What causes the warnings in one kind of ECD calculation but not in another

Find answers or ask questions regarding Dalton calculations.
Please upload an output file showing the problem, if applicable.
(It is not necessary to upload input files, they can be found in the output file.)

Post Reply
xiongyan21
Posts: 286
Joined: 24 Sep 2014, 08:36
First name(s): yan
Last name(s): xiong
Affiliation: CENTRAL CHINA NORMAL UNIVERSITY
Country: China

What causes the warnings in one kind of ECD calculation but not in another

Post by xiongyan21 » 02 Oct 2020, 05:39

In A combined experimental and theoretical study on the formation of interstellar propylene oxide -----a chiral molecule, whose authors are from the United States, published in The Astrophysical Journal, 2018, A chiral molecule is studied.

I use Dalton2018 to calculate its ECD. The geometry is optimized using mp2 and 6-311++G(2d,2p) with GAMESS. There are warnings after the ECD calculation in the terminal, also when nosymmetry is used
88 WARNINGS have been issued by DALTON,
output from 'grep -n -i WARNING' (max 10 warnings):
1245: *** WARNING Complex eigenvalue in reduced response eq. of symmetry 1
1246: *** WARNING real and imaginary part : 48 37 1.605971D-16 4.035815D-02
1248: *** WARNING Complex eigenvalue in reduced response eq. of symmetry 1
1249: *** WARNING real and imaginary part : 48 38 1.580593D-16 -4.035815D-02
1251: *** WARNING Complex eigenvalue in reduced response eq. of symmetry 1
1252: *** WARNING real and imaginary part : 56 44 1.052252D-16 3.209610D-02
1254: *** WARNING Complex eigenvalue in reduced response eq. of symmetry 1
1255: *** WARNING real and imaginary part : 56 45 1.040126D-16 -3.209610D-02
1257: *** WARNING Complex eigenvalue in reduced response eq. of symmetry 1
1258: *** WARNING real and imaginary part : 64 61 -9.924182D-18 3.380033D-02
...

What are the problems?

For the one including frequencies, there are no warnings, only the following message leaving in the terminal
1 informational messages have been issued by DALTON,
output from 'grep -n INFO' (max 10 lines):
238: INFO: Request for parallel calculation (.PARALL) is ignored because only one CPU process available.


The log files are attached.

Very Bet Regards!
Attachments
po1.log
(80.54 KiB) Downloaded 7 times
po.log
(82.82 KiB) Downloaded 4 times
pof.log
(96.11 KiB) Downloaded 11 times
Last edited by xiongyan21 on 07 Oct 2020, 12:33, edited 3 times in total.

hjaaj
Posts: 387
Joined: 27 Jun 2013, 18:44
First name(s): Hans Jørgen
Middle name(s): Aagaard
Last name(s): Jensen
Affiliation: Universith of Southern Denmark
Country: Denmark

Re: What causes the warnings in one kind of ECD calculation but not in another

Post by hjaaj » 06 Oct 2020, 14:38

This indicates that you have a so-called singlet instability, which is not seen that often. It means that you can get lower HF energy by using complex orbitals (not implemented in Dalton), similar to that a triplet instability means UHF gives lower energy than RHF.
Conclusion: you need to use MCSCF to get reasonable values.

xiongyan21
Posts: 286
Joined: 24 Sep 2014, 08:36
First name(s): yan
Last name(s): xiong
Affiliation: CENTRAL CHINA NORMAL UNIVERSITY
Country: China

Re: What causes the warnings in one kind of ECD calculation but not in another

Post by xiongyan21 » 10 Oct 2020, 06:09

I have used mp2 natural orbitals for this. Does it exacerbate the situation?

Very Best regards!

hjaaj
Posts: 387
Joined: 27 Jun 2013, 18:44
First name(s): Hans Jørgen
Middle name(s): Aagaard
Last name(s): Jensen
Affiliation: Universith of Southern Denmark
Country: Denmark

Re: What causes the warnings in one kind of ECD calculation but not in another

Post by hjaaj » 12 Oct 2020, 09:50

Your attached output files above are all for HF.
I assume you have performed MCSCF calculations since then, and choosing active space based on MP2 occupation number improve the situation in all calculations I have seen. Note: if two MP2 occupation numbers are close, e.g. 1.974 and 1.977, you should include both, but so accurate is MP2 not. With respect to "close": for the strongly occupied orbitals you should judge from "2 - occupation number", i.e. 0.026 and 0.023 in may example above.

Post Reply

Who is online

Users browsing this forum: No registered users and 6 guests