[Pw_forum] TAGS for simulate annealing

YY xrhino at gmail.com
Thu Feb 16 07:33:48 CET 2012


Dear all,

I am trying to simulate the annealing process using pw.x.  I set
ion_temperature, tempw, delta_t, nraise in the IONS namelist.  As my
understanding, temperature at the N-step should be T(N) = tempw -
delta_t * N / nraise.  But after a short test, I find Ekin + Etot is a
constant. The temperature fluctuated, and will never go to zero.

I want to know are there any other tags I can use to simulate the
annealing process.  At the end, I attached my input file.

Thanks for your help,
Yuyang
A postdoc at Rensselaer Polytechnic Institute, Troy, NY, USA

INPUT FILE
&control
   calculation='md'
   restart_mode='from_scratch'
   pseudo_dir = './'
   outdir='./tmp'
   prefix='mdtest'
   wf_collect=.true.
   nstep=1250
/
&system
   ibrav=  0,   celldm(1)=1.8897197
   nat=  63,   ntyp= 4
   ecutwfc =25,   ecutrho=150
   occupations='smearing',   smearing='gaussian',   degauss=0.0072
/
&electrons
/
&ions
   ion_temperature = 'reduce-T'
   tempw = 500
   delta_t = 2
   nraise = 5

/
CELL_PARAMETERS cubic
    10.8441000000000000    0.0000000000000000    0.0000000000000000
     0.0000000000000000   10.2440000000000000    0.0000000000000000
     0.0000000000000000    0.0000000000000000   20.0000000000000000
...
K_POINTS {gamma}



More information about the users mailing list