Hello,<div><br></div><div>I kind of did a lot of digging and came up with the following general inputs for PBE and PBE0 for organic, non-periodic systmes.</div><div>What do you think ?</div><div><br></div><div>Can I optimize them further ?</div><div><br></div><div>I guess the system is too small (Li near cyclohexadiene, 14 atoms) to properly test the cutoff in</div><div> &INTERACTION_POTENTIAL<br> POTENTIAL_TYPE TRUNCATED !TRUNCATED 1/r (Coulomb) POTENTIAL FOR SPEED UP<br> CUTOFF_RADIUS 6.0 !SHOULD BE < HALF THE CELL<br> T_C_G_DATA /home/lsd/cp2k_data/t_c_g.dat !DATA FOR THE EVAL OF TRUNCATED GAMMA FUNC<br> &END INTERACTION_POTENTIAL<br></div><div><br></div><div>and</div><div><br></div><div> !&PBE_HOLE_T_C_LR<br> ! CUTOFF_RADIUS 6.0<br> ! SCALE_X 0.25<br> !&END PBE_HOLE_T_C_LR<br></div><div><br></div><div>but quite frankly even on a larger and periodic system (graphene) there seems to be no speed up.</div><div><br></div><div>The EPS_SCHWARZ is set to 1.0E-6 either way.</div><div><br></div><div>Best Regards,</div><div>Dobromir<br><br></div><div class="gmail_quote"><div dir="auto" class="gmail_attr">On Tuesday, December 20, 2022 at 3:56:32 PM UTC+2 DMT wrote:<br/></div><blockquote class="gmail_quote" style="margin: 0 0 0 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">Hello,<div><br></div><div>I am interested in organic molecules, which are to be eventually modeled over surfaces, in solvents in reactions and also in molecular crystals. I am new to the software, so I have questions regarding choice of QS METHOD, BASIS_SET, POTENTIALS and some settings. I have found a few input files, searched the group (some google) and the manual, but there are quite a few things I am not familiar with.</div><div><br></div><div>For basis sets I search the cp2k/data directory of the source code and I search the Basis Set Exchange web site. </div><div>- are there other ways of finding basis sets +/- potentials ?</div><div><br></div><div>So far I am left with the impression that all GTH-POTENTIALs are only available optimized for certain DFT functionals and are hence unsuitable with other functionals ?</div><div>I assume DZVP-GTH basis has a double-zeta valence shell + polarizations ? Does it also exist with diffusion functions (let's say I need density a little further away - an anion, a complex, etc.) ? How good is it compared to TZVP (all-electron) or def2-TZVP ? Or compared to 6-31g* or cc-pVDZ ? Which basis sets for CP2K would be like 6-31g*, TZVP, def2-TZVP and cc-pVDZ ... and why not cc-pVTZ for each of the following QS METHODs: DFTB, GAPW, GAPW_XC, GPW, LRIGPW, RIGPW, XTB.</div><div><br></div><div>Which CP2K basis sets have diffusion shells ?</div><div><br></div><div>!> I cannot deduce the shells of basis sets by reading the basis set files, so is there a way to actually see how many s, p, d and/or f shells does a CP2K basis set have ? I can guess whether there are polarizations or not based on the element and type of shells, but I cannot guess the same for diffusion orbitals, as they are of the same type as the covalent shell orbitals.</div><div><br></div><div>!> What QS METHOD is good for organics and not just for semiconductors or metals ? Which are faster and which are slower ? Which BASIS_SET and/or POTENTIAL should I use for good / scientific / publication quality calculations ?</div><div><br></div><div>Can I use one QS METHOD for a central molecule and another QS METHOD and BASIS_SET for surrounding solvent molecules, without QM/MM ? Or one qs method and basis for a central molecule and another method + basis combination for a surface (graphene or semi-conductor or metal) ?</div><div><br></div><div>Is it a good idea to calculate molecular crystals with qs method + basis different from those for single molecules / complexes.</div><div><br></div><div>The way I understand it, plane wave basis sets may not be the best for organics ?</div><div><br></div><div>!> What type of dispersion is most appropriate for organic molecules / organo-metallic complexes, with or without a solvent. And for molecular crystals ?</div><div><br></div><div>!> I see that there are some new DFT types in CP2K, beyond the kind of DFT that Gaussian does. Should I pick any of them and for which kind of systems ?</div><div><br></div><div>Best Regards,</div><div>Dobromir</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 on the web visit <a href="https://groups.google.com/d/msgid/cp2k/07fff3e3-0a3b-4e86-b6f1-d364072f191cn%40googlegroups.com?utm_medium=email&utm_source=footer">https://groups.google.com/d/msgid/cp2k/07fff3e3-0a3b-4e86-b6f1-d364072f191cn%40googlegroups.com</a>.<br />