Difference between properties value

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
ndhieu92
Posts: 28
Joined: 04 Sep 2019, 07:51
First name(s): Hieu
Middle name(s): Dinh
Last name(s): Nguyen
Affiliation: Kunsan Univ
Country: Korea, Republic of

Difference between properties value

Post by ndhieu92 » 03 Oct 2019, 09:58

Good day everyone,

I am having a difficult thing to understand about the properties in RESPONSE part. I am trying to calculate spin-orbit coupling (SOC) constants for my molecules and am wondering that if I do not figure out the property: X1SPNORB in the RESPONSE part, so dalton will calculate the SOC values with both one and two electron integrals, am I right?

hjaaj
Posts: 326
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: Difference between properties value

Post by hjaaj » 03 Oct 2019, 10:56

I am not quite sure what you mean. Anyway, ".SPIN-ORB" kyword for linear response will include both 1-electron and 2-electron spin-orbit. The "X1SPNORB" is the label for the 1-electron spin-orbit x-component and if you specify that as a property operator, then that specification will only include the 1-electron part.

ndhieu92
Posts: 28
Joined: 04 Sep 2019, 07:51
First name(s): Hieu
Middle name(s): Dinh
Last name(s): Nguyen
Affiliation: Kunsan Univ
Country: Korea, Republic of

Re: Difference between properties value

Post by ndhieu92 » 03 Oct 2019, 13:48

hjaaj wrote:
03 Oct 2019, 10:56
I am not quite sure what you mean. Anyway, ".SPIN-ORB" kyword for linear response will include both 1-electron and 2-electron spin-orbit. The "X1SPNORB" is the label for the 1-electron spin-orbit x-component and if you specify that as a property operator, then that specification will only include the 1-electron part.
I got it. Thank you so much for your reply

ndhieu92
Posts: 28
Joined: 04 Sep 2019, 07:51
First name(s): Hieu
Middle name(s): Dinh
Last name(s): Nguyen
Affiliation: Kunsan Univ
Country: Korea, Republic of

Re: Difference between properties value

Post by ndhieu92 » 04 Oct 2019, 06:17

hjaaj wrote:
03 Oct 2019, 10:56
I am not quite sure what you mean. Anyway, ".SPIN-ORB" kyword for linear response will include both 1-electron and 2-electron spin-orbit. The "X1SPNORB" is the label for the 1-electron spin-orbit x-component and if you specify that as a property operator, then that specification will only include the 1-electron part.
Dear Dr. Hans Jorgen

May I ask you one more thing?

As I know if I use LINEAR and SINGLE RESIDUE, it will be used to calculate between ground and excited states and QUADRATIC and DOUBLE RESIDUE is used to calculate between excited states. However, if i use QUADRATIC and DOUBLE RESIDUE, I just only can get the putput with the transition moments, there is no value of spin-orbit matrix elements.

- For example, the output will look like if I use LINEAR and SINGLE RESIDUE:
@ Operator type: X SPNORB
@ Spin-orbit coupling constant (X SPNORB) : -0.22251827 cm-1
-1.01386786E-06 au

@ Operator type: Y SPNORB
@ Spin-orbit coupling constant (Y SPNORB) : -0.43363459 cm-1
-1.97578458E-06 au

@ Operator type: Z SPNORB
@ Spin-orbit coupling constant (Z SPNORB) : -0.33808853 cm-1
-1.54044470E-06 au

- However, when I use the QUADRATIC and DOUBLE RESIDUE, the output will look like:
@ Transition moment <B | A - <A> | C> in a.u. for
@ A operator label, symmetry, spin: Y1SPNORB 2 0
@ B excited state no., symmetry, spin: 1 1 0
@ C excited state no., symmetry, spin: 1 2 0

@ B and C excitation energies, moment: 0.29276589 0.24693346 -0.00000000.

Please let me know how to deal with this issue.
THANK YOU SO MUCH

hjaaj
Posts: 326
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: Difference between properties value

Post by hjaaj » 04 Oct 2019, 09:42

I do not think double residue with spin-orbit has been tested/used by any of the authors of that part. Anyway, I tried, and with the following input for a test case:

Code: Select all

**RESPONSE
.TRPFLG
*QUADRATIC
.DOUBLE
.SPNORX
.ISPABC
 1 0 1
.ROOTS
 1 0 1 0
**END OF DALTON INPUT
DALTON delivered in the output the total spin-orbit coupling (x-component) between the excited states.
Note that you must specify the spins of A, B, C in <B | A-<A> I C> with the .ISPABC keyword, default is all three singlet. I can see from your output slice that all three operators were treated as singlet.

ndhieu92
Posts: 28
Joined: 04 Sep 2019, 07:51
First name(s): Hieu
Middle name(s): Dinh
Last name(s): Nguyen
Affiliation: Kunsan Univ
Country: Korea, Republic of

Re: Difference between properties value

Post by ndhieu92 » 04 Oct 2019, 12:45

hjaaj wrote:
04 Oct 2019, 09:42
I do not think double residue with spin-orbit has been tested/used by any of the authors of that part. Anyway, I tried, and with the following input for a test case:

Code: Select all

**RESPONSE
.TRPFLG
*QUADRATIC
.DOUBLE
.SPNORX
.ISPABC
 1 0 1
.ROOTS
 1 0 1 0
**END OF DALTON INPUT
DALTON delivered in the output the total spin-orbit coupling (x-component) between the excited states.
Note that you must specify the spins of A, B, C in <B | A-<A> I C> with the .ISPABC keyword, default is all three singlet. I can see from your output slice that all three operators were treated as singlet.
Yes. I also just looked in the manual and see about .SPINORX. Thank you so much for your recommendation. I will keep trying for my calculation. Wish you have a good weekend.

hjaaj
Posts: 326
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: Difference between properties value

Post by hjaaj » 04 Oct 2019, 12:49

a comment: I used .SPNORX, only the x-component of the spin-orbit coupling, instead of .SPIN-ORBIT, because in my test case only the x-component gives a non-zero contribution between symmetries 1 and 3.

ndhieu92
Posts: 28
Joined: 04 Sep 2019, 07:51
First name(s): Hieu
Middle name(s): Dinh
Last name(s): Nguyen
Affiliation: Kunsan Univ
Country: Korea, Republic of

Re: Difference between properties value

Post by ndhieu92 » 05 Oct 2019, 03:42

hjaaj wrote:
04 Oct 2019, 12:49
a comment: I used .SPNORX, only the x-component of the spin-orbit coupling, instead of .SPIN-ORBIT, because in my test case only the x-component gives a non-zero contribution between symmetries 1 and 3.
Hi, Dr. Hans Jogen. Thank you for your comment, in my case, from the paper, it gave three components with non-zero contribution between singlet and triplet excited states. I have tried with a set of .SPINORX .SPINORY .SPINORZ. I try with small molecule like benzene, but I am trying to figure out why it give 3 same values of spin-orbit coupling. Also you can check the output file in attachment.
Attachments
SPINORBIT_rsplrso_quadratic_spnorb_SOC_benzene.out
(73.28 KiB) Downloaded 2 times

ndhieu92
Posts: 28
Joined: 04 Sep 2019, 07:51
First name(s): Hieu
Middle name(s): Dinh
Last name(s): Nguyen
Affiliation: Kunsan Univ
Country: Korea, Republic of

Re: Difference between properties value

Post by ndhieu92 » 05 Oct 2019, 03:46

The value looks like this:

QRTRVE -- linear response calculation for symmetry 1
QRTRVE -- operator label : X SPNORB
QRTRVE -- frequencies : -0.064666
--
QRTRVE -- linear response calculation for symmetry 1
QRTRVE -- operator label : Y SPNORB
QRTRVE -- frequencies : -0.064666
--
QRTRVE -- linear response calculation for symmetry 1
QRTRVE -- operator label : Z SPNORB
QRTRVE -- frequencies : -0.064666
Last edited by ndhieu92 on 05 Oct 2019, 04:42, edited 2 times in total.

ndhieu92
Posts: 28
Joined: 04 Sep 2019, 07:51
First name(s): Hieu
Middle name(s): Dinh
Last name(s): Nguyen
Affiliation: Kunsan Univ
Country: Korea, Republic of

Re: Difference between properties value

Post by ndhieu92 » 05 Oct 2019, 03:47

tt
Last edited by ndhieu92 on 05 Oct 2019, 04:42, edited 1 time in total.

ndhieu92
Posts: 28
Joined: 04 Sep 2019, 07:51
First name(s): Hieu
Middle name(s): Dinh
Last name(s): Nguyen
Affiliation: Kunsan Univ
Country: Korea, Republic of

Re: Difference between properties value

Post by ndhieu92 » 05 Oct 2019, 03:47

tt

hjaaj
Posts: 326
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: Difference between properties value

Post by hjaaj » 07 Oct 2019, 19:20

The three values of -0.064666 are NOT the spin-orbit coupling. The value is the difference between the triplet and singlet excitation energies, and that difference enters the quadratic response formula for the transition moment - see the theory. That difference is of course the same for x, y, and z spin-orbit.

The spin-orbit coupling components are printed at the bottom of your output file in a.u., you must convert them yourself to get them in other units. They are of course very small for benzene. From your output (irrelevant lines deleted, spin-orbit coupling colored red):
@ Transition moment <B | A - <A> | C> in a.u. for
@ A operator label, symmetry, spin: X SPNORB 1 1
@ B excited state no., symmetry, spin: 1 1 0
@ C excited state no., symmetry, spin: 1 1 1

@ B and C excitation energies, moment: 0.20438757 0.13972172 -0.00000001

@ Transition moment <B | A - <A> | C> in a.u. for
@ A operator label, symmetry, spin: Y SPNORB 1 1
@ B excited state no., symmetry, spin: 1 1 0
@ C excited state no., symmetry, spin: 1 1 1

@ B and C excitation energies, moment: 0.20438757 0.13972172 0.00000000

@ Transition moment <B | A - <A> | C> in a.u. for
@ A operator label, symmetry, spin: Z SPNORB 1 1
@ B excited state no., symmetry, spin: 1 1 0
@ C excited state no., symmetry, spin: 1 1 1

@ B and C excitation energies, moment: 0.20438757 0.13972172 0.00000015
The x and y spin-orbit should be the same by symmetry, the small difference is numerical noise.

ndhieu92
Posts: 28
Joined: 04 Sep 2019, 07:51
First name(s): Hieu
Middle name(s): Dinh
Last name(s): Nguyen
Affiliation: Kunsan Univ
Country: Korea, Republic of

Re: Difference between properties value

Post by ndhieu92 » 08 Oct 2019, 02:43

hjaaj wrote:
07 Oct 2019, 19:20
The three values of -0.064666 are NOT the spin-orbit coupling. The value is the difference between the triplet and singlet excitation energies, and that difference enters the quadratic response formula for the transition moment - see the theory. That difference is of course the same for x, y, and z spin-orbit.

The spin-orbit coupling components are printed at the bottom of your output file in a.u., you must convert them yourself to get them in other units. They are of course very small for benzene. From your output (irrelevant lines deleted, spin-orbit coupling colored red):
@ Transition moment <B | A - <A> | C> in a.u. for
@ A operator label, symmetry, spin: X SPNORB 1 1
@ B excited state no., symmetry, spin: 1 1 0
@ C excited state no., symmetry, spin: 1 1 1

@ B and C excitation energies, moment: 0.20438757 0.13972172 -0.00000001

@ Transition moment <B | A - <A> | C> in a.u. for
@ A operator label, symmetry, spin: Y SPNORB 1 1
@ B excited state no., symmetry, spin: 1 1 0
@ C excited state no., symmetry, spin: 1 1 1

@ B and C excitation energies, moment: 0.20438757 0.13972172 0.00000000

@ Transition moment <B | A - <A> | C> in a.u. for
@ A operator label, symmetry, spin: Z SPNORB 1 1
@ B excited state no., symmetry, spin: 1 1 0
@ C excited state no., symmetry, spin: 1 1 1

@ B and C excitation energies, moment: 0.20438757 0.13972172 0.00000015
The x and y spin-orbit should be the same by symmetry, the small difference is numerical noise.
Yes, Thank you so much Dr. Hans Jorgen

ndhieu92
Posts: 28
Joined: 04 Sep 2019, 07:51
First name(s): Hieu
Middle name(s): Dinh
Last name(s): Nguyen
Affiliation: Kunsan Univ
Country: Korea, Republic of

Re: Difference between properties value

Post by ndhieu92 » 08 Oct 2019, 06:08

hjaaj wrote:
04 Oct 2019, 12:49
a comment: I used .SPNORX, only the x-component of the spin-orbit coupling, instead of .SPIN-ORBIT, because in my test case only the x-component gives a non-zero contribution between symmetries 1 and 3.
Dear Dr. Hans Jorgen.

I am trying to calculate the SOC value for another molecule with the atomic mean-field approximation method. However, for these 2 molecules, the output occurred some errors. Could you please help me to understand how to fix these errors?

First output: SPINORBIT_rsplrso_AMFI_quadratic_SOC_BDELA

" there are totally 714 functions
but maxorbs= 700

--- SEVERE ERROR, PROGRAM WILL BE ABORTED ---
Date and time (Linux) : Tue Oct 8 12:12:32 2019
Host name : node02

@ MPI MASTER, node no.: 0
@ Reason: increase maxorbs in symtrafo "

Second output: SPINORBIT_rsplrso_AMFI_SOC_BDEGA

" --- SEVERE ERROR, PROGRAM WILL BE ABORTED ---
Date and time (Linux) : Tue Oct 8 10:28:45 2019
Host name : node07

@ MPI MASTER, node no.: 0
@ Reason: amfi symtra ERROR: numbofcent > maxcent "

Thank you so much and Best regards.
Attachments
SPINORBIT_rsplrso_AMFI_SOC_BDEGA.out
(54.8 KiB) Downloaded 1 time
SPINORBIT_rsplrso_AMFI_quadratic_SOC_BDELA.out
(43.87 KiB) Downloaded 1 time

Post Reply

Who is online

Users browsing this forum: No registered users and 6 guests