Molcas Forum
Support and discussions for Molcas and OpenMolcas users and developers
You are not logged in.
- Topics: Active | Unanswered
Announcement
Please note: The forum's URL has changed. The new URL is: https://molcasforum.univie.ac.at. Please update your bookmarks!
You can choose an avatar and change the default style by going to "Profile" → "Personality" or "Display".Pages: 1
#1 2021-04-16 10:44:21
- FAguilar
- Member
- From: San Sebastián, Spain
- Registered: 2021-04-16
- Posts: 5
- Website
NaN when asking for AMFI
Dear Molcas users,
I'm trying to run some calculations with a basis set that is not in the basis-library in the OpenMolcas directory. The problem is that, if I try to ask for the atomic mean field integrals (to introduce SOC in the last step with RASSI) I get this error:
Basis set specifications :
Symmetry species a
Basis functions 34
SO-integrals were calculated for S: [Ne]3s^2 3p^4
!!! WARNING !!!
NANs encountered
The numbers in the array will now be checked.
There are 465 elements.
Element nr. 1 is NaN
Element nr. 2 is NaN
Element nr. 3 is NaN
Element nr. 4 is NaN
Element nr. 5 is NaN
Element nr. 6 is NaN
Element nr. 7 is NaN
Element nr. 8 is NaN
Element nr. 9 is NaN
Element nr. 10 is NaN
Element nr. 11 is NaN
Element nr. 12 is NaN
Element nr. 13 is NaN
Element nr. 14 is NaN
Element nr. 15 is NaN
Element nr. 16 is NaN
Element nr. 17 is NaN
Element nr. 18 is NaN
Element nr. 19 is NaN
Element nr. 20 is NaN
Element nr. 21 is NaN
Element nr. 22 is NaN
Element nr. 23 is NaN
Element nr. 24 is NaN
Element nr. 25 is NaN
Element nr. 26 is NaN
Element nr. 27 is NaN
Element nr. 28 is NaN
Element nr. 29 is NaN
Element nr. 30 is NaN
Element nr. 31 is NaN
Element nr. 32 is NaN
Element nr. 33 is NaN
Element nr. 34 is NaN
Element nr. 35 is NaN
Element nr. 36 is NaN
Element nr. 37 is NaN
Element nr. 38 is NaN
Element nr. 39 is NaN
Element nr. 40 is NaN
Element nr. 41 is NaN
Element nr. 42 is NaN
Element nr. 43 is NaN
Element nr. 44 is NaN
Element nr. 45 is NaN
Element nr. 46 is NaN
Element nr. 47 is NaN
Element nr. 48 is NaN
Element nr. 49 is NaN
Element nr. 50 is NaN
Element nr. 51 is NaN
Element nr. 52 is NaN
Element nr. 53 is NaN
Element nr. 54 is NaN
Element nr. 55 is NaN
Element nr. 56 is NaN
Element nr. 57 is NaN
Element nr. 58 is NaN
Element nr. 59 is NaN
Element nr. 60 is NaN
Element nr. 61 is NaN
Element nr. 62 is NaN
Element nr. 63 is NaN
Element nr. 64 is NaN
Element nr. 65 is NaN
Element nr. 66 is NaN
Element nr. 67 is NaN
Element nr. 68 is NaN
Element nr. 69 is NaN
Element nr. 70 is NaN
Element nr. 71 is NaN
Element nr. 72 is NaN
Element nr. 73 is NaN
Element nr. 74 is NaN
Element nr. 75 is NaN
Element nr. 76 is NaN
Element nr. 77 is NaN
Element nr. 78 is NaN
Element nr. 79 is NaN
Element nr. 80 is NaN
Element nr. 81 is NaN
Element nr. 82 is NaN
Element nr. 83 is NaN
Element nr. 84 is NaN
Element nr. 85 is NaN
Element nr. 86 is NaN
Element nr. 87 is NaN
Element nr. 88 is NaN
Element nr. 89 is NaN
Element nr. 90 is NaN
Element nr. 91 is NaN
Element nr. 92 is NaN
Element nr. 93 is NaN
Element nr. 94 is NaN
Element nr. 95 is NaN
Element nr. 96 is NaN
Element nr. 97 is NaN
Element nr. 98 is NaN
Element nr. 99 is NaN
Element nr. 100 is NaN
...too many. I give up here.
There were a total of 465 NANs
[ process 0]: xquit (rc = 128): _INTERNAL_ERROR_
abort:
--- Stop Module: seward at Fri Apr 16 11:24:10 2021 /rc=_RC_INTERNAL_ERROR_ ---
*** files: xmldumpThe input is:
&GATEWAY
Title
Test
coord
$HomeDir/singleatom_C1.xyz
basis=BASES
group=nosymm
BasDir=$CurrDir
AMFI
&SEWARD
&SCFwhere "BASES" is the basis of the atom (S). The geometry is a single S atom. If I change the basis or if I remove the AMFI keyword, everything is ok again.
I could run SEWARD, SCF, RASSCF, CASPT2 and RASSI (no SOC) with this basis set and everything was ok, so I don't know where the problem is. The basis seems to be ok (I also did a small geometry optimization using SCF and the final geometry was close to the geometry provided by other basis)
Any help would be appreciated
Thanks in advance,
Fernando
Offline
#2 2021-04-16 13:02:54
- Ignacio
- Administrator
- From: Uppsala
- Registered: 2015-11-03
- Posts: 1,085
Re: NaN when asking for AMFI
What's the (relevant part of the) BASES file? I tried with a copy of CC-PVDZ and it seemed to work fine.
Offline
#3 2021-04-17 09:51:53
- FAguilar
- Member
- From: San Sebastián, Spain
- Registered: 2021-04-16
- Posts: 5
- Website
Re: NaN when asking for AMFI
Thank you for your answer, Ignacio.
Yes, with other basis it works. I did some tests asking for AMFI with other basis that are in the library and I had no problem.
Here is the file of the basis (only S is defined). What is strange is that it seems to work until I ask for AMFI...
BASES
/S.BASES..30s14p2d1f.5s4p2d1f
Reference none
S (30s14p2d1f) -> [5s4p2d1f]
16.0 3
* S-type functions
30 5
0.3741000000D+06
0.5605000000D+05
0.1276000000D+05
0.3615000000D+04
0.1183000000D+04
0.4288000000D+03
0.1678000000D+03
0.6947000000D+02
0.2984000000D+02
0.1272000000D+02
0.5244000000D+01
0.3615000000D+04
0.1183000000D+04
0.4288000000D+03
0.1678000000D+03
0.6947000000D+02
0.2984000000D+02
0.1272000000D+02
0.5244000000D+01
0.2219000000D+01
0.1183000000D+04
0.1678000000D+03
0.6947000000D+02
0.2984000000D+02
0.1272000000D+02
0.5244000000D+01
0.2219000000D+01
0.3490000000D+00
0.7767000000D+00
0.1322000000D+00
0.5425113228D-04 0.0 0.0 0.0 0.0
0.4211790315D-03 0.0 0.0 0.0 0.0
0.2208660086D-02 0.0 0.0 0.0 0.0
0.9199598750D-02 0.0 0.0 0.0 0.0
0.3213735190D-01 0.0 0.0 0.0 0.0
0.9474510879D-01 0.0 0.0 0.0 0.0
0.2238325198D+00 0.0 0.0 0.0 0.0
0.3748153925D+00 0.0 0.0 0.0 0.0
0.3296923919D+00 0.0 0.0 0.0 0.0
0.8466651172D-01 0.0 0.0 0.0 0.0
-0.1203631365D-02 0.0 0.0 0.0 0.0
0.0 -0.6417861173D-05 0.0 0.0 0.0
0.0 -0.1906784738D-03 0.0 0.0 0.0
0.0 -0.1488973130D-02 0.0 0.0 0.0
0.0 -0.1015373592D-01 0.0 0.0 0.0
0.0 -0.4321086128D-01 0.0 0.0 0.0
0.0 -0.1051992395D+00 0.0 0.0 0.0
0.0 0.3204684289D-01 0.0 0.0 0.0
0.0 0.5225213045D+00 0.0 0.0 0.0
0.0 0.5465588535D+00 0.0 0.0 0.0
0.0 0.0 0.4301058736D-05 0.0 0.0
0.0 0.0 0.2119665796D-03 0.0 0.0
0.0 0.0 0.8527390273D-03 0.0 0.0
0.0 0.0 0.4419341710D-02 0.0 0.0
0.0 0.0 -0.1086181378D-02 0.0 0.0
0.0 0.0 -0.7362011711D-01 0.0 0.0
0.0 0.0 -0.3499851706D+00 0.0 0.0
0.0 0.0 0.1160022394D+01 0.0 0.0
0.0 0.0 0.0 1.0 0.0
0.0 0.0 0.0 0.0 1.0
* P-type functions
14 4
0.5744000000D+03
0.1358000000D+03
0.4319000000D+02
0.1587000000D+02
0.6208000000D+01
0.2483000000D+01
0.1358000000D+03
0.4319000000D+02
0.1587000000D+02
0.6208000000D+01
0.2483000000D+01
0.3229000000D+00
0.8688000000D+00
0.1098000000D+00
0.2495012134D-02 0.0 0.0 0.0
0.1985550782D-01 0.0 0.0 0.0
0.9117959218D-01 0.0 0.0 0.0
0.2622309055D+00 0.0 0.0 0.0
0.4468316811D+00 0.0 0.0 0.0
0.3655010168D+00 0.0 0.0 0.0
0.0 -0.7091748681D-05 0.0 0.0
0.0 -0.1065836619D-02 0.0 0.0
0.0 -0.5910788532D-02 0.0 0.0
0.0 -0.2268816359D-01 0.0 0.0
0.0 -0.1079400613D-01 0.0 0.0
0.0 0.1005933734D+01 0.0 0.0
0.0 0.0 1.0 0.0
0.0 0.0 0.0 1.0
* D-type functions
2 2
0.2690000000D+00
0.8190000000D+00
1.0 0.0
0.0 1.0
* F-type functions
1 1
0.5570000000D+00
1.0
0Offline
#4 2021-04-17 13:06:58
- Ignacio
- Administrator
- From: Uppsala
- Registered: 2015-11-03
- Posts: 1,085
Re: NaN when asking for AMFI
It looks like AMFI doesn't like duplicated exponents. Try writing the basis in proper general contraction:
BASES
/S.BASES..30s14p2d1f.5s4p2d1f
Reference none
S (15s9p2d1f) -> [5s4p2d1f]
16.0 3
* S-type functions
15 5
0.3741000000D+06
0.5605000000D+05
0.1276000000D+05
0.3615000000D+04
0.1183000000D+04
0.4288000000D+03
0.1678000000D+03
0.6947000000D+02
0.2984000000D+02
0.1272000000D+02
0.5244000000D+01
0.2219000000D+01
0.3490000000D+00
0.7767000000D+00
0.1322000000D+00
0.5425113228D-04 0.0 0.0 0.0 0.0
0.4211790315D-03 0.0 0.0 0.0 0.0
0.2208660086D-02 0.0 0.0 0.0 0.0
0.9199598750D-02 -0.6417861173D-05 0.0 0.0 0.0
0.3213735190D-01 -0.1906784738D-03 0.4301058736D-05 0.0 0.0
0.9474510879D-01 -0.1488973130D-02 0.0 0.0 0.0
0.2238325198D+00 -0.1015373592D-01 0.2119665796D-03 0.0 0.0
0.3748153925D+00 -0.4321086128D-01 0.8527390273D-03 0.0 0.0
0.3296923919D+00 -0.1051992395D+00 0.4419341710D-02 0.0 0.0
0.8466651172D-01 0.3204684289D-01 -0.1086181378D-02 0.0 0.0
-0.1203631365D-02 0.5225213045D+00 -0.7362011711D-01 0.0 0.0
0.0 0.5465588535D+00 -0.3499851706D+00 0.0 0.0
0.0 0.0 0.1160022394D+01 0.0 0.0
0.0 0.0 0.0 1.0 0.0
0.0 0.0 0.0 0.0 1.0
* P-type functions
9 4
0.5744000000D+03
0.1358000000D+03
0.4319000000D+02
0.1587000000D+02
0.6208000000D+01
0.2483000000D+01
0.3229000000D+00
0.8688000000D+00
0.1098000000D+00
0.2495012134D-02 0.0 0.0 0.0
0.1985550782D-01 -0.7091748681D-05 0.0 0.0
0.9117959218D-01 -0.1065836619D-02 0.0 0.0
0.2622309055D+00 -0.5910788532D-02 0.0 0.0
0.4468316811D+00 -0.2268816359D-01 0.0 0.0
0.3655010168D+00 -0.1079400613D-01 0.0 0.0
0.0 0.1005933734D+01 0.0 0.0
0.0 0.0 1.0 0.0
0.0 0.0 0.0 1.0
* D-type functions
2 2
0.2690000000D+00
0.8190000000D+00
1.0 0.0
0.0 1.0
* F-type functions
1 1
0.5570000000D+00
1.0
0Offline
#5 2021-04-17 16:52:45
- FAguilar
- Member
- From: San Sebastián, Spain
- Registered: 2021-04-16
- Posts: 5
- Website
Re: NaN when asking for AMFI
Dear Ignacio,
Thank you very much!
I didn't realize that some exponents were present several times (I got it in a different format so I adapted it to Molcas format)... And I didn't know that it was a problem for AMFI!
Just for fun I did a small test, changing randomly (by hand) the numbers, starting from the second digit and I'm still getting the same error.
But with the basis you have provided, it works perfectly!
Thank you again!
Best
Offline
Pages: 1