Hi Matthias, <div> </div><div>Many thanks for your suggestion. </div><div>It is really basis set dependent. I've tried again with most basis sets I have, only the MOLOPT-SR types can give me some "right physics" results.</div> <div>Btw, could you please tell me some technical papers about this problem?</div><div> </div><div>Thanks again, </div><div>Manh</div><div> </div><meta http-equiv="content-type" content="text/html; charset=utf-8"><div> <div class="gmail_quote">On Sun, Dec 5, 2010 at 3:06 PM, Matthias Krack <<a href="mailto:matthia...@psi.ch">matthia...@psi.ch</a>> wrote: <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;"> Hi Manh, population analyses are known to be very sensible to the basis set choice. I guess also in this case the TZV2P MOLOPT basis set is the reason for the unexpected result. The MOLOPT basis sets include relatively small exponents which might cause linear dependencies in the overlap matrix. The Lowdin population analysis requires the calculation of S**(1/2) which involves (in CP2K) a diagonalisation of the overlap matrix. However, an ill-conditioned overlap matrix will most likely result in a weird charge partition. Therefore, I would suggest to employ different basis sets, e.g. the MOLOPT-SR or the DZVP/ TZVP basis sets. Best, Matthias <div><div></div><div class="h5"> On 3 Dez., 17:46, Manh <<a href="mailto:manht...@gmail.com">manht...@gmail.com</a>> wrote: > Hello everyone, > > I calculate the Mulliken and Lowdin charges for a water molecule. The > results are: > > ---------------------------------------------------------------- > > MULLIKEN POPULATION ANALYSIS > > # Atom Element Kind Atomic population Net charge > 1 H 1 0.780838 0.219162 > 2 H 1 0.780981 0.219019 > 3 O 2 6.438182 -0.438182 > # Total charge 8.000000 0.000000 > > LOWDIN POPULATION ANALYSIS > > # Atom Element Kind Atomic population Net charge > 1 H 1 1.210449 -0.210449 > 2 H 1 1.210473 -0.210473 > 3 O 2 5.579079 0.420921 > # Total charge 8.000000 0.000000 > > -------------------------------------------------------- > > I think that the Lowdin analysis has some problem. > > Here is my input: > > --------------------------------------------- > > &FORCE_EVAL > METHOD Quickstep > &DFT > BASIS_SET_FILE_NAME ../../../BASIS_MOLOPT > POTENTIAL_FILE_NAME ../../../GTH_POTENTIALS > RESTART_FILE_NAME H2O-RESTART.wfn.wfn > &QS > METHOD GPW > EXTRAPOLATION ASPC > EXTRAPOLATION_ORDER 3 > &END QS > &MGRID > CUTOFF 400 > NGRIDS 5 > &END > &SCF > MAX_SCF 20 > SCF_GUESS RESTART > EPS_SCF 1.0E-6 > &OT > PRECONDITIONER FULL_SINGLE_INVERSE > MINIMIZER CG > &END > &OUTER_SCF > MAX_SCF 10 > EPS_SCF 1.0E-6 > &END > &PRINT > &RESTART > &EACH > QS_SCF 0 > GEO_OPT 2 > &END > ADD_LAST NUMERIC > FILENAME RESTART.wfn > &END > &RESTART_HISTORY OFF > &END > &END > &END SCF > > &XC > &XC_FUNCTIONAL PBE > &END XC_FUNCTIONAL > &END XC > > &PRINT > &MULLIKEN > &END MULLIKEN > > &LOWDIN > &END LOWDIN > &END > > &END DFT > &SUBSYS > &CELL > ABC [angstrom] 30 30 30 > &END CELL > &TOPOLOGY > COORD_FILE_NAME water.xyz > COORDINATE xyz > &END > > &KIND O > BASIS_SET TZV2P-MOLOPT-GTH > POTENTIAL GTH-PBE-q6 > &END KIND > &KIND H > BASIS_SET TZV2P-MOLOPT-GTH > POTENTIAL GTH-PBE-q1 > &END KIND > > &END SUBSYS > > &END FORCE_EVAL > > &GLOBAL > PRINT_LEVEL LOW > PROJECT H2O > RUN_TYPE ENERGY > WALLTIME 285000 > &END GLOBAL > > &MOTION > &GEO_OPT > MAX_ITER 5000 > MAX_FORCE 0.00010 > OPTIMIZER BFGS > &BFGS > &END > &END > &END > > ------------------------------------------------ > the water.xyz file is > ---------------------------------------- > 3 > water > H 8.4351137971 5.1620318678 12.1510352783 > H 8.5534822942 3.6347695084 12.1803075532 > O 7.9011326518 4.3528437024 12.1874564118 > ------------------------------------------------ > > Regards, > Manh -- You received this message because you are subscribed to the Google Groups "cp2k" group. To post to this group, send email to <a href="mailto:cp...@googlegroups.com">cp...@googlegroups.com</a>. To unsubscribe from this group, send email to <a href="mailto:cp2k%2Bun...@googlegroups.com">cp2k+uns...@googlegroups.com</a>. For more options, visit this group at <a href="http://groups.google.com/group/cp2k?hl=en" target="_blank">http://groups.google.com/group/cp2k?hl=en</a>. </div></div></blockquote></div> </div>