the temperature is the best way -to determine if the drag term is working. Typically a value between -0.2 to 2.0 is sufficient to damp oscillations after a few periods. +oscillate undesirably when a Nose/Hoover thermostat is applied, though +this should be less of a problem if Nose/Hoover chains are used. The +optional drag keyword will damp these oscillations in an ad-hoc +fashion, by altering the Nose/Hoover equations so that they no longer +exactly sample the canonical ensemble. A value of 0.0 (no drag) +leaves the Nose/Hoover formalism unchanged. A non-zero value adds a +drag term; the larger the value specified, the greater the damping +effect. Performing a short run and monitoring the temperature is the +best way to determine if the drag term is working. Typically a value +between 0.2 to 2.0 is sufficient to damp oscillations after a few +periods.
IMPORTANT NOTE: Unlike the fix temp/berendsen command which performs thermostatting but NO time integration, this @@ -159,7 +174,7 @@ temp/rescale, fix langevin,
Default:
-The keyword defaults are drag = 0.0. +
The keyword defaults are drag = 0.0 and chain = yes.
@@ -167,4 +182,9 @@ temp/rescale, fix langevin,(Hoover) Hoover, Phys Rev A, 31, 1695 (1985).
+ + +(Martyna) Martyna, Klein, Tuckerman, J Chem Phys, 97, 2635 (1992); +Martyna, Tuckerman, Tobias, Klein, Mol Phys, 87, 1117. +