<div dir="ltr">Hello CP2K users<div><br></div><div>Sorry about these basic questions. So, here the system was periodic in 3D. However if I have a 2D (surface slab) periodic system do I still have to set FULLY_PERIODIC keyword to True. What does "rotational motions are not projected out" mean? What is a translational invariant system (is any periodic system translationally invariant?)? When should one use POISSON_SOLVER block? Also how should we interpret the frequencies in "Snippet output frequency run:" section.</div><div><br></div><div>So I tried to reproduce your results (just changed the basis set and the potentials to DZVP-MOLOPT-GTH and GTH-PBE-q4 resp) and for 8 atom system, I get the following 21 normal modes while I should Ideally get 18 normal modes for 8 atoms (3N-6). So should the three (21-18) extra normal modes be equal to 0 ideally? Now, the normal modes are so close that I don't know which 18 I should consider as the actual normal modes. </div><div><br></div><div><div> 250.512753</div><div> 319.954005</div><div> 399.465765</div><div> 405.037651</div><div> 422.843253</div><div> 435.761877</div><div> 1097.038334</div><div> 1117.184070</div><div> 1124.201100</div><div> 1127.137991</div><div> 1128.074863</div><div> 1128.868198</div><div> 1133.580152</div><div> 1136.804637</div><div> 1141.857240</div><div> 1142.663552</div><div> 1155.931539</div><div> 1163.319124</div><div> 1617.679412</div><div> 1624.770294</div><div> 1627.580761</div></div><div><br></div><div>I am thinking of using the following input file for my 2D surface slab for frequency calculations.</div><div><br></div><div><div>&VIBRATIONAL_ANALYSIS</div><div> FULLY_PERIODIC T </div><div> DX 0.01</div><div> NPROC_REP 8</div><div> &MODE_SELECTIVE</div><div> RESTART_FILE_NAME 4x4-VIBRATIONS-1.mol</div><div> &INVOLVED_ATOMS</div><div> INVOLVED_ATOMS 65 66</div><div> &END INVOLVED_ATOMS</div><div> ATOMS 65 66 </div><div> &END MODE_SELECTIVE </div><div>&END VIBRATIONAL_ANALYSIS</div><div><br></div><div>&FORCE_EVAL</div><div> METHOD Quickstep</div><div> STRESS_TENSOR ANALYTICAL</div><div> &DFT</div><div> BASIS_SET_FILE_NAME ./GTH_BASIS_SETS</div><div> POTENTIAL_FILE_NAME ./GTH_POTENTIALS</div><div> WFN_RESTART_FILE_NAME 4x4-RESTART.wfn </div><div> &MGRID</div><div> CUTOFF 480</div><div> NGRIDS 5</div><div> &END MGRID</div><div> &QS</div><div> WF_INTERPOLATION ASPC</div><div> EXTRAPOLATION_ORDER 3</div><div> &END QS</div><div> &SCF</div><div> EPS_SCF 1.E-6</div><div> SCF_GUESS RESTART</div><div> MAX_SCF 100</div><div> &OUTER_SCF</div><div> EPS_SCF 1.E-6</div><div> MAX_SCF 20</div><div> &END OUTER_SCF</div><div> &OT T</div><div> PRECONDITIONER FULL_SINGLE_INVERSE</div><div> MINIMIZER CG</div><div> LINESEARCH 3PNT</div><div> &END OT</div><div> &END SCF</div><div> &XC</div><div> &XC_FUNCTIONAL PBE</div><div> &END XC_FUNCTIONAL</div><div> &END XC</div><div> &PRINT</div><div> &END PRINT</div><div> &END DFT</div><div> &SUBSYS</div><div> &CELL</div><div>A 11.2005714139949148 0.0000000000000000 0.0000000000000000</div><div>B 5.6002857069974574 9.6999793814213842 0.0000000000000000</div><div>C 0.0000000000000000 0.0000000000000000 26.8589211979727551</div><div>&END CELL</div><div>&COORD</div><div>Pd -0.0000061359210358 -0.0000222840946560 9.9880132312990213</div><div>Pd 2.8000908510293288 0.0000092115121854 9.9879695072021697</div><div>Pd 5.6002734938253287 -0.0000045647646310 9.9879245033445034</div><div>Pd 8.4004340473737855 -0.0000362027794058 9.9879677511115652</div><div>Pd 1.4000634784708132 2.4249744889860554 9.9880132682562639</div><div>Pd 4.2001617373702826 2.4250029632593546 9.9879685687298245</div><div>Pd 7.0003471689337546 2.4249881536022277 9.9879248470317670</div><div>Pd 9.8005050809925134 2.4249618882564654 9.9879671972427317</div><div>Pd 2.8001369378977943 4.8499691363190243 9.9880148797577206</div><div>Pd 5.6002312352107486 4.8499974956142475 9.9879690366522222</div><div>Pd 8.4004168005198139 4.8499832866277810 9.9879251120231469</div><div>Pd 11.2005770533358611 4.8499548963592893 9.9879640638995539</div><div>Pd 4.2002068480680155 7.2749634825025922 9.9880166024601085</div><div>Pd 7.0003058990547880 7.2749930185934337 9.9879692328558018</div><div>Pd 9.8004884696248684 7.2749791367862375 9.9879232637271596</div><div>Pd 12.6006459666106423 7.2749492464615217 9.9879647613984162</div><div>Pd 1.4002245368808004 0.8084157041067350 12.3012388137054103</div><div>Pd 4.2003558131593932 0.8084218859192664 12.3012418639169621</div><div>Pd 7.0005061178333099 0.8084332483716472 12.3012622163864087</div><div>Pd 9.8006578762588088 0.8084217487769948 12.3012772146045251</div><div>Pd 2.8002965326028924 3.2334107313122149 12.3012392548407323</div><div>Pd 5.6004272909567518 3.2334165217953741 12.3012418245909636</div><div>Pd 8.4005772468615536 3.2334291460788043 12.3012628426338733</div><div>Pd 11.2007297065823632 3.2334147570161607 12.3012750335016499</div><div>Pd 4.2003688965050028 5.6584053402129628 12.3012398254336990</div><div>Pd 7.0004993266283773 5.6584119088267411 12.3012414536964982</div><div>Pd 9.8006498524563774 5.6584251222740383 12.3012626779581495</div><div>Pd 12.6008005523876250 5.6584086848353445 12.3012752704691177</div><div>Pd 5.6004393945982400 8.0834002167787382 12.3012396637977126</div><div>Pd 8.4005708513076502 8.0834072604243818 12.3012409646385859</div><div>Pd 11.2007215403971241 8.0834195185537538 12.3012615384488573</div><div>Pd 14.0008718167863044 8.0834038420466356 12.3012780910986823</div><div>Pd 0.0000000000000000 1.6166632302368973 14.5726141319818385</div><div>Pd 2.8001428534987292 1.6166632302368973 14.5726141319818385</div><div>Pd 5.6002857069974583 1.6166632302368973 14.5726141319818385</div><div>Pd 8.4004285604961844 1.6166632302368973 14.5726141319818385</div><div>Pd 1.4000714267493641 4.0416580755922427 14.5726141319818385</div><div>Pd 4.2002142802480922 4.0416580755922427 14.5726141319818385</div><div>Pd 7.0003571337468218 4.0416580755922427 14.5726141319818385</div><div>Pd 9.8004999872455478 4.0416580755922427 14.5726141319818385</div><div>Pd 2.8001428534987287 6.4666529209475891 14.5726141319818385</div><div>Pd 5.6002857069974574 6.4666529209475891 14.5726141319818385</div><div>Pd 8.4004285604961844 6.4666529209475891 14.5726141319818385</div><div>Pd 11.2005714139949148 6.4666529209475891 14.5726141319818385</div><div>Pd 4.2002142802480922 8.8916477663029347 14.5726141319818385</div><div>Pd 7.0003571337468218 8.8916477663029347 14.5726141319818385</div><div>Pd 9.8004999872455496 8.8916477663029347 14.5726141319818385</div><div>Pd 12.6006428407442783 8.8916477663029347 14.5726141319818385</div><div>Pd 0.0000000000000000 0.0000000000000000 16.8589211979727551</div><div>Pd 2.8001428534987287 0.0000000000000000 16.8589211979727551</div><div>Pd 5.6002857069974574 0.0000000000000000 16.8589211979727551</div><div>Pd 8.4004285604961844 0.0000000000000000 16.8589211979727551</div><div>Pd 1.4000714267493644 2.4249948453553460 16.8589211979727551</div><div>Pd 4.2002142802480922 2.4249948453553460 16.8589211979727551</div><div>Pd 7.0003571337468218 2.4249948453553460 16.8589211979727551</div><div>Pd 9.8004999872455478 2.4249948453553460 16.8589211979727551</div><div>Pd 2.8001428534987287 4.8499896907106921 16.8589211979727551</div><div>Pd 5.6002857069974574 4.8499896907106921 16.8589211979727551</div><div>Pd 8.4004285604961844 4.8499896907106921 16.8589211979727551</div><div>Pd 11.2005714139949148 4.8499896907106921 16.8589211979727551</div><div>Pd 4.2002142802480922 7.2749845360660377 16.8589211979727551</div><div>Pd 7.0003571337468218 7.2749845360660377 16.8589211979727551</div><div>Pd 9.8004999872455478 7.2749845360660377 16.8589211979727551</div><div>Pd 12.6006428407442765 7.2749845360660377 16.8589211979727551</div><div>C 2.7695479032678416 1.5018869319588131 8.8099561225681224</div><div>O 2.7497835812269869 1.4846637516465382 7.6749430013606608</div><div>&END COORD</div><div> &KIND Pd </div><div> BASIS_SET DZVP-MOLOPT-SR-GTH </div><div> POTENTIAL GTH-PBE-q18</div><div> &END KIND</div><div> &KIND O</div><div> BASIS_SET DZVP-MOLOPT-GTH</div><div> POTENTIAL GTH-PBE-q6</div><div> &END KIND</div><div> &KIND C</div><div> BASIS_SET DZVP-MOLOPT-GTH</div><div> POTENTIAL GTH-PBE-q4</div><div> &END KIND</div><div> &END SUBSYS</div><div>&END FORCE_EVAL</div><div>&GLOBAL</div><div> PROJECT 4x4</div><div> RUN_TYPE VIBRATIONAL_ANALYSIS</div><div> PRINT_LEVEL LOW</div><div>&END GLOBAL</div><div> &MOTION</div><div> &GEO_OPT</div><div> MAX_ITER 200 </div><div> MAX_FORCE 0.0009725</div><div> OPTIMIZER BFGS </div><div> &END GEO_OPT</div><div> &CONSTRAINT</div><div> &FIXED_ATOMS</div><div> LIST 33..64</div><div> &END FIXED_ATOMS</div><div> &END CONSTRAINT</div><div> &END MOTION</div></div><div><br></div><div>Thank you for any help.</div><div><br></div><div><br></div><div><br>On Friday, April 9, 2010 11:35:20 AM UTC-4, An Ghysels wrote:<blockquote class="gmail_quote" style="margin: 0;margin-left: 0.8ex;border-left: 1px #ccc solid;padding-left: 1ex;">Dear forum users,<p>I am calculating frequencies in periodic structures, from which I'd<br>like to derive thermodynamic quantities like the entropy etc.<br>Therefore I use the VIBRATIONAL_ANALYSIS keyword, where I have put the<br>FULLY_PERIODIC keyword to True, such that rotational motions are not<br>projected out.</p><p>In principle, the second derivatives matrix should have 3 zero<br>eigenvalues (even in non-optimized systems), because the system is<br>translational invariant. However, this is not the case in my<br>calculations. Even in a small test case like diamond with a cubic cell<br>containing 8 atoms, the "zero" frequencies deviate largely from zero.</p><p>(Between brackets: I'm much bothered by this bad performance, because<br>it signals that the numerical accuracy of the frequencies is poor. I<br>aim at studying larger unit cells (eg with 100 to 200 QM atoms) where<br>I surely will encounter floppy modes, which are even more sensible to<br>numerical errors. And those floppy modes play an essential role in the<br>partition function/thermodynamics. Therefore, I wish to improve the<br>accuracy of the calculated frequencies.)</p><p>So my questions are mainly: why is the accuracy so bad? Is this<br>directly related to the eggbox effect, and if so, how to remedy it? Is<br>it something with interpolation grids? And what to do to improve it?</p><p>I tried decreasing the EPS_DEFAULT (down to 10e-16), increasing the<br>E_CUTOFF (up to 1400 Rydberg), increasing the REL_CUTOFF accordingly,<br>adding the keyword COMMENSURATE, changing the interpolation grid (to<br>spline3), but the improvements still do not get me close to "zero"<br>frequencies (still of the order -70 to 50 cm-1).</p><p>Looking forward to your comments and remarks,</p><p>An</p><p>Below: Information on version, input geometry optimization, input<br>frequency run, snippets of output<br>=====<br>Version:<br> CP2K| version string: CP2K version 2.0.1<br>(Development Version)<br> CP2K| is freely available from <a href="http://cp2k.berlios.de/" target="_blank" onmousedown="this.href='http://www.google.com/url?q\75http%3A%2F%2Fcp2k.berlios.de%2F\46sa\75D\46sntz\0751\46usg\75AFQjCNG8a04cBV6g-qX2gL0GLMujwBbc_g';return true;" onclick="this.href='http://www.google.com/url?q\75http%3A%2F%2Fcp2k.berlios.de%2F\46sa\75D\46sntz\0751\46usg\75AFQjCNG8a04cBV6g-qX2gL0GLMujwBbc_g';return true;">http://cp2k.berlios.de/</a><br> CP2K| Program compiled at<br>Wed_Sep__9_15:48:59_CEST_2009<br> CP2K| Program compiled<br>on gengar1<br> CP2K| Program compiled for Linux-<br>x86-64-intel<br>=====<br>Input geomery optimization<br>&GLOBAL<br> PRINT_LEVEL high<br> PROJECT_NAME vib<br> RUN_TYPE GEO_OPT<br> PREFERRED_FFT_LIBRARY FFTW<br>&END GLOBAL</p><p>&MOTION<br> &GEO_OPT<br> OPTIMIZER BFGS<br> MAX_ITER 300<br> &END GEO_OPT<br>&END MOTION</p><p>&VIBRATIONAL_ANALYSIS<br> FULLY_PERIODIC T<br>&END VIBRATIONAL_ANALYSIS</p><p>&FORCE_EVAL</p><p> METHOD QS<br> &DFT<br> BASIS_SET_FILE_NAME GTH_BASIS_SETS<br> POTENTIAL_FILE_NAME GTH_POTENTIALS<br> &MGRID<br> CUTOFF 100<br> REL_CUTOFF 23<br> COMMENSURATE<br> &END MGRID<br> &QS<br> EPS_DEFAULT 1.0E-10<br> &END QS<br> &SCF<br> EPS_SCF 1.0E-5<br> MAX_SCF 100<br> SCF_GUESS RESTART<br> &OUTER_SCF<br> EPS_SCF 1.0E-5<br> MAX_SCF 20<br> &END<br> &OT<br> ALGORITHM IRAC<br> MINIMIZER CG<br> PRECONDITIONER FULL_ALL<br> ENERGY_GAP 0.001<br> &END OT<br> &PRINT<br> &RESTART<br> BACKUP_COPIES 0<br> &END RESTART<br> &END PRINT<br> &END SCF<br> &XC<br> &XC_FUNCTIONAL PBE<br> &END XC_FUNCTIONAL<br> &END XC<br> &POISSON<br> POISSON_SOLVER PERIODIC<br> PERIODIC XYZ<br> &EWALD<br> EWALD_TYPE EWALD<br> ALPHA 0.3<br> GMAX 11<br> O_SPLINE 6<br> &END EWALD<br> &END POISSON<br> &PRINT<br> &END PRINT<br> &END DFT</p><p> &SUBSYS<br> &CELL<br> ABC 3.567 3.567 3.567<br> PERIODIC XYZ<br> &END CELL<br> &COORD<br>C -0.066491 -0.063731 -0.038851<br>C 0.019572 1.759040 1.767428<br>C 1.756129 0.023158 1.769616<br>C 1.767376 1.770784 0.004363<br>C 0.841267 0.843123 0.828198<br>C 2.663349 0.903043 2.661920<br>C 0.899007 2.665284 2.661387<br>C 2.644659 2.648722 0.926362<br> &END COORD<br> &TOPOLOGY<br> &GENERATE<br> CREATE_MOLECULES<br> &END GENERATE<br> &END TOPOLOGY</p><p> &KIND C<br> BASIS_SET DZVP-GTH<br> POTENTIAL GTH-PBE-q4<br> &END KIND</p><p> &END SUBSYS<br>&END FORCE_EVAL<br>=====<br>Input Frequency run<br>&GLOBAL<br> PRINT_LEVEL high<br> PROJECT_NAME vib<br> RUN_TYPE VIBRATIONAL_ANALYSIS<br> PREFERRED_FFT_LIBRARY FFTW<br>&END GLOBAL</p><p>&MOTION<br> &GEO_OPT<br> OPTIMIZER BFGS<br> MAX_ITER 300<br> &END GEO_OPT<br>&END MOTION</p><p>&VIBRATIONAL_ANALYSIS<br> FULLY_PERIODIC T<br>&END VIBRATIONAL_ANALYSIS</p><p>&FORCE_EVAL</p><p> METHOD QS<br> &DFT<br> BASIS_SET_FILE_NAME GTH_BASIS_SETS<br> POTENTIAL_FILE_NAME GTH_POTENTIALS<br> &MGRID<br> CUTOFF 100<br> REL_CUTOFF 23<br> COMMENSURATE<br> &END MGRID<br> &QS<br> EPS_DEFAULT 1.0E-10<br> &END QS<br> &SCF<br> EPS_SCF 1.0E-5<br> MAX_SCF 100<br> SCF_GUESS RESTART<br> &OUTER_SCF<br> EPS_SCF 1.0E-5<br> MAX_SCF 20<br> &END<br> &OT<br> ALGORITHM IRAC<br> MINIMIZER CG<br> PRECONDITIONER FULL_ALL<br> ENERGY_GAP 0.001<br> &END OT<br> &PRINT<br> &RESTART<br> BACKUP_COPIES 0<br> &END RESTART<br> &END PRINT<br> &END SCF<br> &XC<br> &XC_FUNCTIONAL PBE<br> &END XC_FUNCTIONAL<br> &END XC<br> &POISSON<br> POISSON_SOLVER PERIODIC<br> PERIODIC XYZ<br> &EWALD<br> EWALD_TYPE EWALD<br> ALPHA 0.3<br> GMAX 11<br> O_SPLINE 6<br> &END EWALD<br> &END POISSON<br> &PRINT<br> &END PRINT<br> &END DFT<br> &SUBSYS<br> &CELL<br> ABC 3.567 3.567 3.567<br> PERIODIC XYZ<br> &END CELL<br> &COORD<br> C -0.0393192769 -0.0522499977 -0.0007602780<br> C 0.0299677090 1.7769206120 1.7824574256<br> C 1.7642084626 0.0471181783 1.7911208859<br> C 1.7938452241 1.7853211459 0.0053883203<br> C 0.8664406756 0.8762954420 0.8435833733<br> C 2.6939879368 0.9029246308 2.6806928814<br> C 0.9092781460 2.6785011492 2.6771532650<br> C 2.6481574040 2.6661172907 0.9483093511<br> &END COORD<br> &TOPOLOGY<br> &GENERATE<br> CREATE_MOLECULES<br> &END GENERATE<br> &END TOPOLOGY</p><p> &KIND C<br> BASIS_SET DZVP-GTH<br> POTENTIAL GTH-PBE-q4<br> &END KIND</p><p> &END SUBSYS<br>&END FORCE_EVAL<br>=====<br>Snippet output frequency run:<br>If I look at the values with the lowest absolute value and I convert<br>them to cm-1, it is poor accuracy:</p><p> VIB| Cartesian Low frequencies ----0.44072 -0.10876E-01<br>0.12458 0.69806<br> VIB| Cartesian Low frequencies --- 0.71490 1.4715<br>2.0898 3.8120<br> VIB| Cartesian Low frequencies --- 5.8007<br> VIB| Frequencies after removal of the rotations and translations<br> VIB| Internal Low frequencies ----0.14095 0.50894<br>1.4072 2.0559<br> VIB| Internal Low frequencies --- 3.6794 5.7900<br>25.027 25.504<br> VIB| Internal Low frequencies --- 25.855 25.997<br>26.180 26.684<br> VIB| Internal Low frequencies --- 26.818 26.970<br>27.202 27.461<br> VIB| Internal Low frequencies --- 27.643 28.937<br>54.059 54.351<br> VIB| Internal Low frequencies --- 54.817</p><p> VIB| NORMAL MODES - CARTESIAN DISPLACEMENTS<br> VIB|<br> VIB| 1<br>2 3<br> VIB|Frequency (cm^-1) -82.399299 156.573040<br>260.352529<br>...<br> <br>------------------------------<wbr>------------------------------<wbr>-------------------<br> ---- MULTIGRID<br>INFO ----<br> <br>------------------------------<wbr>------------------------------<wbr>-------------------<br> count for grid 1: 868 cutoff<br>[a.u.] 50.00<br> count for grid 2: 276 cutoff<br>[a.u.] 12.50<br> count for grid 3: 80 cutoff<br>[a.u.] 3.12<br> count for grid 4: 0 cutoff<br>[a.u.] 0.78<br> total gridlevel count : 1224</p><p></p><p></p><p></p><p></p><p></p><p></p><p></p><p></p><p></p><p></p><p></p><p></p><p></p><p></p><p></p><p></p><p></p><p></p><p></p><p></p><p></p><p></p><p></p><p></p></blockquote></div></div>