Molecule ' s Energy

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

Molecule ' s Energy

Daniel Hoegerl
Hi everyone,

I'll immediately start, so:

I'm trying to calculate the gibbs energy of a molecule, and first found the class OBMol and it's method "GetEnergy()". I'm creating the OBMol object from a SMILES string. However, GetEnergy() allways (well, I tried with a lot molecules) returns 0.0.

Then I recognized the OBFieldForce class (example used in the obenergy tool) and I got rather large energy values in kJ/mol for molecules of at most 240 amu. Finally I built in the energy minimizing step (using the pFF->ConjugateGradients(10000)), but only the energy values of small molecules became smaller - the larger molecules stayed at their high positive energies (i.e. > 1e10).

So my first question is: what is this energy in a chemical sense - I thought it to be the "heat of formation", oder the free enthalpy or even the gibbs energy!?

And why is OBMol->GetEnergy() always 0.0?

Thanks in advance,

Daniel




------------------------------------------------------------------------------
SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
http://p.sf.net/sfu/solaris-dev2dev
_______________________________________________
OpenBabel-discuss mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
Reply | Threaded
Open this post in threaded view
|

Re: Molecule ' s Energy

Noel O'Boyle
Administrator
On 14 February 2010 19:10, Daniel Hoegerl <[hidden email]> wrote:
> Hi everyone,
>
> I'll immediately start, so:
>
> I'm trying to calculate the gibbs energy of a molecule, and first found the class OBMol and it's method "GetEnergy()". I'm creating the OBMol object from a SMILES string. However, GetEnergy() allways (well, I tried with a lot molecules) returns 0.0.
>
> Then I recognized the OBFieldForce class (example used in the obenergy tool) and I got rather large energy values in kJ/mol for molecules of at most 240 amu. Finally I built in the energy minimizing step (using the pFF->ConjugateGradients(10000)), but only the energy values of small molecules became smaller - the larger molecules stayed at their high positive energies (i.e. > 1e10).
>
> So my first question is: what is this energy in a chemical sense - I thought it to be the "heat of formation", oder the free enthalpy or even the gibbs energy!?

Molecular mechanics forcefield energies are not derived from
thermodynamics. They treat bonds and angles (etc..) as a system of
'springs' (see Cramer or Jensen's Computational Chemistry textbooks
for more background). So what's the point? Well, their relative values
can be used to determine the most stable conformer for example. I'm
not sure whether it's meaningful to compare the FF energy for one
molecule to the FF energy for another, though.

> And why is OBMol->GetEnergy() always 0.0?

I think this value is only non-zero if you are reading the result of a
QM calculation (e.g. a Gaussian log file) (we welcome documentation
patches!).

> Thanks in advance,
>
> Daniel
>
>
>
>
> ------------------------------------------------------------------------------
> SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
> Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
> http://p.sf.net/sfu/solaris-dev2dev
> _______________________________________________
> OpenBabel-discuss mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
>

------------------------------------------------------------------------------
SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
http://p.sf.net/sfu/solaris-dev2dev
_______________________________________________
OpenBabel-discuss mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
Reply | Threaded
Open this post in threaded view
|

Re: Molecule ' s Energy

Chris Morley-3
In reply to this post by Daniel Hoegerl
People from various branches of chemistry have contributed to
OpenBabel and the terms they use are not always compatible.

If you were expecting numbers like Enthalpy of Formation of methane at
300K to be about -74.8 kJ/mol, and wanting to estimate that for any
molecule (or even handle it), I am afraid OpenBabel cannot help much
at present.

Elsewhere I have implemented thermodynamic property estimation from
structure by group additivity and have in mind to port it to
OpenBabel. However I have had this in mind for a very long time, so
don't hold your breath.

OBMol::GetEnergy() accesses data previously input from from, for
example, Gaussian format. It is a zero-point energy on a computational
chemistry baseline (is this separated electrons and nuclei?) and needs
much reworking to put it on the thermodynamic basis I think you are
interested in.

Chris

Daniel Hoegerl wrote:

> Hi everyone,
>
> I'll immediately start, so:
>
> I'm trying to calculate the gibbs energy of a molecule, and first found the class OBMol and it's method "GetEnergy()". I'm creating the OBMol object from a SMILES string. However, GetEnergy() allways (well, I tried with a lot molecules) returns 0.0.
>
> Then I recognized the OBFieldForce class (example used in the obenergy tool) and I got rather large energy values in kJ/mol for molecules of at most 240 amu. Finally I built in the energy minimizing step (using the pFF->ConjugateGradients(10000)), but only the energy values of small molecules became smaller - the larger molecules stayed at their high positive energies (i.e. > 1e10).
>
> So my first question is: what is this energy in a chemical sense - I thought it to be the "heat of formation", oder the free enthalpy or even the gibbs energy!?
>
> And why is OBMol->GetEnergy() always 0.0?

------------------------------------------------------------------------------
SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
http://p.sf.net/sfu/solaris-dev2dev
_______________________________________________
OpenBabel-discuss mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
Reply | Threaded
Open this post in threaded view
|

Re: Molecule ' s Energy

Geoffrey Hutchison
In reply to this post by Daniel Hoegerl
> Then I recognized the OBFieldForce class (example used in the obenergy tool) and I got rather large energy values in kJ/mol for molecules of at most 240 amu. Finally I built in the energy minimizing step (using the pFF->ConjugateGradients(10000)), but only the energy values of small molecules became smaller - the larger molecules stayed at their high positive energies (i.e. > 1e10).

You've gotten two good answers. The third would be -- what force field are you using? In principal, MMFF94 is intended to reproduce heats of formation, although Noel indicated a number of problems with this. I have no idea what UFF energies are supposed to indicate -- the paper suggests that relative energies are supposed to reproduce activation barriers, but I don't think that's been tested.

If you really want fairly accurate heats of formation, I suggest a semiempirical quantum calculation, for example with MOPAC (e.g., AM1 or PM6) which has been specifically parameterized for such a task.

BTW, if you read MOPAC output files with Open Babel, then the OBMol energy will reflect the heat of formation calculated by MOPAC.

Best regards,
-Geoff

P.S. This particular property will be disappearing in Open Babel v3. Energy is a function of the computation method, not the molecule itself.
------------------------------------------------------------------------------
SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
http://p.sf.net/sfu/solaris-dev2dev
_______________________________________________
OpenBabel-discuss mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/openbabel-discuss