I ran a static calculation on a MoS2 monolayer using the following OT settings:<div><br></div><div>&OT</div><div> algorithm irac</div><div> rotation true</div><div> preconditioner full_single_inverse</div><div> energy_gap -1</div><div> minimizer diis</div><div>&end</div><div><br></div><div>I use these settings by default because, while full all and strict may result in faster convergence for some systems. Allowing for rotations in general is more robust in my experience. For the first time, however, an issue has come up when I ran the calculation it converged with a band gap of -1.66eV and a final energy of -3195.190101039151614 Ha. On a hunch, I reran it using FULL_ALL and Algorithm strict and this gave a band gap of 1.62eV and an energy of -3195.310508319606470 Ha. The energy difference is almost exactly two times the "correct" band gap so I looked at the MOs and found this </div><div><br></div><div><p> -0.05112749 -0.04139593 -0.04138555 0.02140865</p>
<p> Fermi Energy [eV] : 0.582559</p>
<p><br></p>
<p> Lowest Eigenvalues of the unoccupied subspace spin 1</p>
<p> -----------------------------------------------------</p>
<p> OT| Eigensolver reached convergence in 1 iterations</p>
<p> -0.03958992 0.01966848 0.03075463 0.03075623</p>
<br></div><div>By the looks of it, the OT solver with IRAC converged to some kind of excited state. I have not seen this behavior before so I am wondering:</div><div><br></div><div>(1) Is this an expected consequence of not enforcing strict orthogonality?</div><div>(2) Is there something to do to fix the issue without giving up the ability to have Rotation True for fractional occupations?</div><div><br></div><div>I have attached the output file for the two calculations.</div><div><br></div><div>Best</div><div>Nick</div><div><br></div><div><br></div><div><br></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 on the web visit <a href="https://groups.google.com/d/msgid/cp2k/96a1b582-6c84-4842-b19e-dd503d2f4745n%40googlegroups.com?utm_medium=email&utm_source=footer">https://groups.google.com/d/msgid/cp2k/96a1b582-6c84-4842-b19e-dd503d2f4745n%40googlegroups.com</a>.<br />