Dissociation energy for the NO molecule

Click For Summary

Discussion Overview

The discussion centers on the dissociation energy of the NO molecule, specifically seeking benchmarks for post-Hartree-Fock calculations and equilibrium energy values using the cc-pVTZ basis set. Participants share their computational experiences and results related to geometry optimization and energy calculations.

Discussion Character

  • Technical explanation
  • Mathematical reasoning
  • Debate/contested

Main Points Raised

  • One participant requests benchmarks for the dissociation energy of the NO molecule, specifically looking for results using the cc-pVTZ basis set.
  • Another participant shares their geometry optimization results at the CCSD(T)/cc-pVTZ level, providing a final energy value.
  • Questions arise about the speed and efficiency of the calculation, with a participant noting that CCSD(T) is fast for small systems like NO but scales poorly for larger systems.
  • Participants discuss the performance of different computational programs, highlighting that some may handle certain basis sets better than others.
  • There is an inquiry into the efficiency of CCSDT compared to CCSD(T), with a reference to benchmarks that may use modified basis sets, noting discrepancies in energy results.

Areas of Agreement / Disagreement

Participants express varying opinions on the efficiency and performance of different computational methods and programs, indicating that no consensus exists on the best approach or the accuracy of the results presented.

Contextual Notes

Some participants mention the limitations of specific computational methods and the dependence on the choice of basis sets, which may affect the results. There is also an acknowledgment of the variability in performance across different computational programs.

Morberticus
Messages
82
Reaction score
0
Hi,

Does anyone know where I can find post-hartree-fock calculation benchmarks for dissociation energy of the NO molecule. Or even just the equilibrium energy of the NO molecule?

I am looking for cc-pVTZ basis set results but I can only seem to find augmented basis set results.

Thanks.
 
Chemistry news on Phys.org
I did a geometry optimisation in qchem at CCSD(T)/cc-pVTZ level. Hope this helps.
Z-matrix Print:
$molecule
0,2
1 N
2 O 1 1.154418
$end
Final energy is -129.70273042654063
 
TAMEPJLAH said:
I did a geometry optimisation in qchem at CCSD(T)/cc-pVTZ level. Hope this helps.
Z-matrix Print:
$molecule
0,2
1 N
2 O 1 1.154418
$end
Final energy is -129.70273042654063

That was fast! Thanks. How long did this calculation take? How many cores? It seems much faster than the method I was trying.
 
Morberticus said:
That was fast! Thanks. How long did this calculation take? How many cores? It seems much faster than the method I was trying.
This calculation should be a matter of seconds on a single core.

Note the horrific scaling of CCSD(T) (7th order). That means it gets very slow quickly for larger systems. That also means that it is very fast for small systems, like NO.

As a side note: The performance of different programs differs widely for different methods and basis sets. For example, many programs cannot deal well with generally contracted basis sets (like cc-pVnZ), and then it can easily happen that one program is spending two days calculating integrals where another program does the same job in five minutes...
So you generally want to use a program which is known to handle a certain kind of task well (e.g., Molpro is a good choice for coupled cluster calculations).
 
I think this computer has 4 cores, but as cgk said this calculation is fast for small systems. I also recommend cfour for cupled cluster calculations, its free of charge for non-commercial users :) http://www.cfour.de/
 
Interesting. Would either of you know how efficient a CCSDT, as opposed to CCSD(T) calculation would be? I ask because I have found a set of benchmarks, but I think they are using slightly tweaked basis sets (They optimise the polarisation exponent).

http://adsabs.harvard.edu/abs/2000JChPh.113..485F

Their result is half an eV off mine, while the result you reported is almost identical to mine (I used a CI based method).
 

Similar threads

Replies
1
Views
3K
  • · Replies 0 ·
Replies
0
Views
2K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 6 ·
Replies
6
Views
4K
  • · Replies 18 ·
Replies
18
Views
24K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 6 ·
Replies
6
Views
8K
Replies
0
Views
2K
  • · Replies 16 ·
Replies
16
Views
3K
  • · Replies 4 ·
Replies
4
Views
3K