Juerg,<div><br></div><div> I tried the new RT_PROPAGATION feature on both water and metal cluster systems. By checking the energy conservation along with other properties, it seems that the new feature works pretty well. </div>
<div><br></div><div> I have another question. Does CP2k allow user to perturb the wave function at the beginning of a real time propagation? By enabling this feature, the evolution of a system after an ultra-short-time electric pulse being applied can be studied, and many important physical properties, such as photoabsorption spectrum and dynamic polarizability, can be obtained. Theoretically, the idea is quite simple as shown below:</div>
<div><br></div><div> Given a wave function psi(t) at time t, the wave function after the ultra-short-time electric field perturbation psi_new(t) can be expressed as psi_new(t) = exp( i * K * r) * psi(t), where K is the electric field strength, and r is the direction of the field. Thereafter, the system evolves as usual. </div>
<div><br></div><div> I do not know if CP2k developers have a plan for this feature, or I will be happy to contribute with the developer's help. </div><div><br></div><div> Thanks.</div><div><br></div><div>Hanning Chen</div>
<div><br></div><div>Department of Chemistry</div><div>Northwestern University</div><div>Evanston, IL 60208</div><div><br></div><div> <br><br><div class="gmail_quote">On Sat, Apr 18, 2009 at 6:20 AM, Juerg Hutter <span dir="ltr"><<a href="mailto:hut...@pci.uzh.ch">hut...@pci.uzh.ch</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;"><div class="im"><br>
<br>
<br>
> Dear CP2k users,<br>
> I am learning to run an electron dynamics calculation with fixed nuclei<br>
> position. In the input file, I guess the Section<br>
> &CP2K_INPUT/&FORCE_EVAL/&DFT/&REAL_TIME_PROPAGATION is the place. Is there<br>
> any way to specify a time step for the wave function propagation, or the<br>
> time step is automatically determined by the variable %EXP_ACCURACY%?<br>
><br>
</div>Hi<br>
<br>
The timestep is set in MOTION%MD%TIMESTEP<br>
<br>
See also the inputs in the regression testing directory<br>
cp2k/tests/QS/regtest-rtp<br>
<br>
regards<br>
<br>
Juerg Hutter<br>
<br>
PS: This is a very new feature, carefully test your runs and we would be<br>
happy to get feedback.<br>
<div class="im"><br>
> Thanks.<br>
><br>
> Hanning Chen<br>
><br>
> Department of Chemistry<br>
> Northwestern University<br>
> Evanston, IL 60208<br>
><br>
> ><br>
><br>
<br>
</div><br>
</blockquote></div><br></div>