<div style="line-height:1.7;color:#000000;font-size:14px;font-family:arial">According to my experience, the general process is right, but, this general process is normally working for crystal rather than amorphous, liquid and gas. Since the contribution from vibration is not that much in those matter, for example, transition contributes much more in gas phase. Therefore, Monte Carlo or Molecular Dynamic could be better choice in these cases.<br>Technically, in your calculation, H2O molecule may exist in a box, therefore only the Gamma phonon calculation is meaningful. Since q2r means fitting the finite samplings of dynamic matrics to infinite interatomic force constants, only one point is not that reliable to find good aproach of IFCs. Actually, since vibrational free energy is defined with the term of hv, you can derectly obtain Fvib from your first step.<br><br><div>--<br>GAO Zhe<br>CMC Lab, Materials Science &amp; Engineering Department,<br>Seoul National University, South Korea<br>        

&nbsp;</div><div id="divNeteaseMailCard"></div><br><pre><br>At&nbsp;2012-07-10&nbsp;22:44:56,"Torstein&nbsp;Fjermestad"&nbsp;&lt;torstein.fjermestad@kjemi.uio.no&gt;&nbsp;wrote:

&gt;&nbsp;Dear&nbsp;all,

&gt;

&gt;&nbsp;In&nbsp;one&nbsp;of&nbsp;my&nbsp;projects&nbsp;I&nbsp;would&nbsp;need&nbsp;to&nbsp;compute&nbsp;the&nbsp;Gibbs&nbsp;free&nbsp;energies&nbsp;

&gt;&nbsp;of&nbsp;structures&nbsp;along&nbsp;a&nbsp;reaction&nbsp;path.&nbsp;After&nbsp;looking&nbsp;through&nbsp;the&nbsp;QE&nbsp;

&gt;&nbsp;webpage,&nbsp;the&nbsp;pw_forum,&nbsp;and&nbsp;the&nbsp;documentation&nbsp;in&nbsp;the&nbsp;quantum&nbsp;espresso&nbsp;

&gt;&nbsp;installation&nbsp;directory&nbsp;I&nbsp;think&nbsp;I&nbsp;understand&nbsp;the&nbsp;general&nbsp;procedure:

&gt;

&gt;&nbsp;1.&nbsp;one&nbsp;calculates&nbsp;the&nbsp;phonons

&gt;&nbsp;2.&nbsp;one&nbsp;computes&nbsp;the&nbsp;interatomic&nbsp;Force&nbsp;Constants&nbsp;with&nbsp;the&nbsp;q2r.x&nbsp;program

&gt;&nbsp;3.&nbsp;one&nbsp;uses&nbsp;the&nbsp;matdyn.x&nbsp;to&nbsp;produce&nbsp;input&nbsp;for&nbsp;the&nbsp;F_QHA.x&nbsp;program

&gt;&nbsp;4.&nbsp;with&nbsp;the&nbsp;F_QHA.x&nbsp;program&nbsp;one&nbsp;can&nbsp;compute&nbsp;the&nbsp;Gibbs&nbsp;free&nbsp;energy&nbsp;and&nbsp;

&gt;&nbsp;other&nbsp;thermodynamical&nbsp;properties.

&gt;

&gt;&nbsp;(please&nbsp;correct&nbsp;me&nbsp;if&nbsp;I&nbsp;have&nbsp;misunderstood&nbsp;this&nbsp;general&nbsp;procedure)

&gt;

&gt;&nbsp;The&nbsp;problem&nbsp;is&nbsp;that&nbsp;I&nbsp;have&nbsp;so&nbsp;far&nbsp;only&nbsp;managed&nbsp;to&nbsp;complete&nbsp;step&nbsp;1,&nbsp;to&nbsp;

&gt;&nbsp;compute&nbsp;the&nbsp;phonons.&nbsp;I&nbsp;am&nbsp;currently&nbsp;doing&nbsp;test&nbsp;calculations&nbsp;on&nbsp;a&nbsp;water&nbsp;

&gt;&nbsp;molecule&nbsp;and&nbsp;this&nbsp;is&nbsp;the&nbsp;input&nbsp;file&nbsp;of&nbsp;the&nbsp;phonon&nbsp;calculation:

&gt;

&gt;&nbsp;phonons&nbsp;of&nbsp;H2O&nbsp;in&nbsp;super&nbsp;cell&nbsp;at&nbsp;Gamma

&gt;&nbsp;&nbsp;&amp;inputph

&gt;&nbsp;&nbsp;&nbsp;outdir='./',

&gt;&nbsp;&nbsp;&nbsp;prefix='h2o',

&gt;&nbsp;&nbsp;&nbsp;amass(1)=1.01,

&gt;&nbsp;&nbsp;&nbsp;amass(2)=16.01,

&gt;&nbsp;&nbsp;&nbsp;fildyn='h2o.dynG',

&gt;&nbsp;&nbsp;&nbsp;trans=&nbsp;.true.

&gt;&nbsp;&nbsp;/

&gt;&nbsp;0.0&nbsp;0.0&nbsp;0.0

&gt;

&gt;&nbsp;I&nbsp;then&nbsp;try&nbsp;to&nbsp;run&nbsp;the&nbsp;q2r.x&nbsp;calculation&nbsp;with&nbsp;the&nbsp;following&nbsp;input:

&gt;

&gt;&nbsp;test&nbsp;of&nbsp;q2r&nbsp;calculation

&gt;&nbsp;&amp;input

&gt;&nbsp;&nbsp;&nbsp;fildyn='h2o.dynG',

&gt;&nbsp;&nbsp;&nbsp;flfrc='h2o-output-q2r.fc',

&gt;&nbsp;&nbsp;&nbsp;zasr='no'

&gt;

&gt;&nbsp;This&nbsp;calculation&nbsp;crashes&nbsp;presumably&nbsp;because&nbsp;it&nbsp;expects&nbsp;a&nbsp;file&nbsp;called&nbsp;

&gt;&nbsp;h2o.dynG0.&nbsp;But&nbsp;from&nbsp;where&nbsp;do&nbsp;I&nbsp;get&nbsp;this&nbsp;file?&nbsp;As&nbsp;far&nbsp;as&nbsp;I&nbsp;can&nbsp;see&nbsp;the&nbsp;

&gt;&nbsp;phonon&nbsp;calculation&nbsp;only&nbsp;produces&nbsp;the&nbsp;file&nbsp;h2o.dynG,&nbsp;not&nbsp;a&nbsp;h2o.dynG0&nbsp;

&gt;&nbsp;file.

&gt;

&gt;&nbsp;I&nbsp;am&nbsp;completely&nbsp;lost&nbsp;at&nbsp;the&nbsp;moment&nbsp;so&nbsp;I&nbsp;would&nbsp;appreciate&nbsp;if&nbsp;anyone&nbsp;

&gt;&nbsp;could&nbsp;help&nbsp;me&nbsp;with&nbsp;carrying&nbsp;out&nbsp;the&nbsp;q2r&nbsp;calculation&nbsp;correctly.

&gt;&nbsp;

&gt;&nbsp;Thank&nbsp;you&nbsp;very&nbsp;much&nbsp;in&nbsp;advance,

&gt;

&gt;&nbsp;Yours&nbsp;sincerely,

&gt;

&gt;&nbsp;Torstein&nbsp;Fjermestad,

&gt;&nbsp;University&nbsp;of&nbsp;Oslo,

&gt;&nbsp;Norway

&gt;

&gt;

&gt;

&gt;

&gt;

&gt;

&gt;

&gt;

&gt;

&gt;

&gt;&nbsp;

&gt;_______________________________________________

&gt;Pw_forum&nbsp;mailing&nbsp;list

&gt;Pw_forum@pwscf.org

&gt;http://www.democritos.it/mailman/listinfo/pw_forum

</pre></div><br><br><span title="neteasefooter"><span id="netease_mail_footer"></span></span>