<div>Hi,</div><div><br /></div><div>I have doubt if your system is fit for OT method; it looks like to be metal-based and may require smearing.</div><br /><div class="gmail_quote"><div dir="auto" class="gmail_attr">在2026年4月9日星期四 UTC+8 19:18:49<Yinyu chan> 写道:<br/></div><blockquote class="gmail_quote" style="margin: 0 0 0 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">Dear CP2K community,<div><br></div><div>I'am a new user of CP2K(v2026.1) and I'm running an AIMD simulation of a metal-electrolyte interface (Au-Pd alloy/NaCl(aq)) with electric field.<br></div><div><br></div><div>However, I encountered an issue where the SCF starts fast but then becomes slow, as shown in the *ener file. But when I restarted the AIMD at steps 5000(SCF_GUESS RESTART) and 5378(SCF_GUESS ATOMIC), the SCF convergence starts fast again, but then becomes slow once more. I would like to know whether this phenomenon is normal, and how to keep the SCF convergence fast during the AIMD. </div><div><br></div><div>I've attached the inputs and outputs for reference. I would appreciate it if anyone could offer suggestions on this issue.</div><div><br></div><div>Thank you for your time and support.</div><div><br></div><div>Best regards,</div><div>Yinyu Chan</div></blockquote></div>
<p></p>
-- <br />
You received this message because you are subscribed to the Google Groups "cp2k" group.<br />
To unsubscribe from this group and stop receiving emails from it, send an email to <a href="mailto:cp2k+unsubscribe@googlegroups.com">cp2k+unsubscribe@googlegroups.com</a>.<br />
To view this discussion visit <a href="https://groups.google.com/d/msgid/cp2k/7af7726b-8138-4aa5-8a36-9a68dfd0f13en%40googlegroups.com?utm_medium=email&utm_source=footer">https://groups.google.com/d/msgid/cp2k/7af7726b-8138-4aa5-8a36-9a68dfd0f13en%40googlegroups.com</a>.<br />