<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<tt>Dear Jie Peng,<br>
<br>
suppose you were running a model harmonic system in 1 dimension.<br>
<br>
M a = - K x<br>
<br>
at fixed energy E.<br>
<br>
The kinetic energy would fluctuate harmonically between 0 (at
maximum/minimum elongation) and E at the equilibrium distance .<br>
<br>
On average the Kinetic energy would be E/2 and its fluctuation
some big fraction of E^2<br>
<br>
Something like sigma^2 = 1/T \int_0^T (E cos^2(2pi t/T) -E/2)^2
dt = E^2 1/T \int_0^T (cos(4pi t/T)/2)^2 dt = (E/2)^2 1/2pi
\int_0^2pi cos^2(x) dx = (E/2)^2 / 2<br>
<br>
or<br>
<br>
sigma = 1/sqrt(2) * E/2 = 1/sqrt(2) avg EKin<br>
<br>
with 1 degree of freedom the mean square fluctuation of the
kinetic energy is 70% of its average !<br>
<br>
you have 3 atoms in your cell hence 9 degrees of freedom.
Assuming each contributes independently to the average this goes
down by a factor 1/sqrt(9)=1/3<br>
<br>
actually more likely just 1/sqrt(6) as the total momentum is
conserved so only 6 modes at Gamma are actually excited...<br>
<br>
If you perform your simulation in a bigger supercell with more
atoms (more degrees of freedom) the average will be more
stable ( proportionally to 1/sqrt(#deg.of.freedom-3 ) ...
moreover the thermal excitations of vibrational modes will be
sampled more faithfully.<br>
<br>
best<br>
<br>
stefano<br>
<br>
<br>
</tt>
<div class="moz-cite-prefix">On 13/04/2018 21:39, Jie Peng wrote:<br>
</div>
<blockquote type="cite"
cite="mid:CAJTOooHctbCQ+b-WGbu8kxVVNU7vHVJRQcRRno5fYX4aUTa6ug@mail.gmail.com">
<div dir="ltr">Dear all
<div><br>
</div>
<div>I have been running MD simulations on HfS2 using cp.x code
in Quantum espresso. I start from initial configuration
obtained from pwscf vc-relax, and relax the system using cp.x
by consecutive steps of: electron relaxation->ionic
relaxation->cell relaxation. Then, I just directly start a
NVE simulation starting from the equilibrium configuration. I
expect the system to almost stay stationary or the temperature
should be very small since I am allowing dynamics in a system
that is already in equilibrium. However, what I see is a huge
fluctuation in the <i>tmpp</i> output of cp.x, as I attach a
figure showing variation of tmpp (Ionic temperature) with
simulation time</div>
<div><img src="cid:part1.93951B92.1BE23FC8@sissa.it" class=""
width="481" height="288"></div>
<div><br>
I did this because it is suggested in the user guide you
should apply an initial displacement to the atoms in your
system after the relaxation since otherwise there will not be
any dynamics. But what I see here is a large fluctuation of
the system temperature.</div>
<div><br>
</div>
<div>The thinking or questions here are</div>
<div><br>
</div>
<div>1.Does the tmpp represents the physical temperature of the
system here? I think it should be since it is the temperature
corresponding to kinetic energy of the ions.</div>
<div><br>
2.It above point is true, why is the temperature varying so
fiercely? Am I setting incorrect parameters, for instance the
timestep or the fictitious mass? But I took those from
previous simulation steps where I did the relaxation, and they
all worked well since they successfully drived my system to
equilibrium, satisfying the convergence threshold on total
energy, forces acting on atoms, and the fictitious electron
kinetic energy. I am confused at this point.</div>
<div><br>
</div>
<div>The input file for NVE simulation is attached here:</div>
<div><br>
</div>
<div>
<div><i>&control</i></div>
<div><i> calculation='cp',</i></div>
<div><i> title='Halfnium disulfide'</i></div>
<div><i> restart_mode='restart',</i></div>
<div><i> ndr=53,</i></div>
<div><i> ndw=54,</i></div>
<div><i> nstep=50000,</i></div>
<div><i> iprint=10</i></div>
<div><i> isave=100,</i></div>
<div><i> tstress = .true.</i></div>
<div><i> tprnfor = .true.</i></div>
<div><i> dt=10,</i></div>
<div><i> wf_collect=.true.</i></div>
<div><i> etot_conv_thr=1e-6</i></div>
<div><i> forc_conv_thr=1e-3</i></div>
<div><i> ekin_conv_thr=1e-5</i></div>
<div><i> prefix='HfS2',</i></div>
<div><i> pseudo_dir='/home/jpeng/HfS2/potential'</i></div>
<div><i> outdir='./tmp/',</i></div>
<div><i> /</i></div>
<div><i> &system</i></div>
<div><i> ibrav= 4,</i></div>
<div><i> a=3.6529</i></div>
<div><i> c=5.6544</i></div>
<div><i> nat= 3, ntyp= 2,</i></div>
<div><i> ecutwfc =50</i></div>
<div><i> vdw_corr='DFT-D',</i></div>
<div><i> ! lspinorb=.true.</i></div>
<div><i> ! noncolin=.true.</i></div>
<div><i> ! ecutrho=300</i></div>
<div><i> ! nbnd=14</i></div>
<div><i>! occupations='smearing'</i></div>
<div><i>! smearing='gaussian'</i></div>
<div><i>! degauss=0.01</i></div>
<div><i> ! nspin=2</i></div>
<div><i> ! starting_magnetization(1)=0.1</i></div>
<div><i>! Hf 95.94 Hf.pbe-mt_fhi.UPF</i></div>
<div><i>! S 32.065 S.pbe-mt_fhi.UPF</i></div>
<div><i>/</i></div>
<div><i> &electrons</i></div>
<div><i> electron_dynamics='verlet'</i></div>
<div><i> electron_velocities='zero'</i></div>
<div><i> emass=400</i></div>
<div><i> emass_cutoff=1</i></div>
<div><i>/</i></div>
<div><i> &ions</i></div>
<div><i> ion_dynamics = 'verlet'</i></div>
<div><i> ion_damping=0.1</i></div>
<div><i>! ion_nstepe=10</i></div>
<div><i> /</i></div>
<div><i> &cell</i></div>
<div><i> cell_dynamics = 'none'</i></div>
<div><i> </i></div>
<div><i>/</i></div>
<div><i>ATOMIC_SPECIES</i></div>
<div><i> Hf 95.94 Hf.pbe-mt_fhi.UPF</i></div>
<div><i> S 32.065 S.pbe-mt_fhi.UPF</i></div>
<div><i>ATOMIC_POSITIONS (crystal)</i></div>
<div><i>Hf -0.000000000 -0.000000000 -0.000000000</i></div>
<div><i>S 0.666666667 0.333333333 0.257234636</i></div>
<div><i>S 0.333333333 0.666666667 -0.257234636</i></div>
<div><br>
</div>
<div>Anyone could help me on it? Thank you very much.</div>
<div><br>
</div>
<div>Best</div>
<div>Jie</div>
-- <br>
<div class="gmail_signature">
<div dir="ltr">
<div>
<div style="font-size:12.8px">------------------------------------------------------------------------------------------------------------------------<br>
Jie Peng</div>
<div style="font-size:12.8px">PhD student<br>
2134 Glenn Martin Hall, Mechanical Engineering,
University of Maryland<br>
College Park, Maryland, USA<br>
Phone:(+1) 240-495-9445<br>
</div>
<div style="font-size:12.8px">Email: <a
href="mailto:jiepeng@umd.edu" target="_blank"
moz-do-not-send="true">jiepeng@umd.edu</a><br>
</div>
</div>
<div><br>
</div>
</div>
</div>
</div>
</div>
<br>
<fieldset class="mimeAttachmentHeader"></fieldset>
<br>
<pre wrap="">_______________________________________________
users mailing list
<a class="moz-txt-link-abbreviated" href="mailto:users@lists.quantum-espresso.org">users@lists.quantum-espresso.org</a>
<a class="moz-txt-link-freetext" href="https://lists.quantum-espresso.org/mailman/listinfo/users">https://lists.quantum-espresso.org/mailman/listinfo/users</a></pre>
</blockquote>
<br>
</body>
</html>