Which test illustrates SHG calculation in Dalton

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.)

taylor
Posts: 599
Joined: 15 Oct 2013, 05:37
First name(s): Peter
Middle name(s): Robert
Last name(s): Taylor
Affiliation: Tianjin University
Country: China

Re: Which test illustrates SHG calculation in Dalton

Post by taylor » 20 Mar 2020, 08:39

This may be too late to be useful to you, but I would make the following points:
1) Calculated polarizabilities and particularly hyperpolarizabilities are usually strongly affected by the inclusion of correlation effects. Hartree-Fock and most DFT functionals that I am aware of will only give sensible-looking results through cancellation of errors.
2) The inclusion of diffuse functions in calculations is essential, including diffuse polarization functions. Basis sets such as 6-311G are the worst place to start because they are fundamentally poor descriptions of the atomic electronic structure. Augmenting them is not the way to go. Better basis sets are needed from the start.
3) For hyperpolarizabilities there are vibrational as well as electronic contributions that can be significant in polyatomic molecules.
4) Comparison with experiments is a much trickier business than looking at a published experimental number. It is very often the case that the uncertainties given in experimental papers are derived purely from the precision of the original measurement. But the original measurement is not the measurement of the hyperpolarizability! It is a measurement of something (light scattering, SHG, birefringence) from which the experimentalists deduce a hyperpolarizability. Such analysis of experiment includes all sorts of assumptions that may or may not be warranted, and whose influence on the uncertainties in the final result are often ignored. One should be very careful about understanding exactly how the final result was obtained from the original experiment, and what sources of uncertainty might arise.

Best regards
Pete

xiongyan21
Posts: 299
Joined: 24 Sep 2014, 08:36
First name(s): yan
Last name(s): xiong
Affiliation: CENTRAL CHINA NORMAL UNIVERSITY
Country: China

Re: Which test illustrates SHG calculation in Dalton

Post by xiongyan21 » 22 Mar 2020, 03:28

I first used B3LYP 6-311G++(2d,2p) by GAMESS to optimize HF, and then used NWCHEM7.0.0, GAMESS and Dalton2016.2 to calculate the hyperpolarizability of HF.

Dalton2016.2
using the same basis set and c2v symmetry gives
ZDIPLEN (unrel.) -0.0000 ZDIPLEN (unrel.) 0.0000 ZDIPLEN (unrel.) 0.0000 8.8915460
-0.1312 0.0656 0.0656 9.7304868

NWCHEM gives
CCSD Lambda Response polarizability / au
Frequency = 0.0000000 / au
Wavelength = Infinity / nm
beta(Z,Z,Z) = 9.739286463 / au
...
CCSD Lambda Response polarizability / au
Frequency = -0.0656000 / au
Wavelength = -694.5633079 / nm
...
CCSD Quadratic Response
Static Hyperpolarizability ''' ''Why still called static?'''''
-----------------------------------------------
beta(Z,Z,Z) = 8.461760589 / au
The original NWCHEM QA test using the following geometry

symmetry c2v
H 0 0 0
F 0 0 1.7328795 in terms of a.u.

gives

CCSD Quadratic Response
Static Hyperpolarizability
-----------------------------------------------
...
beta(Z,Z,Z) = 8.630759039 / au

1 H 1.0000 0.00000000 0.00000000 -1.55959155
2 F 9.0000 0.00000000 0.00000000 0.17328795



GAMESS HF gives using basis set accdz and c2v symmetry
Beta tensor [in au]( -0.131200; 0.065600, 0.065600)
x y z
...
zz. -0.000000 -0.000000 -10.171063

Reference'' Frequency-dependent hyperpolarizability of hydrogen fluoride'' gives(calculated)
static beta(z,z,z)=-8.058
SHG beta(z,z,z)=-9.430 at Re=1.7328 a.u.

Very Best Regards!
Last edited by xiongyan21 on 02 Apr 2020, 15:49, edited 6 times in total.

xiongyan21
Posts: 299
Joined: 24 Sep 2014, 08:36
First name(s): yan
Last name(s): xiong
Affiliation: CENTRAL CHINA NORMAL UNIVERSITY
Country: China

Re: Which test illustrates SHG calculation in Dalton

Post by xiongyan21 » 22 Mar 2020, 03:31

I redraw the molecule, then use mp2 and 6-311G++(2d,2p) to find the following equilibrium geometry of HF having c2v geometry.
DALTO2016.2 gives
+---------------------------------------------------------------------------------------+
! FINAL CCSD RESULTS FOR THE FIRST HYPERPOLARIZABILITIES !
+---------------------------------------------------------------------------------------+


A operator B operator C operator property
------------------------------------------------------------------------------------------------------

ZDIPLEN (unrel.) -0.0000 ZDIPLEN (unrel.) 0.0000 ZDIPLEN (unrel.) 0.0000 -8.6941344
-0.1312 0.0656 0.0656 -9.5090740



------------------------------------------------------------------------------------------------------


With GAMESS
...
Beta tensor [in au]( -0.000000; 0.000000, 0.000000)
x y z
zx. 0.704144 -0.000000 -0.000000
zz. -0.000000 -0.000000 9.926728
Beta tensor [in au]( -0.131200; 0.065600, 0.065600)
x y z
xz. 0.848328 -0.000000 -0.000000
zx. 0.679234 -0.000000 -0.000000
zz. -0.000000 -0.000000 10.603235

The above reference gives (calculated)
SHG beta(zzz)=-9.430 beta(zxx)=-1.107 beta(xzx)=-1.276
static beta(zzz)=-8.722,-8.750 beta(zxx)=-1.045,-1.102
one indexed data there:static beta(zzz)=-9.838, beta(zxx)=-0.802
SCF:beta(zzz)=-8.808, beta(zxx)=-0.434

NWCHEM7.0.0 gives
CCSD Lambda Response polarizability / au
Frequency = -0.0656000 / au
Wavelength = -694.5633079 / nm
...
Cpu & wall time / sec 0.2 0.2

CCSD Quadratic Response
Static Hyperpolarizability
-----------------------------------------------
beta(X,Z,X) = -0.581141625 / au
beta(Z,X,X) = -0.581141625 / au
beta(Z,Z,Z) = -8.275124709 / au
-----------------------------------------------
CCSD Lambda Response polarizability / au
Frequency = 0.0000000 / au
Wavelength = Infinity / nm
-------------------------------------------------------|-----------------------------------------------

Cpu & wall time / sec 0.1 0.1

CCSD Quadratic Response
Static Hyperpolarizability
-----------------------------------------------
beta(Z,X,X) = -0.770738651 / au
beta(Z,Z,Z) = -9.518810663 / au
-----------------------------------------------
Cpu & wall time / sec 13.1 13.5

Very Best Reards!
Last edited by xiongyan21 on 06 May 2020, 16:46, edited 6 times in total.

xiongyan21
Posts: 299
Joined: 24 Sep 2014, 08:36
First name(s): yan
Last name(s): xiong
Affiliation: CENTRAL CHINA NORMAL UNIVERSITY
Country: China

Re: Which test illustrates SHG calculation in Dalton

Post by xiongyan21 » 22 Mar 2020, 04:23

Perhaps owing to the network traffic problem, I just can read Dr. Taylor's suggestion. Aug-cc-pvdz already adds diffuse.

6-31G is just for warming up exercises, and for the inquiry of units conversion, and agreements of different softwares with the same method.
Measurements of second- and third-order nonlinear polarizabilities for HF and HCI [/i].

Dalton2016.2 HF gives, using d-aug-cc-pv-dz,

B-freq = 0.065600 C-freq = 0.065600 beta(Z;Z,Z) = -8.06145767
B-freq = 0.065600 C-freq = 0.065600 beta(Z;X,X) = 0.11285876
B-freq = 0.065600 C-freq = 0.065600 beta(X;Z,X) = -0.00022904

I am trying to repeat this for HF and calculate this organic molecules in solutions.

Very Best Regards!
Last edited by xiongyan21 on 06 May 2020, 16:52, edited 10 times in total.

taylor
Posts: 599
Joined: 15 Oct 2013, 05:37
First name(s): Peter
Middle name(s): Robert
Last name(s): Taylor
Affiliation: Tianjin University
Country: China

Re: Which test illustrates SHG calculation in Dalton

Post by taylor » 25 Mar 2020, 18:24

I am a bit lost now as to what your question is, or at least what it seems to have evolved into. Are you asserting that three programs that claim to do exactly the same calculation give different results? If so there are ways to check this (certainly in the static limit this can be done using finite-field methods).

If your concern is the lack of agreement between calculation and experiment, I attempted to point out some issues in this respect in my earlier posting. But this is already perhaps going beyond the function of the Dalton mailing forum, which is focussed more in assisting people to use the program. If no computational method gives satisfactory agreement with experiment for your purposes, I don't see how we can do more to help!

Best regards
Pete

xiongyan21
Posts: 299
Joined: 24 Sep 2014, 08:36
First name(s): yan
Last name(s): xiong
Affiliation: CENTRAL CHINA NORMAL UNIVERSITY
Country: China

Re: Which test illustrates SHG calculation in Dalton

Post by xiongyan21 » 29 Mar 2020, 14:12

''Measurements and Calculations of the Hyperpolarizabilities of Atoms and Small Molecules in the Gas Phase gives experimental'' beta=-11.0+-1.0 a.u. at 694.3 nm.

Table XIII. in this Chemical Reviews article gives beta (parallel) of -7.3 a.u. using ab initio.
Last edited by xiongyan21 on 09 Nov 2020, 04:34, edited 15 times in total.

taylor
Posts: 599
Joined: 15 Oct 2013, 05:37
First name(s): Peter
Middle name(s): Robert
Last name(s): Taylor
Affiliation: Tianjin University
Country: China

Re: Which test illustrates SHG calculation in Dalton

Post by taylor » 29 Mar 2020, 14:43

With all due respect, this does not answer my question. I did not ask what the experimental value was, I asked whether you were asserting that different programs gave different results for what should be the same calculation. If that is the case, then there is presumably a problem with one or more of the programs, and if it is a Dalton problem I think this is something that developers reading this forum might be able to help with.

If your concern is that you do not get the same value as experiment, that is an issue I tried to address previously. I think while we try to help when people get the wrong answer, in the sense that Dalton does not work, or gives a manifestly incorrect answer, it is difficult to help when the issue is simply that people don't like the answer! Other than trying to make some suggestions about what possibilities exist for improving the calculation.

Best regards
Pete

xiongyan21
Posts: 299
Joined: 24 Sep 2014, 08:36
First name(s): yan
Last name(s): xiong
Affiliation: CENTRAL CHINA NORMAL UNIVERSITY
Country: China

Re: Which test illustrates SHG calculation in Dalton

Post by xiongyan21 » 29 Mar 2020, 17:12

I am not interested in your question, and I think you can do this easily by yourself.

User avatar
magnus
Posts: 524
Joined: 27 Jun 2013, 16:32
First name(s): Jógvan Magnus
Middle name(s): Haugaard
Last name(s): Olsen
Affiliation: Aarhus University
Country: Denmark

Re: Which test illustrates SHG calculation in Dalton

Post by magnus » 29 Mar 2020, 19:23

Then what exactly are you interested in? What is the point of your posts? This is a Dalton support forum so please stop posting unless you want support.

taylor
Posts: 599
Joined: 15 Oct 2013, 05:37
First name(s): Peter
Middle name(s): Robert
Last name(s): Taylor
Affiliation: Tianjin University
Country: China

Re: Which test illustrates SHG calculation in Dalton

Post by taylor » 01 Apr 2020, 15:23

I am a little surprised at your response after a genuine attempt to understand what it was you were posting about. I have no interest whatsoever in the hyperpolarizability of anything, at least not at present so there is certainly no question for me to answer! I am just trying to understand what you wanted help with. As Magnus has said, if you aren't looking for help, what is the purpose of your posts?

I am genuinely concerned at the possibility that Dalton and other codes do not get the same result for what is nominally the same calculation: this seemed to be what you originally were posting? I am simply trying to establish if there may be a problem with Dalton (or with the other codes).

Best regards
Pete

xiongyan21
Posts: 299
Joined: 24 Sep 2014, 08:36
First name(s): yan
Last name(s): xiong
Affiliation: CENTRAL CHINA NORMAL UNIVERSITY
Country: China

Re: Which test illustrates SHG calculation in Dalton

Post by xiongyan21 » 02 Apr 2020, 04:20

Dear All
Can Dalton calculate nonlinear optical properties of polymers ?
Very Best Regards!

hjaaj
Posts: 394
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: Which test illustrates SHG calculation in Dalton

Post by hjaaj » 02 Apr 2020, 10:18

Fully QM not for bulk, only a single polymer strand, as big as your computer allows. With quite a lot of work you should be able to select one strand for QM and then describe the other polymer strands in the environment by polarizable embedding (the .PEQM option).

xiongyan21
Posts: 299
Joined: 24 Sep 2014, 08:36
First name(s): yan
Last name(s): xiong
Affiliation: CENTRAL CHINA NORMAL UNIVERSITY
Country: China

Re: Which test illustrates SHG calculation in Dalton

Post by xiongyan21 » 02 Apr 2020, 10:29

Dear Prof. Jensen
Thanks a lot.

Very Best Regards!

xiongyan21
Posts: 299
Joined: 24 Sep 2014, 08:36
First name(s): yan
Last name(s): xiong
Affiliation: CENTRAL CHINA NORMAL UNIVERSITY
Country: China

Re: Which test illustrates SHG calculation in Dalton

Post by xiongyan21 » 05 Apr 2020, 12:18

According to Table VIII in reference Molecular vibrational and rotational motion in static and dynamic fields, axial beta(vibrational) calculated can be 10.041,10.99 and 10.51, and beta(averaged over the vibrational gound state) calculated can be -10.505 and -8.47. It is stated in this reference there is a radical change in beta(zzz) when vibration is taken into consideration. It is also said the discrepency between the calculated and the theoretical one is caused by the measured is for HF dimer.
Last edited by xiongyan21 on 12 Nov 2020, 13:04, edited 1 time in total.

xiongyan21
Posts: 299
Joined: 24 Sep 2014, 08:36
First name(s): yan
Last name(s): xiong
Affiliation: CENTRAL CHINA NORMAL UNIVERSITY
Country: China

Re: Which test illustrates SHG calculation in Dalton

Post by xiongyan21 » 06 Apr 2020, 04:59

The aug-cc-pvdz CCSD calculation of beta(ZZZ) of HF using DALTON2016.2 and NWCHEM7.0.0 can give -11.795a.u., and -13.501a.u., respectively. NWCHEM 7.0.0 can give chi(parallel)=-4.53*10^-32 esu, close to the experimental value -4.70*10^-32 esu., but with d-aug-cc-pvdz, CCSD in NWCHEM7.0.0 only gives that of -2.88*10^-32 esu.

No obvious changes caused by the employment of aug-cc-pvdz instead of d-aug-cc-pvdz using Dalton2016.2 and accd instead of accdc using GAMESS.

With HF Dalton2016.2 gives chi(parallel) around -3.58*10^-32 esu.

Very Best Regards!

xiongyan21
Posts: 299
Joined: 24 Sep 2014, 08:36
First name(s): yan
Last name(s): xiong
Affiliation: CENTRAL CHINA NORMAL UNIVERSITY
Country: China

Re: Which test illustrates SHG calculation in Dalton

Post by xiongyan21 » 23 May 2020, 12:26

I think we should be very careful when using the calculated BETA quantum chemically.

Very Best Regards!

xiongyan21
Posts: 299
Joined: 24 Sep 2014, 08:36
First name(s): yan
Last name(s): xiong
Affiliation: CENTRAL CHINA NORMAL UNIVERSITY
Country: China

Re: Which test illustrates SHG calculation in Dalton

Post by xiongyan21 » 10 Nov 2020, 03:10

I am now trying to repeat the gammar of PQM, a singlet diradical system, published in J. Phy. Chem.A.

In this regard, perhaps DALTON is particularly useful.

in Science, an article discussing the first hyperpolarizabilities of some conjugated organic molecules, whose authors are from California Institute of Technology and DuPont de Nemours and Company, is a lilltle challenging, because the molecules are of more than 30 atoms and perhaps have many conformations.

An article by Prof. Mauroulis from Univerity of Patras in Greece discusses the basis set limit and size of correlation effects on the hyperpolarizability of water, compared with those deduced from EFISH measurements. This should be repeated.

Very Best Regards!
Last edited by xiongyan21 on 13 Nov 2020, 05:41, edited 1 time in total.

xiongyan21
Posts: 299
Joined: 24 Sep 2014, 08:36
First name(s): yan
Last name(s): xiong
Affiliation: CENTRAL CHINA NORMAL UNIVERSITY
Country: China

Re: Which test illustrates SHG calculation in Dalton

Post by xiongyan21 » 11 Nov 2020, 09:26

I am using the following input to calculate static beta and gamma of water
**DALTON INPUT
.RUN WAVE FUNCTIONS
**INTEGRAL
.DIPLEN
**WAVE FUNCTIONS
.CC
*CC INPUT
.CCS
.CC2
.CCSD
**RESPONSE
*CUBIC
.DIPLEN
**END OF DALTON INPUT

but at the end, the program stop at
*******************************************************************************
*******************************************************************************
* *
* *
* END OF COUPLED CLUSTER CALCULATION *
* *
* *
*******************************************************************************
*******************************************************************************


CPU and wall time for CC : 0.271 0.295


...


.-------------------------------------.
| End of Coupled Cluster Section (CC) |
`-------------------------------------'

Total CPU time used in DALTON: 0.31 seconds
Total wall time used in DALTON: 0.34 seconds

NWCHEM7.0.0 gives beta(ZZZ) around 0.7/au(<6%) deviating from that from ccsd in Prof. Mauroulis' article if ccsd and aug-cc-pvdz are used, and aug-cc-pvqz beta(ZZZ) is close to CCSD(T) result in that reference. Alpha(XX)(0.1%) , (alpha(yy)(2%), alpha(ZZ)(1%), beta(zxx)(11.9%) and beta(zyy)(1%) also agree, perhaps except for that of beta(zxx).

Using GAMESS, the static beta(2,1,1) with aug -cc-pvtz and pbe0 gives a relative error of a little more than 5% when compared with the beta(ZZZ) in that reference, but I don't know whether they can be compared. When TDHF and 6-311++G(2d,2p) are employed, the static beta(YXX) and beta(XXY) , having identical values, have a relative error within 1% when compared with beta(zzz) metioned just, but beta(ZZZ) is zero. But the second and the third largest ones are far from those.

Pervious hydrogen fluoride beta calculation used the input for hyper-Raman, thus recalculation of this way perhaps is needed.
This may make the TDDFT and TDHF calculation of the molecules in the above Science issue possible.

GAMESS calculation uses C1 symmetry, and C2v symmetry gives nearly the same or not significantly different results using TDDFT, as well as the same numerical results when using TDHF, but the directions are different, e.g., beta(YYY) becomes beta(ZZZ).

The average beta is -17.99 a.u. according to the formula in that reference and data from NWCHEM7.0.0 aug-cc-pvqz and ccsd, compared with -17.51 a.u.(CCSD[T] ) and the experimental values of -19,2+-0.9 a.u or -22.0+-0.9 a.u. indexed there.

From the GAMESS TDHF output, the average beta is 12.45 a.u., lower than the absolute value of the largest component, of which I don't know the algorithmic formula. I will read the GAMESS reference.

Prof. Gordon, et al. once simulated 401 water molecules using TDDFT/MD and explicit solvation model for SHG first hyperpolarizability and two-photo cross section, already published in 2014, in which it seems one can see the large absolute value is a bulk property depending on time from
the figure.


Very Best Regards!
Last edited by xiongyan21 on 15 Nov 2020, 04:44, edited 4 times in total.

xiongyan21
Posts: 299
Joined: 24 Sep 2014, 08:36
First name(s): yan
Last name(s): xiong
Affiliation: CENTRAL CHINA NORMAL UNIVERSITY
Country: China

Re: Which test illustrates SHG calculation in Dalton

Post by xiongyan21 » 13 Nov 2020, 04:39

Actually, it seems the calculation of polarizabilities and hyperpolarizabilities is a challenge for computational chemistry.


Very Best Regards!

xiongyan21
Posts: 299
Joined: 24 Sep 2014, 08:36
First name(s): yan
Last name(s): xiong
Affiliation: CENTRAL CHINA NORMAL UNIVERSITY
Country: China

Re: Which test illustrates SHG calculation in Dalton

Post by xiongyan21 » 21 Nov 2020, 06:31

Before, Dr. Gu in Hiroshima University, and Prof. Aoiki in Department of Mathematics, Hiroshima Kokusai Gakuin University, Prof. IMAMURA in University of Ottawa, Prof. KIRTMAN in Department of Chemistry and Biochemistry , University of California , Santa Barbara, et al, using the elongation method originally proposed by them, a theoretical polymer synthesis tool, have obtained the static alphaes and betaes of water clusters with 3, 11 and 21 water moleculers, respectively by PM3.
I have tried the static and dynamic alpha, static and dynamic beta from SHG, Pockels effect and optical rectification of polyglycerine(n=7), whose geometry is from GAMESS test, with sto-3g and 6-31G by TDHF, respectively, using GAMESS, and found there are only negligible differences of average beta as well as beta-V values, and identical average alpha values, when compared the results from the calculations without elongation, but I am not very sure whether the two methods really work together. I also will try a method told me by Prof. Jensen in this forum with Dalton2018.

Very Best Regard!
Last edited by xiongyan21 on 21 Nov 2020, 13:47, edited 12 times in total.

hjaaj
Posts: 394
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: Which test illustrates SHG calculation in Dalton

Post by hjaaj » 21 Nov 2020, 08:57

**RESPONSE is for MCSCF and DFT response, for CC response you must use the CC response input. See how in the manual.

xiongyan21
Posts: 299
Joined: 24 Sep 2014, 08:36
First name(s): yan
Last name(s): xiong
Affiliation: CENTRAL CHINA NORMAL UNIVERSITY
Country: China

Re: Which test illustrates SHG calculation in Dalton

Post by xiongyan21 » 22 Nov 2020, 03:33

Thanks a lot.
There is another way to deal with clusters and large molecules in GAMESS called fragment molecular orbitals method, and Dr. F. Zahariev in Prof. Grodon's group and Ames Lab has been awarded two hundred thousand supercomputer time to validate the programs developed.

I have already tried two tests from GAMESS, i.e., CCSD(T) energy of a cluster of 6 water molecules with 6-31G(d,p) runnning parallelly, finishing less than one minute, and the gradient calculation of 3 glycerine molecules using parameters which can produce cube files, either a sparse or a Gaussian one. I have not tried how to use them to calculate properties.
I have also tried the test of the RHF analytic gradient of sixty four water molecules using EFMO with 6-31++G(d,p), obtaining dipole moments, LMO alphaes, dispersion interaction energies of separated pairs, etc., simulataneously runing on 3 threads within 4 minitues. I have got the Hessian and total energy with negligible difference from those in the original test. Here, imaginary frequency polarizabilities appear. In addition. the different orbital localization method may make a difference according to the outputs. The mp2 gradient of sixty four water molecules with 6-31++G(d,p) has also been tried using the input from the test of GAMESS, within 8.5 minitues running on 3 threads, also getting the results with negligible difference from those in the original test.

Can LSDATON do similarly?

Very Best Regards

xiongyan21
Posts: 299
Joined: 24 Sep 2014, 08:36
First name(s): yan
Last name(s): xiong
Affiliation: CENTRAL CHINA NORMAL UNIVERSITY
Country: China

Re: Which test illustrates SHG calculation in Dalton

Post by xiongyan21 » 23 Nov 2020, 12:04

Can LSDALTON calculate nonlinear optical properties?

Very Best Regards!

Post Reply

Who is online

Users browsing this forum: No registered users and 7 guests