Why Oxygen is not supported by pcS-n?

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
keyares

Why Oxygen is not supported by pcS-n?

Post by keyares » 14 Mar 2015, 22:34

In reference Basis Set Convergence of Nuclear Magnetic Shielding Constants Calculated by Density Functional Methods http://140.123.79.88/~silvercy/Data/Reference/pcs.pdf
pcS-n can work on B-Ne

Why I can not run it on Oxygen?


ERROR: Nuclear charge 8 is an unsupported element for basis pcS-3
ERROR: You must choose another basis set, or enter it manually.



--- SEVERE ERROR, PROGRAM WILL BE ABORTED ---
Date and time (Linux) : Sat Mar 14 22:23:03 2015

My input Mol:

BASIS
pcS-2
b3lypme
pcS-2 nmr
ATOMTYPES=3
CHARGE=8.0 ATOMS=1
O 1.0776786488 0.0383655824 -0.2038735241
CHARGE=6.0 ATOMS=2
C_1 -0.3434413042 -0.0069972700 -0.1398305225
C_2 -0.8640033981 -1.2601826139 0.5587148884
CHARGE=1.0 ATOMS=6
H_1 -0.6872058673 0.0183066175 -1.1791160867
H_2 -0.7462933425 0.8955197546 0.3488847451
H_3 1.4082325380 0.0539650541 0.7050084320
H_4 -0.4953463933 -2.1600951072 0.0574576838
H_5 -1.9598154222 -1.2874711730 0.5586140418
H_6 -0.5278114591 -1.2934258446 1.6023343424

My input dal:
**DALTON INPUT
.RUN PROPERTIES
*PCM
.SOLVNT
CLFORM
*PCMCAV
**WAVE FUNCTIONS

.DFT
mPW1PW91
*CONFIGURATION INPUT
.SYMMETRY
1
.SPIN MUL
1
**PROPERTIES
.SHIELD
.SPIN-SPIN
*SPIN-S
0.90
**END OF DALTON INPUT


Thank you very much!

RikaK
Posts: 11
Joined: 09 Jun 2014, 03:40
First name(s): Rika
Last name(s): Kobayashi
Affiliation: Australian National University
Country: Australia

Re: Why Oxygen is not supported by pcS-n?

Post by RikaK » 14 Mar 2015, 23:35

Hello,
I haven't checked Dalton2015 yet but I can run your inputs without that error (though you have an error in your input later on):

Atomic type no. 1
--------------------
Nuclear charge: 8.00000
Number of symmetry independent centers: 1
Number of basis sets to read; 2
Basis set file used for this atomic type with Z = 8 :
"/home/900/rxk900/dalton/jobs/basis/pcS-2"

but I needed to provide my own pcS bases. I got mine from the EMSL basis-set exchange but I think Frank Jensen told me later that he is happy to provide them in dalton format on request.
Rika

keyares

Re: Why Oxygen is not supported by pcS-n?

Post by keyares » 15 Mar 2015, 11:13

Thank you very much. So I just send an email to Dr Frank Jensen to ask if he could provide me this basis file?

keyares

Re: Why Oxygen is not supported by pcS-n?

Post by keyares » 15 Mar 2015, 11:17

RikaK wrote:Hello,
I haven't checked Dalton2015 yet but I can run your inputs without that error (though you have an error in your input later on):

Atomic type no. 1
--------------------
Nuclear charge: 8.00000
Number of symmetry independent centers: 1
Number of basis sets to read; 2
Basis set file used for this atomic type with Z = 8 :
"/home/900/rxk900/dalton/jobs/basis/pcS-2"

but I needed to provide my own pcS bases. I got mine from the EMSL basis-set exchange but I think Frank Jensen told me later that he is happy to provide them in dalton format on request.
Rika
Thank you very much

taylor
Posts: 545
Joined: 15 Oct 2013, 05:37
First name(s): Peter
Middle name(s): Robert
Last name(s): Taylor
Affiliation: Tianjin University
Country: China

Re: Why Oxygen is not supported by pcS-n?

Post by taylor » 15 Mar 2015, 11:39

Rika has given you a pretty good pointer in the right direction, and I'm just adding a couple of comments here. The situation with respect to these basis sets is somewhat complicated, which is in no way either Frank Jensen's or the Dalton developers' fault. Well, possibly partly a Dalton issue in that the EMSL library "Dalton export" format is not ideal for the Dalton program, and we won't discuss that further here because there are multiple postings about it, and a quick look (full disclosure: I wrote it) at the little utility aces2dalton that comes with the Dalton release will cover all the details.

Anyway, the Dalton release comes with a partial set of pc-X type basis sets, but Frank Jensen produced a much more comprehensive set that was released via EMSL, all optimized according to a particular set of criteria. Subsequent to that, Frank has basically regenerated sets according to the same criteria, but with more optimized contraction patterns. I am sure that Frank would be delighted to provide you with these sets (you see how easy it is to make work for other people? That's how you get to be director of a supercomputer centre...) but you can extract the sets from EMSL in the old contraction patterns and use aces2dalton to create basis libraries for yourself.

Either way, you should make sure in your researches going forward that you are making apples-to-apples comparisons! That is, you should either use the original Dalton-supplied basis sets throughout, or you should download replacements and use those throughout. The "old" sets and the current ones will differ, and results will not necessarily be consistent.

Best regards
Pete

RikaK
Posts: 11
Joined: 09 Jun 2014, 03:40
First name(s): Rika
Last name(s): Kobayashi
Affiliation: Australian National University
Country: Australia

Re: Why Oxygen is not supported by pcS-n?

Post by RikaK » 15 Mar 2015, 11:46

keyares wrote:Thank you very much. So I just send an email to Dr Frank Jensen to ask if he could provide me this basis file?
Yes, I would say so (Frank may even see this forum) unless the other developers know of an official file. In the meantime, you could have a play with the attached. It's my basis from EMSL which only contains atoms I was interested in and may require proofreading. I remember having some formatting issues.
Attachments
pcS-2.txt
(12.83 KiB) Downloaded 203 times

keyares

Re: Why Oxygen is not supported by pcS-n?

Post by keyares » 15 Mar 2015, 11:59

taylor wrote:Rika has given you a pretty good pointer in the right direction, and I'm just adding a couple of comments here. The situation with respect to these basis sets is somewhat complicated, which is in no way either Frank Jensen's or the Dalton developers' fault. Well, possibly partly a Dalton issue in that the EMSL library "Dalton export" format is not ideal for the Dalton program, and we won't discuss that further here because there are multiple postings about it, and a quick look (full disclosure: I wrote it) at the little utility aces2dalton that comes with the Dalton release will cover all the details.

Anyway, the Dalton release comes with a partial set of pc-X type basis sets, but Frank Jensen produced a much more comprehensive set that was released via EMSL, all optimized according to a particular set of criteria. Subsequent to that, Frank has basically regenerated sets according to the same criteria, but with more optimized contraction patterns. I am sure that Frank would be delighted to provide you with these sets (you see how easy it is to make work for other people? That's how you get to be director of a supercomputer centre...) but you can extract the sets from EMSL in the old contraction patterns and use aces2dalton to create basis libraries for yourself.

Either way, you should make sure in your researches going forward that you are making apples-to-apples comparisons! That is, you should either use the original Dalton-supplied basis sets throughout, or you should download replacements and use those throughout. The "old" sets and the current ones will differ, and results will not necessarily be consistent.

Best regards
Pete
Thank you very much!
I will backup the old version.

frj
Posts: 11
Joined: 09 May 2014, 08:37
First name(s): Frank
Last name(s): Jensen
Affiliation: Aarhus University
Country: Denmark

Re: Why Oxygen is not supported by pcS-n?

Post by frj » 15 Mar 2015, 17:20

The most recent version of the pcS basis sets are with segmented contraction (pcSseg-n), and they are available for all elements H-Kr, in qualities from 0-4 (unpolarized DZ to polarized 5Z), and they are avaiilable from the EMSL.
Unfortunately the Dalton basis set format has changed between versions, and not all are compatible with the EMSL supplied Dalton format.
I'll be happy to supply the basis sets in what I believe is the most current Dalton format, just send me an email (frj@chem.au.dk).
Frank

Post Reply

Who is online

Users browsing this forum: No registered users and 6 guests