Semiempirical Geometry Optimization calculation with periodic boundary condition
1,274 views
Skip to first unread message
bharat
Nov 6, 2013, 11:57:56 AM11/6/13
to
Hello Everyone,
I am trying to perform geometry optimization calculation using semiempirical methods AM1, PM3, RM1, and PM6 with periodic boundary condition for bulk water (64 water molecules, initial structure is from M. D. Ben et al. J. Phys. Chem. Lett. 4, 3753 (2013)) . It is found that SCF is not converged and the structures are very bad. (O-H bond lengths increased to 5-8 A and structure is not converged.)
Does anyone have experience using semiempirical methods for PBC calculation? Please see the .xyz file using AM1 method.
All suggestions are appreciated.
Thank you.
Bharat Sharma
Alexey Nesternko
Nov 6, 2013, 11:29:48 PM11/6/13
to cp...@googlegroups.com
Can you attach cp2k input files for direct reproduce of your calculations (ALL files that needed for calculation)?
среда, 6 ноября 2013 г., 20:13:26 UTC+4 пользователь bharat написал:
среда, 6 ноября 2013 г., 20:13:26 UTC+4 пользователь bharat написал:
Bharat Sharma
Nov 7, 2013, 6:34:22 AM11/7/13
to cp...@googlegroups.com
Hello Alexey,
Thank you for your reply. Here is my input. I took from cp2k TEST folder. This input works perfectly for NON-PERIODIC systems to do geometry optimization.
Thank you.
Bharat
**************************************************
&FORCE_EVAL
METHOD Quickstep
&DFT
CHARGE 0
UKS T
&QS
METHOD AM1
&SE
&END SE
&END QS
&SCF
SCF_GUESS ATOMIC
&END SCF
&END DFT
&SUBSYS
&CELL
ABC 12.42 12.42 12.42
PERIODIC XYZ
&END CELL
&COORD
................................................
...............................................
...............................................
&END COORD
&END SUBSYS
&END FORCE_EVAL
&MOTION
&GEO_OPT
MAX_ITER 5000
&END GEO_OPT
&END MOTION
&GLOBAL
PROJECT 64-water
RUN_TYPE GEO_OPT
&END GLOBAL
************************************************************
--
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/groups/opt_out.
David Stelter
Nov 7, 2013, 1:33:30 PM11/7/13
to cp...@googlegroups.com
Hi Bharat,
You can simply supply more SCF iterations to try to force a convergence to solve your SCF problem. By default, MAX_SCF is 50 iterations with an accuracy (EPS_SCF) of 1x10^-5. These parameters can simply be added to your input file under the SCF section. Hopefully, the SCF convergence will fix the geometry problems! I am unable to test this for you as did not provide coordinates in the input file for your system.
Best,
David
Bharat Sharma
Nov 7, 2013, 1:38:32 PM11/7/13
to cp...@googlegroups.com
Hello David,
Thank you for your suggestions. I will try. Here is my complete INPUT with coordinates.
Thank you.
Bharat
*********************************************************
&FORCE_EVAL
METHOD Quickstep
&DFT
CHARGE 0
UKS T
&QS
METHOD AM1
&SE
&END SE
&END QS
&SCF
SCF_GUESS ATOMIC
&END SCF
&END DFT
&SUBSYS
&CELL
ABC 12.42 12.42 12.42
PERIODIC XYZ
&END CELL
&COORD
O 3.8230943318 -4.5198948649 5.2582488668
H 4.4117593821 -5.4463322402 6.0596860669
O -3.5612056858 1.8633025312 4.1469785659
H -3.3823543540 0.6906309631 4.4224847660
H -3.4885955188 2.0946812564 5.3996183543
O 4.7086948517 3.0093829507 5.5464511841
H 3.4262158642 3.4253180971 5.4782028172
H 4.4579128961 2.5841332560 4.0875459268
O -2.7045923244 -0.2102004489 -3.5420257152
H -1.9939093671 0.4901891677 -2.6434576509
H -1.6770767416 -0.8062198960 -3.8915363498
O -3.6216543968 4.7004584483 5.7565880477
H -4.8833823363 4.8251865842 5.8114967791
H -3.9447391516 4.6873989245 4.4858624201
O 5.3421447081 -0.1217505313 2.3239281663
H 6.3886685734 -0.5183444773 2.8431670261
H 6.2956040964 0.7869438015 1.9798473892
O 4.1596195298 -5.7536764673 -0.2321266367
H 3.0364542503 -5.4113326871 0.3505072545
H 3.8249932555 -4.7905788631 -1.1998930336
O -2.7677897367 -3.4362924990 -1.2237401267
H -3.0571228916 -2.2194319904 -1.8147172497
H -2.0936433688 -2.4089247165 -0.6039159980
O 5.7326722939 -3.1420476877 3.1441398518
H 4.4891144628 -3.1901078429 3.5256475925
H 5.0077941888 -3.0691162469 2.0818914121
O 1.8655927568 4.0018105285 5.6091613765
H 1.6481294638 2.7815183286 5.7207957003
H 1.9231894440 3.9394825623 6.8720670966
O 4.1375052157 -4.1169889939 -3.0371524865
H 2.9170329830 -3.7327190948 -3.6019884434
H 4.4913764104 -2.7576576960 -2.6754176093
O 1.9871654379 -1.7808562073 1.0132132857
H 1.4956276767 -2.4016243012 1.9335445590
H 1.7274141884 -2.0552735562 -0.0686962102
O 0.6513325188 -5.7978739003 -4.7563067307
H 1.7949082911 -6.2902958474 -4.8694750267
H 0.4252927215 -5.7724103191 -6.1363048623
O 4.9831165787 3.9049035266 -3.4623226377
H 4.5251075625 2.7865839706 -3.1950850168
H 5.2621534715 3.8351355152 -2.2067690949
O 2.8343986251 -1.7265674941 5.0957018393
H 3.8901917364 -2.1104141142 5.6756518968
H 3.2521343809 -0.6334705960 4.6349747217
O -0.1479206385 1.9981716633 5.9116582713
H 0.1422339976 0.8030977940 6.1578598984
H -0.0864235441 1.7386060492 4.6196599471
O 1.2272551990 -3.0117695026 3.4523637307
H 2.0783696915 -2.5961838818 4.3674958664
H 1.0660132084 -4.0130496587 4.2664825333
O -2.5397248533 2.8806483799 2.0741109075
H -1.5098655941 2.2599706114 2.2589599419
H -2.6209631153 2.6946708819 3.2721332728
O 0.6738899383 -0.6128060815 -5.6128026799
H 1.7202403738 -0.8676291902 -6.3604735082
H 1.4302320591 -0.5183677872 -4.5721651977
O -1.2753495181 -0.8673605398 0.1044321785
H -1.2223619399 0.3974601211 -0.0527350062
H -1.0144907302 -0.7791400986 1.3121236061
O -3.9571563578 3.1480835597 -3.0975427170
H -4.2080297976 4.4192573404 -3.1385984521
H -2.8570588583 3.7128862665 -3.1985251337
O 5.1859100090 4.4800678666 2.1984827712
H 4.4013978525 3.6881982415 1.3959385690
H 4.1130998223 5.1534828329 2.6583198701
O 4.1854609441 1.2393397940 -3.0854099930
H 4.8184865712 0.8510213936 -4.2248788205
H 4.4096483092 0.0421519641 -2.6105001389
O -4.9447204667 5.9384731949 -2.7966460519
H -5.4472177991 6.5314233234 -1.7816421359
H -6.0926128191 6.4385014826 -3.1472691900
O 5.0297403238 -2.3455559345 0.5702079817
H 5.3511690005 -1.2931260170 1.3181707304
H 3.5422535190 -2.0532842005 0.7433884076
O 1.8945371353 1.0957984297 1.7508553995
H 3.0381214506 1.4329047552 1.3208879374
H 2.1961106991 -0.1311690464 1.1933286407
O -5.5733698128 -5.0902387913 -0.3758861288
H -6.7041675030 -5.3442819612 -0.0740418294
H -5.9938383703 -3.9336998317 -0.5704589087
O -4.7278846581 4.7949699714 2.7855690867
H -6.0114205857 4.7509272423 2.5392932894
H -4.7167766151 3.5028746336 2.2913331297
O 5.3279775793 0.6992201889 -5.5706417256
H 6.4511479381 1.1765754344 -5.4352792743
H 5.4008580313 1.6004173063 -6.3629147992
O 0.8727740244 -1.7432658898 -1.2923943573
H 0.3010790073 -2.8472258063 -1.6988076221
H -0.2312850960 -1.6202825726 -0.9366604220
O 3.7609385121 2.5391347684 0.3450413833
H 4.9680346520 2.8111829643 -0.4882643248
H 3.1897630072 3.3926450321 -0.5119973793
O -1.0649776727 -1.2464857344 2.9638278887
H -0.2662866134 -2.2400625291 3.0959394170
H -2.0135723256 -1.8278612038 3.1254280051
O 2.1170545031 -5.5419702172 1.2919514207
H 1.0652550305 -6.0955245369 1.1380415200
H 2.6502653892 -6.0715573210 2.2797572146
O -0.6513987669 -3.5783155169 -2.9718593782
H -1.7277174488 -4.0936767941 -2.3447919780
H -0.0470593646 -4.8344690552 -3.1855804540
O -5.4663654588 2.1677032262 1.5863235537
H -4.1741193232 1.8958277158 1.8709523739
H -5.1789879857 1.8119220894 0.3533003666
O 2.1170555880 -3.4111546528 -5.3519595376
H 1.2448042154 -4.2708608193 -4.8375544944
H 1.5313281141 -2.3789870678 -5.4479503673
O -1.2622635000 3.6406646822 -3.2684201995
H -0.9201859484 4.2430093332 -4.6325699442
H -0.8527584070 4.9261011387 -2.8346591715
O -5.3130145771 -4.9085599129 4.5436909515
H -6.3993891551 -4.4666875233 4.0177638092
H -4.8221097928 -5.3385790504 3.2854237959
O 2.8243576869 5.7420587632 3.5093813753
H 3.4475357352 6.5754870329 4.2995477798
H 2.6227959313 5.1161303531 4.5954154470
O -4.1917296707 1.3570064133 -5.3818910995
H -3.4278543842 0.9236461294 -4.5147705972
H -3.9229377409 2.4070081331 -4.5758684708
O 3.5258010318 0.8383199111 4.2631475877
H 2.7296826381 0.6907149867 3.1933008167
H 4.3836547868 -0.0095241513 3.5390456395
O -4.5025690356 -0.7557459026 3.9354780593
H -5.3535292154 -1.4061913692 4.7062150294
H -4.2451336382 -1.9946125810 3.8500291124
O -0.4380654439 1.4390700425 3.1940423464
H 0.6811252395 1.5058735601 2.6051605347
H -0.2592668167 0.1778218376 3.1576033675
O -1.9149706811 5.0900162910 0.4377449270
H -2.4223869635 4.0255635368 0.9899205117
H -2.5958133388 5.9525704817 1.0582087947
O -1.6256152064 -1.9580160385 -4.8956854261
H -0.4695093212 -2.3465714100 -4.3524809611
H -0.5339384492 -1.3086116820 -5.5246608862
O -5.0936835915 1.7522574988 -1.0713684229
H -4.8873921039 2.6748581414 -1.9895576687
H -5.5049370565 0.9526258996 -2.0772896905
O 4.7822538107 -1.2717756755 -2.0872865313
H 6.0945661610 -1.0719544575 -2.2139370457
H 4.9082025622 -1.3564020380 -0.9122921692
O -0.6237849041 1.2386854086 -1.5381137773
H -0.7135342896 2.3303753598 -2.3219243091
H 0.3172258322 0.9530112803 -2.2583936246
O 0.1855806308 5.7703380995 -1.2658899442
H -0.1006795170 7.0279668765 -1.1306583500
H -0.9124231374 5.5846283801 -0.4996291954
O 5.5359127346 -1.7857832652 5.7183116653
H 6.2442680474 -2.3479676118 6.6132244823
H 5.7193581804 -0.8592768797 6.8560328838
O -2.3722188766 -5.5675554953 -4.2183479483
H -2.4473443707 -6.8468293563 -4.5397236813
H -3.5392273884 -5.9010509485 -3.6919759280
O -5.0566656629 -1.2804952949 -3.1233232526
H -4.9578252937 -2.4424189805 -3.8437836216
H -3.9125179983 -0.8778083124 -3.3653093055
O 2.4224926830 4.3124582854 -1.3098664930
H 1.2497658249 4.3317602980 -1.0441246827
H 2.5522650042 5.5249850912 -0.8611320513
O -5.7464780604 -3.7920156704 -5.0281968210
H -4.7650000054 -4.1243466295 -5.8797989835
H -6.4243495564 -4.6083956605 -5.9642399492
O -0.1809305646 5.1101453408 3.0090554933
H -0.7913900935 4.1570625107 3.5843672894
H -0.8773429285 4.7266812044 1.9472771533
O -2.9262978460 -3.9760589026 -6.3928557269
H -2.7508267626 -4.8734613988 -5.4631686351
H -2.1069579541 -3.1742462964 -5.8150905348
O -2.7839196045 -3.3452806415 3.5485724397
H -3.0074881941 -4.1845269137 4.4176725045
H -3.0999557455 -3.9315476706 2.5911848767
O 2.3806062254 4.7398054413 -4.1381208167
H 3.5130758683 4.5192007325 -3.6012586758
H 1.9719332501 4.5198151860 -2.9804608713
O 1.4450109296 0.2935787476 -3.1284579098
H 1.7883538450 -0.7672155657 -2.6285197465
H 2.6149551735 0.4677648699 -3.0598789725
O -0.9957515532 4.4986978562 6.0443757494
H -0.8631216686 3.0355913937 5.9592751927
H -2.2633009107 4.3536458576 5.8027402791
O -0.5144255324 -3.8966385064 0.2806810323
H 0.4975005805 -4.0055251828 1.3089433825
H -1.2356001907 -4.4891030043 1.4358564378
O 5.8036109755 5.7970313839 6.0042588118
H 5.2413091025 4.6535298826 5.4779581385
H 6.4462508991 6.2123485591 4.8260009215
O -0.1561553634 -5.1035308246 4.7274965740
H -0.6465244271 -6.3073794820 4.0619144434
H -1.4024382439 -4.8118220454 5.2205370964
O -3.6879351221 -5.2988773511 1.8854650848
H -4.3341877560 -6.3605752021 1.9637700336
H -4.4414028513 -5.4658411288 0.7926068865
&END COORD
&END SUBSYS
&END FORCE_EVAL
&MOTION
&GEO_OPT
MAX_ITER 5000
&END GEO_OPT
&END MOTION
&GLOBAL
PROJECT 64-water
RUN_TYPE GEO_OPT
&END GLOBAL
*********************************************************
hut...@pci.uzh.ch
Nov 7, 2013, 2:01:07 PM11/7/13
to cp...@googlegroups.com
Hi
periodic NDDO calculations are not a standard application.
This type of application needs a lot of care and insight
into the implemented methods. With other words, it will not
work if you are just using defaults.
There are many input options that need careful consideration,
especially options in the sections
&SE ... &END
&POISSON
&EWALD
&END
&END
Have a look in the input files in the regtest sections.
For example
...tests/SE/regtest-3-4/Al2O3.inp
There are also some basic algorithmic problems that are associated
with the way the standard NDDO methods handle Coulomb interactions.
There is one NDDO method (PNNL) that is free from this problem,
unfortunately, there is no real parameter list available.
To make this clear again. It only makes sense to try to use this
options if you are willing to dig deep into the theory and algorithms
of these simulation methods.
regards
Juerg
--------------------------------------------------------------
Juerg Hutter Phone : ++41 44 635 4491
Physical Chemistry Institute FAX : ++41 44 635 6838
University of Zurich E-mail: hut...@pci.uzh.ch
Winterthurerstrasse 190
CH-8057 Zurich, Switzerland
---------------------------------------------------------------
-----cp...@googlegroups.com wrote: -----
To: cp...@googlegroups.com
From: bharat
Sent by: cp...@googlegroups.com
Date: 11/06/2013 05:13PM
Subject: [CP2K:4771] Semiempirical Geometry Optimization calculation with periodic boundary condition
Hello Everyone,
I am trying to perform geometry optimization calculation using semiempirical methods AM1, PM3, RM1, and PM6 with periodic boundary condition for bulk water. It is found that SCF is not converged and the structures are very bad. (O-H bond lengths increased to 5-8 A and structure is not converged.)
Does anyone have experience using semiempirical methods for PBC calculation?
periodic NDDO calculations are not a standard application.
This type of application needs a lot of care and insight
into the implemented methods. With other words, it will not
work if you are just using defaults.
There are many input options that need careful consideration,
especially options in the sections
&SE ... &END
&POISSON
&EWALD
&END
&END
Have a look in the input files in the regtest sections.
For example
...tests/SE/regtest-3-4/Al2O3.inp
There are also some basic algorithmic problems that are associated
with the way the standard NDDO methods handle Coulomb interactions.
There is one NDDO method (PNNL) that is free from this problem,
unfortunately, there is no real parameter list available.
To make this clear again. It only makes sense to try to use this
options if you are willing to dig deep into the theory and algorithms
of these simulation methods.
regards
Juerg
--------------------------------------------------------------
Juerg Hutter Phone : ++41 44 635 4491
Physical Chemistry Institute FAX : ++41 44 635 6838
University of Zurich E-mail: hut...@pci.uzh.ch
Winterthurerstrasse 190
CH-8057 Zurich, Switzerland
---------------------------------------------------------------
-----cp...@googlegroups.com wrote: -----
To: cp...@googlegroups.com
From: bharat
Sent by: cp...@googlegroups.com
Date: 11/06/2013 05:13PM
Subject: [CP2K:4771] Semiempirical Geometry Optimization calculation with periodic boundary condition
Hello Everyone,
I am trying to perform geometry optimization calculation using semiempirical methods AM1, PM3, RM1, and PM6 with periodic boundary condition for bulk water. It is found that SCF is not converged and the structures are very bad. (O-H bond lengths increased to 5-8 A and structure is not converged.)
Does anyone have experience using semiempirical methods for PBC calculation?
Bharat Sharma
Nov 7, 2013, 2:07:53 PM11/7/13
to cp...@googlegroups.com
Hello Hutter,
Thank you for your reply. I will go through your suggestions.
Bharat
Reply all
Reply to author
Forward
0 new messages