<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=Content-Type content="text/html; charset=iso-8859-1">
<META content="MSHTML 6.00.2600.0" name=GENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY style="COLOR: #000080; FONT-FAMILY: Times New Roman" bgColor=#ffffff>
<DIV>Dear all, </DIV>
<DIV> </DIV>
<DIV>For an isobaric-isothermal MD simulation, I need to set the target electron
kinetic energy. According to the manual, it's suggested to perform a
short run without thermostats, and then take the "natural" electron kinetic
energy as the target value (see, e.g, the CPMD manual, version 3.5.2 pp
62). Still, I found this "target" value very capricious. After all, I
arbitrarily set the kinetic energy to be 0.0001 a.u . During
simulation I found that the output EKINC (the fictitious electron kinetic
energy, ~ 0.007 a.u. after 1000 time steps) differs significantly from this
desired kinetic energy 0.0001 a.u. I think this is abnormal, because EKINC
should fluctuate around the desired (also predefined) kinetic energy. Can anyone
tell me why?</DIV>
<DIV> </DIV>
<DIV>The ion temperature is also a pain in the neck. I have tried a number of
NOSE parameters, but the temperature fluctuation is still too much. For example,
the target ion temperature is 5000 Kelvin, and the fluctuation reaches as much
as 1000 K. A possible reason is that the target electron kinetic energy is
set too small and causes the damping of the ions. Do I make any sense?</DIV>
<DIV> </DIV>
<DIV>Besides, any rule-of-thumb to set the CELL kinetic energy? </DIV>
<DIV>In a nutshell, I wish to have a guidance in choosing the Nose-Hover
chain parameters. I appreciate your help.</DIV>
<DIV> </DIV>
<DIV> </DIV></BODY></HTML>