Orbitals of a molecule enclosed by charged points


Clicked A Few Times
Dear NWChem developers

Can NWChem calculate DFT (or HF) energies and orbitals of a molecule enclosed by charged points, like as QM/MM?

My best,
Mikiya

Forum Vet
Mikiya,

Yes, NWChem can do that. See http://nwchemgit.github.io/index.php/Release61:QMMM for details on our QM/MM implementation.

Depending on the version of NWChem you are using, you can also add arbitrary point charges (for example to generate a Madelung potential). See http://nwchemgit.github.io/index.php/Release62:Bq . This the documentation of the upcoming release, but this functionality should already be in NWChem-6.1.1

Bert

Clicked A Few Times
Bert,

Great!! Thanks!! I am using 6.1.1 and will try it!

Mikiya

Clicked A Few Times
BQ-Strange
Dear developers;

I am new user of NWCHEM (less than a week) and my major is solid state.
Firstly thanks for your nice job of BQ which is very useful;
As I seen the
http://nwchemgit.github.io/index.php/Release62:Bq

The reading order ins "x y z q" and this also what I found in
/src/bq inside one of the subroutine (I am not sure).

However when I tried my example as well as example in the link the reading order
looks to be "q x y z".

To describe more I used the example in the link:

INPUT <<

bq
2.384   1.738   1.380  -0.9
2.448 1.608 0.416 0.45
1.560 1.268 1.608 0.45
end

>> OUTPUT

Number of centers:   3
1 Bq 1.73800000 1.38000000 -0.90000000 charge 2.38400000
2 Bq 1.60800000 0.41600000 0.45000000 charge 2.44800000
3 Bq 1.26800000 1.60800000 0.45000000 charge 1.56000000
Total Bq charge: 6.39200000000000

I tried in a different way means::

<< INPUT
  q        x       y      z
-0.9 2.384 1.738 1.380
0.45 2.448 1.608 0.416
0.45 1.560 1.268 1.608

>> OUTPUT

Name: default                                                                 
Number of centers: 3
1 Bq 2.38400000 1.73800000 1.38000000 charge -0.90000000
2 Bq 2.44800000 1.60800000 0.41600000 charge 0.45000000
3 Bq 1.56000000 1.26800000 1.60800000 charge 0.45000000
Total Bq charge: 0.000000000000000E+000

Now it looks fine.

I thought it may come from writing in fortran routine but this not the case.

I put the q=0 to see whether I can get the same calculation as no-charge calculation.
The result is, YES I can get it when I put the charge in the first column.

Originally, I did test for CH4 (-1), and increasing the charge (+) should bring down the
HOMO because of Madelung potential and the result is same. I have correct results
by locating charge in first column.

I am not familiar with the NWCHEM routines and I couldn't find the bug therefore
I report it.

Please let me know I am correct and column should be displace or not.

Best regards;
Shahab
--
ººººººººººººººººººººººººººººººººººººººººººººººººººººººººº
S. Shahab Naghavi
(SISSA)
Scuola Internazionale Superiore di Studi Avanzati
Via Bonomea 265, Tireste, Italy
Phone: +39-403787448
Email(1): snaghavi@sissa.it
ººººººººººººººººººººººººººººººººººººººººººººººººººººººººº

Forum Vet
There have been some inconsistencies, which will be resolved in the upcoming release. In NWChem 6.1 the format is indeed "q x y z".

Bert


Forum >> NWChem's corner >> NWChem functionality