Re: [CP2K:5232] Stress Tensor forcing SCF Wavefunction Optimization
1,019 views
Skip to first unread message
hut...@chem.uzh.ch
May 3, 2014, 10:33:44 AM5/3/14
to cp...@googlegroups.com
Hi
I cannot see any problematic behavior when running a similar
system (you don't give us your exact input).
Could it be that the additional SCF iterations you see, are just the
finite difference calculations from the numerical stress evaluation?
BTW why not using the analytic stress evaluation?
Also, if you are using DFTB and periodic boundary conditions I would
strongly advice to use EWALD!
regards
Juerg
--------------------------------------------------------------
Juerg Hutter Phone : ++41 44 635 4491
Institut für Chemie FAX : ++41 44 635 6838
Universität Zürich E-mail: hut...@chem.uzh.ch
Winterthurerstrasse 190
CH-8057 Zürich, Switzerland
---------------------------------------------------------------
I cannot see any problematic behavior when running a similar
system (you don't give us your exact input).
Could it be that the additional SCF iterations you see, are just the
finite difference calculations from the numerical stress evaluation?
BTW why not using the analytic stress evaluation?
Also, if you are using DFTB and periodic boundary conditions I would
strongly advice to use EWALD!
regards
Juerg
--------------------------------------------------------------
Juerg Hutter Phone : ++41 44 635 4491
Institut für Chemie FAX : ++41 44 635 6838
Universität Zürich E-mail: hut...@chem.uzh.ch
Winterthurerstrasse 190
CH-8057 Zürich, Switzerland
---------------------------------------------------------------
Rafael Soler-Crespo
May 12, 2014, 11:29:44 PM5/12/14
to
Hi Juerg,
Thanks for your reply. Apparently, this got lost somewhere in space, so I just saw it. I attach my full input at the end of my post.
The system is a sheet of oxidized graphite oxide and I am trying to get an idea of the stress tensor during cell optimization. Based on your post, yes, you are right -- the multiple SCF wavefunction optimizations are just the finite differences to find the stress tensor. I tried using the diagonal analytical keyword but it wouldn't work for some reason. The error was:
*** SCF run NOT converged ***
Core Hamiltonian energy: -514.15564300498284
Repulsive potential energy: 15.42812723286143
Electronic energy: 0.81658900421239
Dispersion energy: 0.00000000000000
Total energy: -497.91092676790907
***************************************************************************
*** 18:26:20 ERRORL2 in qs_dftb_coulomb:build_dftb_coulomb processor 0 ***
*** :: err=-300 condition FAILED at line 271 ***
***************************************************************************
I am unsure why (and trying both on a locally compiled copy and also on a cluster copy didn't work) I got this error. I will also look into using Ewald summations as, yes, my system is both periodic and I am using DFTB. I forgot I turned it off to test some input a long time ago and never returned it to on!
Now I am finding that if I take a snapshot (same atomic coordinates) of the system and recalculate the stress doing a single point (ENERGY_FORCES) calculation it doesn't give me the same stress values. This is probably because the wavefunction is not the same during the cell optimization. Would you agree with my conclusion, and do you have any hints or tips for obtaining accurate results of the stress tensor in CP2K calculations for systems like mine?
Thanks a million for your response and help in advance!
Rafael
===================================
INPUT
&FORCE_EVAL
STRESS_TENSOR DIAGONAL_ANALYTICAL
&DFT
&QS
METHOD DFTB
&DFTB
SELF_CONSISTENT T
DO_EWALD F
DISPERSION F
&PARAMETER
PARAM_FILE_PATH /home/ras536/dftb_parameter_sets/cp2k/scc/
PARAM_FILE_NAME parameter_table
UFF_FORCE_FIELD uff_table
&END PARAMETER
&END DFTB
&END QS
&SCF
SCF_GUESS ATOMIC
MAX_SCF 200
&OT ON
LINESEARCH 3PNT
PRECONDITIONER FULL_SINGLE
&END OT
&END SCF
&END DFT
&SUBSYS
&CELL
# Change with the system coordinates
ABC 21.30 22.14 100
ALPHA_BETA_GAMMA 90 90 90
PERIODIC XYZ
SYMMETRY ORTHORHOMBIC
# End of changes
&END CELL
&COORD
# Paste here system coordinates
C -8.2554109139 -9.8380490000 -1.6801378957
C -6.9603351788 -9.8380490000 -1.0891408567
C -4.2077759278 -9.8380490000 -0.1860590641
C -2.8047748598 -9.8380490000 -0.6483639658
C 0.0305892406 -9.8380490000 -0.7150005959
C 1.5204352714 -9.8380490000 -0.8875680764
C 4.2690168758 -9.8380490000 -0.3760461092
C 5.6664878525 -9.8380490000 -0.4625534107
C 8.4484394028 -9.8380490000 -1.0534139503
C 9.7255781492 -9.8380490000 -1.6742526456
C -8.4934336677 -7.1450588005 -0.7827875533
C -7.0439867935 -7.2373328254 -0.6570825690
C -4.1763859225 -7.2904474044 -0.0793908609
C -2.7886680738 -7.3678348562 0.1494769378
C 0.0178930434 -7.4922386793 0.2233465195
C 1.4362238243 -7.3305334791 0.0153523063
C 4.3072200959 -7.3730543584 -0.0323089425
C 5.7027726205 -7.3779693175 0.0553353058
C 8.5981901881 -7.4077687091 -0.0953368521
C 9.9552305820 -7.2057395822 -0.6606185678
C -8.5331267996 -4.9483020547 0.3406093844
C -7.0724367777 -4.9564599509 0.2657712613
C -4.2254424230 -4.9846974287 0.7071689382
C -2.8480576373 -5.0882055399 1.0069512422
C 0.0704776256 -5.2273656984 1.4576809278
C 1.5257384248 -5.1081956570 1.1763373988
C 4.2793779988 -4.9936842255 0.6343906667
C 5.7071843984 -4.9113081353 0.1905227833
C 8.5552943829 -4.9957621971 0.6352164812
C 9.9189472104 -4.9939128450 0.5829994583
C -8.6002141940 -2.4819289823 0.3741555523
C -7.1563188658 -2.4454452797 0.7116466576
C -4.3069012779 -2.5216286004 1.0850320691
C -2.8791481429 -2.5543815024 1.4631591816
C 0.0388908363 -2.6260056301 1.8026088121
C 1.5435289812 -2.6441866489 1.8122642874
C 4.2647290881 -2.5425531361 0.7709570099
C 5.7360443357 -2.4940389231 0.6446448620
C 8.5099570392 -2.5179613785 0.1399341536
C 9.8721387228 -2.6411693850 -0.4011131502
C -8.4596011734 -0.0818263347 -0.3925756458
C -7.0936995431 -0.0443212834 0.2175360225
C -4.2696487161 0.0001302190 0.8565730368
C -2.8248985105 -0.0843239729 0.6959641828
C 0.1325543046 -0.0578172287 1.2380994318
C 1.5244604184 0.0052842805 1.7886408986
C 4.4064702533 -0.0221312965 1.0209496686
C 5.6121787770 -0.0464685329 0.1157682169
C 8.4534206770 -0.1049637767 -0.4199252111
C 9.8563988959 -0.1437484903 -0.8402668336
C -8.5467454195 2.3532929023 -0.3010984903
C -7.1219726175 2.3874681466 -0.0098605373
C -4.3151789651 2.4829173878 0.7291164593
C -2.8280302271 2.4935603169 0.8494013989
C 0.0400047393 2.5507542999 1.2060003356
C 1.5225558810 2.5683173807 1.4245379389
C 4.5367289495 2.5216544772 1.0636492603
C 5.8374928297 2.4316470815 0.3735257264
C 8.5233373896 2.3386910061 -0.6326970839
C 9.9150883808 2.3913769374 -0.6665557150
C -8.5514445928 4.8621601929 -0.4374486331
C -7.1149147430 4.8158156210 -0.4428978207
C -4.2341671084 4.9415211135 -0.0028013005
C -2.7572522644 4.8691927236 0.0315161057
C 0.0363238454 4.8427722188 0.1550498238
C 1.5073395029 4.7940200263 0.1241373993
C 4.4058959873 5.0335214517 0.4013105731
C 5.8415572798 5.0141215284 -0.0529185737
C 8.5677091315 4.8628974386 -0.9011544249
C 9.9243054687 4.8382533422 -0.3154870541
C -8.5785988586 7.3208566965 -0.8812910916
C -7.0859748318 7.3055543123 -1.1253976473
C -4.1784573301 7.2726438967 -1.0526749672
C -2.7206040161 7.2332650301 -0.8223777357
C 0.0940719511 7.2059048429 -0.6611814788
C 1.4399475863 7.2448802650 -0.4340429073
C 4.2454803743 7.3274491113 -0.5512818488
C 5.6854421612 7.3566202427 -0.9311017225
C 8.5393584777 7.4231842542 -0.7947132985
C 9.8189258148 7.4524790797 0.0017242812
C -8.3732267677 9.6612076320 -0.3589679883
C -7.0051511115 9.7755665711 -0.7041647465
C -4.2222400713 9.7461527022 -0.4047688945
C -2.7261578687 9.8045406793 -0.5267790495
C 0.0911966208 9.7547547599 -1.1007446745
C 1.3476230119 9.7888534697 -0.3018801496
C 4.3439053738 9.8029462979 -0.0739978476
C 5.7135153097 9.8113572154 -0.6010994960
C 8.5289252374 9.8932420524 -0.8478175239
C 9.9871250606 10.0682630023 -0.6589134998
C -9.2300000000 -7.8325637473 -1.7536439658
C -6.2731676559 -8.5557452059 -0.8725724256
C -4.9351808173 -8.5554172499 -0.2197641234
C -2.0704457686 -8.5704609705 -0.3555369196
C -0.6063485758 -8.6588286762 -0.2747244544
C 2.1833164547 -8.5373715834 -0.4947031257
C 3.6193971798 -8.5942803810 -0.2741664115
C 6.3957988846 -8.6471152497 -0.1032894572
C 7.8658522400 -8.6647078304 -0.3936925714
C 10.6500000000 -7.8592320412 -1.7610761829
C -9.2300000000 -6.2665845556 0.2010576526
C -6.3881315619 -6.2875644977 0.3614184746
C -4.8977218596 -6.1428534643 0.2429583275
C -2.1242771475 -6.2711961217 0.7113046670
C -0.6957422206 -6.5292807572 1.1461770651
C 2.1888552860 -6.2126099822 0.6057356014
C 3.5795668231 -6.1875225776 0.3899015044
C 6.4524409374 -6.2053121678 0.3215992249
C 7.8374927331 -6.2197122713 0.3892422550
C 10.6500000000 -5.9729374753 -0.2788794784
C -9.2300000000 -3.7943937956 0.2547955533
C -6.3569092266 -3.6875161719 0.5047847628
C -4.9297888572 -3.7129824661 0.8554043837
C -2.1777099679 -3.8910725962 1.5464153272
C -0.6897478397 -3.9440275381 1.6806052834
C 2.2316486052 -3.8971569891 1.3971083452
C 3.6004895148 -3.8417692248 1.0463822284
C 6.4174662317 -3.6975585666 0.6770607764
C 7.8921126614 -3.6723730235 0.8937721592
C 10.6500000000 -3.8702304200 -0.1015888358
C -9.2300000000 -1.3643971547 -0.3946000942
C -6.4239754622 -1.1829277204 0.5875498637
C -5.0625760414 -1.2025026840 1.2393381572
C -2.0899365660 -1.3591238299 1.0325078484
C -0.6022622833 -1.3704798000 1.2576697101
C 2.2770142953 -1.3021813022 1.7826004797
C 3.6117105941 -1.2456095524 1.1228126380
C 6.4240571173 -1.2672147737 0.4450576208
C 7.8176374693 -1.3058406979 0.0948670931
C 10.6500000000 -1.4089354912 -0.7174852064
C -9.2300000000 1.1690965759 -0.5296955925
C -6.4190432860 1.2200688718 0.2300154157
C -4.9200055446 1.1937960809 0.2227836837
C -2.0699137548 1.2114864538 0.6218006355
C -0.6030423171 1.2243624865 0.9301967541
C 2.2990791703 1.2964366528 1.6652055840
C 3.8083336035 1.2486383846 1.4059399459
C 6.4237474346 1.2227985235 0.0904971675
C 7.7870937684 1.1307610658 -0.4081797618
C 10.6500000000 1.1572259982 -0.8711299063
C -9.2300000000 3.6048663212 -0.3940892390
C -6.4496204946 3.6594216864 -0.0775582487
C -5.0416364017 3.7417617647 0.3749721528
C -2.0839594190 3.7319213011 0.5289052670
C -0.6753282603 3.7198759329 0.5928422298
C 2.2748716577 3.6809570200 0.7683983178
C 3.7577996443 3.8072184769 0.9719675100
C 6.4087620795 3.6522315561 -0.2206615081
C 7.7919825795 3.5923343899 -0.7826897269
C 10.6500000000 3.5963070964 -0.5356003656
C -9.2300000000 6.0801776209 -0.4461744497
C -6.3794889562 5.9930139084 -0.9376796211
C -4.8975806596 5.9794704444 -0.8598551256
C -2.0354212531 5.9953390133 -0.3835130465
C -0.6293789958 5.9869511796 -0.3527784755
C 2.2037798592 6.0616371756 -0.1183521781
C 3.5669153457 6.1034744510 -0.1055506858
C 6.4062984138 6.0967107948 -0.8843790940
C 7.8568772699 6.1447110541 -1.0693363526
C 10.6500000000 6.1444582970 -0.0842165521
C -9.2300000000 8.5863543297 -0.4901248403
C -6.2927826370 8.5450776841 -0.9024141746
C -4.9215269697 8.5334655000 -0.8653521562
C -1.9940875265 8.5164822035 -0.6272077266
C -0.6206966288 8.4115176199 -1.2471555320
C 2.0966787479 8.5220744732 -0.1284748887
C 3.5828957507 8.5415635646 -0.0402279786
C 6.4078656451 8.6415829221 -0.8824985224
C 7.8469833696 8.6384833950 -0.8840406513
C 10.6500000000 8.7379434947 -0.2136071864
C -9.2300000000 11.4995858204 -1.1160295455
C -6.3855524443 11.0678050000 -0.6840123785
C -4.9791153776 11.0678050000 -0.3130180005
C -2.1681493315 11.0678050000 -1.1972113971
C -0.6524133221 11.0678050000 -1.1653715291
C 2.0849982762 11.0678050000 -0.1778413260
C 3.5623284838 11.0678050000 -0.0387246169
C 6.3702821610 11.0678050000 -0.6122889482
C 7.7818188869 11.0678050000 -0.8354309890
C 10.6500000000 11.2615247735 -1.3277095855
O -6.6289090421 -6.8320070089 1.7057488478
O -0.7610428528 8.0991394985 -2.6569148907
O -4.8440885735 4.8443600074 1.3465768260
O -4.6777363828 -9.2253342445 1.0989554957
O 10.2079764542 0.4934541487 -2.1276211008
O 10.1834111926 -2.2872437197 -1.8194138459
H -6.0183534352 -7.5931201854 1.8487058250
H -0.7850313932 8.9604531716 -3.1420253848
O 9.7864808712 -7.9014405994 0.6665056612
O 1.2677257521 10.5993865538 0.9400193675
O -2.4240923761 -1.0745716810 -0.3805919937
O -6.3098448737 -7.5059310232 -1.8978221721
O -5.3375697440 -1.0965373352 2.7056546258
O -2.6166192167 10.9209562979 -2.5778386721
O 7.3464299452 -9.0322427143 0.9743154124
O -8.1758245819 -2.3056573684 1.7996205538
O -6.5707343797 -4.1791396933 -0.8939126621
O 0.0348346675 10.4976279582 -2.3617001579
H -4.5124573410 -1.3414433795 3.1831919275
H -2.4760365573 11.7942184241 -3.0152739481
O 4.4048079834 3.4479539428 2.2230009769
O -4.7219891251 10.2161314670 0.8965614283
O -2.6044949790 -8.8118862401 -1.7300593484
O -8.5860226476 -8.7861613327 -2.6011121064
O 9.3109418613 7.5528426776 1.3721215308
O 5.6653747209 -4.7076142666 -1.2920345798
O -2.0307910841 9.2450245497 0.6481865681
O 4.1631920709 10.5365606695 1.2045232050
O -5.5426461217 5.5383606761 -2.1026189311
O -8.0148463242 7.3145429968 -2.2722065085
H 10.0569505282 7.3957517196 1.9910989603
H 5.1173462960 -3.9070169743 -1.4629850022
O 3.2436019737 1.3753032380 2.7768907508
O 0.7569176903 -2.6404575147 3.0704622096
O 10.0212883421 -8.7988046173 -2.5962751314
O 3.4620105543 -1.8643827415 -0.2894609186
O 8.0710287666 -3.3598363282 2.3425638891
O 1.8898161109 -10.0017392183 -2.2995000711
O -2.6961778811 -3.2004280676 2.7724742822
O 10.3387791368 9.7787800452 0.7595749790
O 2.0280443388 3.6561582903 -0.7000866911
O -8.4327653282 -0.9730147860 -1.5900432959
H 7.7633542676 -4.1437149163 2.8510348445
H 1.6225598633 -9.1680418180 -2.7521434528
O 10.2101596538 5.3045960579 1.0609541213
O 0.5950173047 2.8568724072 2.5381760859
O 6.6459289701 3.4848886424 -1.6894099158
O 1.8094946012 -7.4688323403 -1.4534950741
O -9.3338135292 -6.9266098967 1.4905933018
O 5.1725145944 -0.1464129148 -1.3004555794
O 1.9650249103 -0.6203794480 3.0421169148
O -8.8610806315 10.8968173003 0.1181628156
O 10.7245274048 -4.8639703317 -1.3165430505
O -3.2551559457 7.2058272345 -2.2128854064
H -8.4185149155 -7.0026933010 1.8664717280
H 4.5400962153 -0.9090831059 -1.3515399900
O 2.7960335320 8.3673865830 1.1779573297
O 5.4921779218 5.5105761764 1.2989177131
O -1.1023150923 1.2742361849 -0.4676734805
O 8.4970986658 5.4276234963 -2.2770587677
O -0.8030637041 -7.2710506570 2.4149780960
O -4.5326111653 1.1718174011 -1.2105148634
O -0.2997587917 -4.8057459707 2.8230959424
O -3.6910711849 2.4517493789 2.0542504659
O 0.1278029727 -0.9413933163 0.0475544553
O 4.5716939689 7.3861013893 -1.9848736350
H -0.8342460257 -6.5891825276 3.1320820245
H -4.8886509524 0.3476323338 -1.6120829717
# End of system coordinates
&END COORD
&END SUBSYS
&PRINT
&STRESS_TENSOR
FILENAME __STD_OUT__
LOG_PRINT_KEY TRUE
ADD_LAST NUMERIC
&EACH
CELL_OPT 1
&END EACH
&END STRESS_TENSOR
&END PRINT
&END FORCE_EVAL
&GLOBAL
PROJECT geo_opt_go_patch
RUN_TYPE CELL_OPT
PRINT_LEVEL LOW
&END GLOBAL
&MOTION
&CELL_OPT
EXTERNAL_PRESSURE 0 0 0 0 0 0 0 0 0
KEEP_ANGLES TRUE
KEEP_SYMMETRY TRUE
MAX_ITER 1000
&END CELL_OPT
# Uncomment to allow printing of stress tensor.
# Also need to uncomment the stress tensor evaluation.
&PRINT
&STRESS
&END STRESS
&END PRINT
# End of uncomment
&END MOTION
==================================
hut...@chem.uzh.ch
May 13, 2014, 3:12:18 AM5/13/14
to cp...@googlegroups.com
Hi
I don't know why your SCF didn't converge, but you will get
different stress tensors for MD and simple single point calculations,
as in the former you have a contribution from the atomic velocities.
regards
Juerg
--------------------------------------------------------------
Juerg Hutter Phone : ++41 44 635 4491
Institut für Chemie FAX : ++41 44 635 6838
Universität Zürich E-mail: hut...@chem.uzh.ch
Winterthurerstrasse 190
CH-8057 Zürich, Switzerland
---------------------------------------------------------------
-----cp...@googlegroups.com wrote: -----
To: cp...@googlegroups.com
From: Rafael Soler-Crespo
Sent by: cp...@googlegroups.com
Date: 05/13/2014 01:27AM
Subject: Re: [CP2K:5272] Stress Tensor forcing SCF Wavefunction Optimization
Hi Juerg,
Thanks for your reply. Apparently, this got lost somewhere in space, so I just saw it. I attach my full input at the end of my post.
The system is a sheet of oxidized graphite oxide and I am trying to get an idea of the stress tensor during cell optimization. Based on your post, yes, you are right -- the multiple SCF wavefunction optimizations are just the finite differences to find the stress tensor. I tried using the diagonal analytical keyword but it wouldn't work for some reason. The error was:
*** SCF run NOT converged ***
Core Hamiltonian energy: -514.15564300498284
Repulsive potential energy: 15.42812723286143
Electronic energy: 0.81658900421239
Dispersion energy: 0.00000000000000
Total energy: -497.91092676790907
***************************************************************************
*** 18:26:20 ERRORL2 in qs_dftb_coulomb:build_dftb_coulomb processor 0 ***
*** :: err=-300 condition FAILED at line 271 ***
***************************************************************************
I am unsure why (and trying both on a locally compiled copy and also on a cluster copy didn't work) I got this error. I will also look into using Ewald summations as, yes, my system is both periodic and I am using DFTB. I forgot I turned it off to test some input a long time ago and never returned it to on!
Now I am finding that if I take a snapshot (same atomic coordinates) of the system and recalculate the stress doing a single point (ENERGY_FORCES) calculation it doesn't give me the same stress values. This is probably because the wavefunction is not the same during the cell optimization. Would you agree with my conclusion, and do you have any hints or tips for obtaining accurate results of the stress tensor in CP2K calculations for systems like mine?
Thanks a million for your response and help in advance!
Rafael
===================================
INPUT
&FORCE_EVAL
STRESS_TENSOR DIAGONAL_NUMERICAL
On Saturday, May 3, 2014 9:33:44 AM UTC-5, jgh wrote:Hi
--
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns...@googlegroups.com.
To post to this group, send email to cp...@googlegroups.com.
Visit this group at http://groups.google.com/group/cp2k.
For more options, visit https://groups.google.com/d/optout.
I don't know why your SCF didn't converge, but you will get
different stress tensors for MD and simple single point calculations,
as in the former you have a contribution from the atomic velocities.
regards
Juerg
--------------------------------------------------------------
Juerg Hutter Phone : ++41 44 635 4491
Institut für Chemie FAX : ++41 44 635 6838
Universität Zürich E-mail: hut...@chem.uzh.ch
Winterthurerstrasse 190
CH-8057 Zürich, Switzerland
---------------------------------------------------------------
To: cp...@googlegroups.com
From: Rafael Soler-Crespo
Sent by: cp...@googlegroups.com
Date: 05/13/2014 01:27AM
Subject: Re: [CP2K:5272] Stress Tensor forcing SCF Wavefunction Optimization
Hi Juerg,
Thanks for your reply. Apparently, this got lost somewhere in space, so I just saw it. I attach my full input at the end of my post.
The system is a sheet of oxidized graphite oxide and I am trying to get an idea of the stress tensor during cell optimization. Based on your post, yes, you are right -- the multiple SCF wavefunction optimizations are just the finite differences to find the stress tensor. I tried using the diagonal analytical keyword but it wouldn't work for some reason. The error was:
*** SCF run NOT converged ***
Core Hamiltonian energy: -514.15564300498284
Repulsive potential energy: 15.42812723286143
Electronic energy: 0.81658900421239
Dispersion energy: 0.00000000000000
Total energy: -497.91092676790907
***************************************************************************
*** 18:26:20 ERRORL2 in qs_dftb_coulomb:build_dftb_coulomb processor 0 ***
*** :: err=-300 condition FAILED at line 271 ***
***************************************************************************
I am unsure why (and trying both on a locally compiled copy and also on a cluster copy didn't work) I got this error. I will also look into using Ewald summations as, yes, my system is both periodic and I am using DFTB. I forgot I turned it off to test some input a long time ago and never returned it to on!
Now I am finding that if I take a snapshot (same atomic coordinates) of the system and recalculate the stress doing a single point (ENERGY_FORCES) calculation it doesn't give me the same stress values. This is probably because the wavefunction is not the same during the cell optimization. Would you agree with my conclusion, and do you have any hints or tips for obtaining accurate results of the stress tensor in CP2K calculations for systems like mine?
Thanks a million for your response and help in advance!
Rafael
===================================
INPUT
&FORCE_EVAL
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns...@googlegroups.com.
To post to this group, send email to cp...@googlegroups.com.
Visit this group at http://groups.google.com/group/cp2k.
For more options, visit https://groups.google.com/d/optout.
Rafael Soler-Crespo
May 13, 2014, 10:19:58 AM5/13/14
to
Hi Juerg,
And I take it to mean that the function will not work if I am using a virial formulation of the stress tensor (and thus forcing me to always use the numerical evaluation). Does this seem correct to you? I would prefer to use the virial formulation of the stress tensor but since the simulation isn't converging I have little choice as it is. Furthermore, this check is within an do_ewald loop (and I don't even have Ewald enabled for this calculation).
You are correct. However, I am not using MD on this run. In the one I posted, I am trying to perform a cell optimization. I didn't take off the keywords for the MD printing of stress but those are never printed (as I have no MD steps). What I don't understand is why I can get a stress tensor using the numerical evaluation of the stress tensor but when I use the analytic formulation of the stress tensor, the code always crashes after the initial SCF wavefunction optimization. Do you have any ideas what might be going on to prevent the analytical formulation from processing correctly? I checked line 271 of the referenced code:
Is it something related to DFTB not being able to use analytical formulations of the stress tensor?
Thanks,
Rafael
Rafael
And i
hut...@chem.uzh.ch
May 13, 2014, 10:24:28 AM5/13/14
to cp...@googlegroups.com
Hi
the stress tensor for direct summation of Coulomb interactions is not
implemented. It is assumed, that you are using Ewald for periodic systems.
regards
Juerg
--------------------------------------------------------------
Juerg Hutter Phone : ++41 44 635 4491
Institut für Chemie FAX : ++41 44 635 6838
Universität Zürich E-mail: hut...@chem.uzh.ch
Winterthurerstrasse 190
CH-8057 Zürich, Switzerland
---------------------------------------------------------------
-----cp...@googlegroups.com wrote: -----
To: cp...@googlegroups.com
From: Rafael Soler-Crespo
Sent by: cp...@googlegroups.com
Date: 05/13/2014 04:17PM
Subject: Re: [CP2K:5280] Stress Tensor forcing SCF Wavefunction Optimization
Hi Juerg,
You are correct. However, I am not using MD on this run. In the one I posted, I am trying to perform a cell optimization. I didn't take off the keywords for the MD printing of stress but those are never printed (as I have no MD steps). What I don't understand is why I can get a stress tensor using the numerical evaluation of the stress tensor but when I use the analytic formulation of the stress tensor, the code always crashes after the initial SCF wavefunction optimization. Do you have any ideas what might be going on to prevent the analytical formulation from processing correctly? I checked line 271 of the referenced code:
cppostcondition(.NOT.use_virial,cp_failure_level,routinep,error,failure)
And I take it to mean that the function will not work if I am using a virial formulation of the stress tensor (and thus forcing me to always use the numerical evaluation). Does this seem correct to you? I would prefer to use the virial formulation of the stress tensor but since the simulation isn't converging I have little choice as it is...
Is it something related to DFTB not being able to use analytical formulations of the stress tensor?
Thanks,
Rafael
And i
On Tuesday, May 13, 2014 2:12:18 AM UTC-5, jgh wrote:Hi
the stress tensor for direct summation of Coulomb interactions is not
implemented. It is assumed, that you are using Ewald for periodic systems.
regards
Juerg
--------------------------------------------------------------
Juerg Hutter Phone : ++41 44 635 4491
Institut für Chemie FAX : ++41 44 635 6838
Universität Zürich E-mail: hut...@chem.uzh.ch
Winterthurerstrasse 190
CH-8057 Zürich, Switzerland
---------------------------------------------------------------
-----cp...@googlegroups.com wrote: -----
To: cp...@googlegroups.com
From: Rafael Soler-Crespo
Sent by: cp...@googlegroups.com
Subject: Re: [CP2K:5280] Stress Tensor forcing SCF Wavefunction Optimization
Hi Juerg,
You are correct. However, I am not using MD on this run. In the one I posted, I am trying to perform a cell optimization. I didn't take off the keywords for the MD printing of stress but those are never printed (as I have no MD steps). What I don't understand is why I can get a stress tensor using the numerical evaluation of the stress tensor but when I use the analytic formulation of the stress tensor, the code always crashes after the initial SCF wavefunction optimization. Do you have any ideas what might be going on to prevent the analytical formulation from processing correctly? I checked line 271 of the referenced code:
cppostcondition(.NOT.use_virial,cp_failure_level,routinep,error,failure)
Is it something related to DFTB not being able to use analytical formulations of the stress tensor?
Thanks,
Rafael
And i
Rafael Soler-Crespo
May 13, 2014, 10:40:02 AM5/13/14
to cp...@googlegroups.com
Hi Juerg,
I placed Ewald summations (SPME as per the if loop on the referenced code by the error) on my input and now the stress tensor is working correctly under the analytical formulation.
A million thanks for your help!
Rafael
Reply all
Reply to author
Forward
0 new messages