 |
SEARCH
TOOLBOX
LANGUAGES
Forum Menu
CDFT: Unable to improve the multipliers 0
From NWChem
Viewed 3287 times, With a total of 7 Posts
|
|
|
Clicked A Few Times
Threads 2
Posts 6
|
|
| 4:25:19 AM PST - Fri, Jan 18th 2013 |
|
I am running into some trouble with a CDFT run. I have two c60 molecules and want to constrain an extra charge to only one of them. The input looks like this:
geometry noautoz noautosym
C +81.489381446 +13.112890436 +166.873903869
C +78.962304344 +17.150679902 +166.106934122
C +75.609428653 +11.624753147 +164.558898327
C +80.887578721 +13.497489794 +168.136940890
C +81.654772483 +14.057818439 +165.864884969
C +80.474467315 +14.811617587 +168.338957375
C +79.213643788 +15.082316355 +169.005986226
C +80.055767208 +12.397859908 +168.590953045
C +78.846344668 +12.657058644 +169.229980338
C +78.416032945 +14.027387306 +169.440997770
C +80.926744528 +12.422388720 +164.172843251
C +80.755153509 +11.436659416 +165.224863371
C +81.367753872 +13.705317838 +164.485853591
C +81.030471236 +11.774860562 +166.546893144
C +80.144462506 +11.333141652 +167.608923400
C +79.581825586 +10.643040003 +164.907862784
C +78.732217206 +10.218321663 +165.925891753
C +79.019136327 +10.571522382 +167.304922607
C +79.737898879 +14.679844708 +162.916839020
C +79.278289090 +13.342114349 +162.590828471
C +80.760030750 +14.857246056 +163.845851841
C +79.860811311 +12.236887167 +163.205830243
C +79.029599901 +11.136556636 +163.659842388
C +77.839854110 +13.394612717 +162.404838337
C +77.042142721 +12.339583651 +162.839849352
C +77.649966410 +11.187454858 +163.481851436
C +79.564407112 +16.765580466 +164.842897456
C +78.497973811 +16.580078913 +163.875884453
C +80.672535122 +15.922764450 +164.827881479
C +78.582769771 +15.559061399 +162.933855997
C +77.409842437 +14.765041397 +162.615855237
C +77.236750739 +16.849777517 +164.541913134
C +76.111424031 +16.088158777 +164.237912341
C +76.200419907 +15.023940077 +163.255882696
C +80.646958495 +15.797246871 +167.287937414
C +79.492229444 +16.676463569 +167.303953690
C +81.224260727 +15.427847580 +166.075901872
C +77.523469287 +17.203178275 +165.919943824
C +78.605920656 +16.234145092 +168.364983778
C +77.226087120 +16.285143336 +168.185993719
C +76.674060939 +16.778860003 +166.937972792
C +74.766005535 +14.308709513 +164.972931879
C +75.225215257 +15.646739922 +165.298942963
C +75.500033428 +15.985340605 +166.620972213
C +74.601614137 +13.363681495 +165.981950768
C +75.781819807 +12.610282420 +163.506878728
C +75.368408352 +13.924010146 +163.708894683
C +75.031925860 +11.993652884 +165.770933867
C +76.691379409 +10.655319366 +167.003938284
C +77.294082276 +10.270920050 +165.739901616
C +76.763957671 +10.745436432 +164.542882052
C +75.583852028 +11.498735485 +167.018954259
C +76.518388242 +12.742055299 +168.928996553
C +77.673916899 +11.862638570 +168.912979739
C +77.758512827 +10.841120972 +167.970951282
C +75.496256383 +12.564153862 +168.000983897
C +75.329042510 +14.999411269 +167.672992326
C +76.394775693 +15.185112855 +168.640005337
C +76.977397931 +14.079685640 +169.255007637
C +74.888232671 +13.716282648 +167.359981983
C +72.434019535 +1.721432016 +166.588379359
C +69.624051952 +5.636960878 +166.808023047
C +76.046464092 +7.067439885 +165.871815118
C +71.873175616 +1.930709576 +165.266097872
C +71.815895037 +2.280483436 +167.703427675
C +70.717821604 +2.692155936 +165.113361198
C +70.611580686 +3.639621896 +164.018493844
C +72.973312651 +2.086791450 +164.332247834
C +72.872082783 +2.995594072 +163.282283565
C +71.666901507 +3.788914668 +163.122889853
C +74.005511418 +2.914277756 +168.636261426
C +74.650258453 +2.330558011 +167.474128202
C +72.618179461 +2.889326982 +168.749135316
C +73.880838598 +1.746764566 +166.471508118
C +74.213738690 +1.972293146 +165.076558534
C +75.784952205 +3.165218321 +167.124300598
C +76.105283890 +3.381292413 +165.786825860
C +75.302478375 +2.772911387 +164.741423897
C +72.615674629 +5.204131077 +169.590283172
C +74.062272054 +5.229970088 +169.472854998
C +71.909082338 +4.058483080 +169.235086254
C +74.742707639 +4.108510454 +169.006042627
C +75.842960682 +4.263504368 +168.071724630
C +74.452594959 +6.554117861 +169.024727274
C +75.507916453 +6.703304541 +168.128903184
C +76.216750616 +5.533621346 +167.641877480
C +70.185059152 +5.427190798 +168.130411133
C +70.922153679 +6.621683085 +168.499478981
C +70.667572538 +4.172102513 +168.491068209
C +72.112020955 +6.512721214 +169.214160122
C +73.247181979 +7.347114463 +168.865231870
C +70.817350372 +7.569569543 +167.405367441
C +71.906252217 +8.370280200 +167.070569944
C +73.146169115 +8.255842002 +167.815214951
C +70.072676163 +3.274821975 +166.275387301
C +69.569104434 +4.583269300 +165.899936702
C +70.611036370 +3.073916284 +167.543611917
C +70.014620072 +6.961621448 +166.360360843
C +69.902940410 +4.809201422 +164.505340214
C +70.276825739 +6.079437372 +164.075710884
C +70.334526380 +7.177508332 +165.022852827
C +73.685938800 +8.621809714 +165.558915259
C +72.239073422 +8.596244359 +165.675897213
C +71.469352092 +8.013010354 +164.673208273
C +74.303870397 +8.061722032 +164.443964672
C +75.401649579 +7.650460580 +167.034437296
C +74.246519783 +8.411966847 +166.881223089
C +75.508715324 +7.268517880 +164.603616085
C +75.935220781 +4.915156051 +164.016256299
C +76.495512526 +4.705243147 +165.339139486
C +76.550237245 +5.759194424 +166.247287612
C +75.452083841 +6.170123177 +163.656273167
C +73.503836962 +5.138500297 +162.557371983
C +74.007426975 +3.829406184 +162.933360663
C +75.197679089 +3.720157850 +163.647375516
C +74.210806853 +6.283837825 +162.912514798
C +72.114180460 +7.428675763 +163.511204469
C +71.376905948 +6.234767594 +163.141448789
C +72.057453615 +5.113184621 +162.674934286
C +73.501538100 +7.453353303 +163.398220198
end
basis
* library 3-21g
end
memory 1500 mb
charge -1.0
dft
xc xpbe96 cpbe96
direct
iterations 100
odft
mult 2
cdft 1 60 charge -1
end
task dft
The error occurs early...
Memory utilization after 1st SCF pass:
Heap Space remaining (MW): 47.69 47686129
Stack Space remaining (MW): 49.11 49113505
convergence iter energy DeltaE RMS-Dens Diis-err time
---------------- ----- ----------------- --------- --------- --------- ------
d= 0,l s=0.0,diis 1 -4540.6409092047 -2.81D+04 2.30D-03 3.38D+00 172.6
1.94D-03 3.38D+00
CDFT multipliers:
1 -0.0233316383
iter = 9
CDFT: Unable to improve the multipliers 0
------------------------------------------------------------------------
An error occured in the Runtime Database
------------------------------------------------------------------------
------------------------------------------------------------------------
An error occured in the Runtime Database
------------------------------------------------------------------------
For more information see the NWChem manual at
http://www.nwchem-sw.org/index.php/NWChem_Documentation
For further details see manual section:
2:2:CDFT: Unable to improve the multipliers:: 0
(rank:2 hostname:thinc008 pid:5049):ARMCI DASSERT fail. ../../ga-5-1-1/armci/src/common/armci.c:ARMCI_Error():208 cond:0
CDFT: Unable to improve the multipliers 0
------------------------------------------------------------------------
The latter error is repeated for tasks 3-8 as well. Has anyone encountered this before?
|
|
|
-
Edoapra Forum:Admin, Forum:Mod, bureaucrat, sysop
|
|
|
Forum Vet
Threads 9
Posts 1444
|
|
| 3:42:24 PM PST - Fri, Jan 18th 2013 |
|
Fragment Guess
|
Bob,
You might want to use a fragment guess made of two separate c60 molecules to make the cdft work a bit easier.
I have upload an input file that shows how to do it in practice
http://www.nwchem-sw.org/images/C120.nw
Cheers, Edo
|
|
|
|
|
|
Clicked A Few Times
Threads 2
Posts 6
|
|
| 3:00:04 AM PST - Sat, Jan 19th 2013 |
|
Edo,
Did you try to run your input? Unfortunately, for me it fails just the same way as before at the first CDFT iteration... 
Bob
|
|
Edited On 3:01:54 AM PST - Sat, Jan 19th 2013 by Bbob79
|
|
|
|
|
|
Clicked A Few Times
Threads 3
Posts 9
|
|
| 9:15:05 AM PST - Mon, Jan 21st 2013 |
|
Quote:Bbob79 Jan 19th 3:00 amEdo,
Did you try to run your input? Unfortunately, for me it fails just the same way as before at the first CDFT iteration...
Bob
Bob,
In the Edo's input, "odft" seems to be missing in CDFT of the C120.
Namely, I recommend you using
set geometry c120
dft
cdft 1 60 charge -1
vectors input fragment c60_a.mos c60_b.mos output c120.mos
odft
mult 2
direct
end
task dft ignore
Futofuji
|
|
|
-
Edoapra Forum:Admin, Forum:Mod, bureaucrat, sysop
|
|
|
Forum Vet
Threads 9
Posts 1444
|
|
| 10:24:14 AM PST - Mon, Jan 21st 2013 |
|
Quote:Bbob79 Jan 19th 3:00 amEdo,
Did you try to run your input? Unfortunately, for me it fails just the same way as before at the first CDFT iteration...
Bob
Bob
you're right. Let me have another look. A possible fix might be first to converge with no cdft (or at least run a few iterations without
reaching full convergence), then try cdft.
May I asking why you are using cdft here? Is it because you really want once c60 to have a -1 charge or as an aid for convergence or for some other reason?
Cheers, Edo
|
|
|
-
Edoapra Forum:Admin, Forum:Mod, bureaucrat, sysop
|
|
|
Forum Vet
Threads 9
Posts 1444
|
|
| 10:26:16 AM PST - Mon, Jan 21st 2013 |
|
Quote:Futofuji Jan 21st 9:15 am
Bob,
In the Edo's input, "odft" seems to be missing in CDFT of the C120.
Namely, I recommend you using
Futofuji
Futofuji,
adding odft is redundant (but harmless), since it is inherited from the previous dft sections.
Cheers, Edo
|
|
|
|
|
|
Clicked A Few Times
Threads 2
Posts 6
|
|
| 10:36:46 AM PST - Mon, Jan 21st 2013 |
|
Quote:Edoapra Jan 21st 6:24 pmQuote:Bbob79 Jan 19th 3:00 amEdo,
Did you try to run your input? Unfortunately, for me it fails just the same way as before at the first CDFT iteration...
Bob
Bob
you're right. Let me have another look. A possible fix might be first to converge with no cdft (or at least run a few iterations without
reaching full convergence), then try cdft.
May I asking why you are using cdft here? Is it because you really want once c60 to have a -1 charge or as an aid for convergence or for some other reason?
Cheers, Edo
Edo,
Thanks again. I am curious to learn what you would recommend. And, yes, I do want to constrain one c60 to have -1 charge. On the long run, I want to consider a larger cluster with that charge constraint.
Bob
|
|
|
-
Edoapra Forum:Admin, Forum:Mod, bureaucrat, sysop
|
|
|
Forum Vet
Threads 9
Posts 1444
|
|
| 11:11:06 AM PST - Mon, Jan 21st 2013 |
|
Quote:Bbob79 Jan 21st 10:36 am
Edo,
Thanks again. I am curious to learn what you would recommend. And, yes, I do want to constrain one c60 to have -1 charge. On the long run, I want to consider a larger cluster with that charge constraint.
Bob
Bob,
I forgot to put the "convergence nolevelshifting" line in the last dft section.
Unfortunately, cdft does not seem to like level shifting.
I have uploaded a modifed c120.nw at
http://www.nwchem-sw.org/images/C120.nw
I have to tell you that I have tried to run this input for a few cycles and it
seems that cdft does not crash but it does not seem likely to reach convergence.
Edo
|
|
|
AWC's:
2.5.10 MediaWiki - Stand Alone Forum Extension
Forum theme style by: AWC
| |
