Re: CCL:accuracy of relative energies with DFT

 --- Marcel Swart <m.swart*&*few.vu.nl> wrote:
 > Hi Silviu,
 >
 > as you can see in my Mol.Phys. 102 (2004) p. 2467
 > paper, PBE gives an
 > accuracy of 11.8 kcal/mol for the atomization
 > energies of the G2/97 set,
 > when used with a TZP basis set;
 > note also that there are other GGA functionals that
 > perform much better.
 >
 > For the atomization energies, you test the absolute
 > value of the energy,
 > while for the energies of the different isomers, you
 > compare the
 > relative
 > energy; I would expect the error in the atomization
 > energy to be much
 > larger
 > than that in the relative energy of isomers; the
 > latter is probably
 > smaller
 > than the 1.0 kcal/mol you mentioned.
 Hi Dr. Swart, is this error just your guess or you
 have some evidence, either theoretical or
 experimental?
 Thanks.
 Vincent
 > NB.
 > Please let me know if you have trouble locating the
 > paper,
 > and I will send you an off-print.
 >
 > On Jun 22, 2005, at 11:30 PM, Silviu Zilberman
 > wrote:
 >
 > > Hi,
 > >
 > > I have been studying transition metal complexes
 > with DFT (PBE
 > > xc-functional). I think that a typical accuracy of
 > PBE is around 10
 > > kcal/mol.
 > >
 > > I was wondering, when comparing very similar
 > isomers, should I expect
 > > higher accuracy in the relative total energies. I
 > have seen papers
 > > reporting differences of ~1 kcal/mol, but is it
 > reliable?
 > >
 > > Thanks, SIlviu.
 >
 >
 > dr. Marcel Swart
 >
 > Theoretische Chemie
 > Vrije Universiteit Amsterdam
 > Faculteit der Exacte Wetenschappen
 >
 > De Boelelaan 1083
 > 1081 HV Amsterdam
 > The Netherlands
 >
 > Tel		+31-(0)20-5987619
 > Fax		+31-(0)20-5987629
 > E-mail	m.swart*&*few.vu.nl
 > Web		http://www.few.vu.nl/~swart
 >
 >
 >
 __________________________________________________
 Do You Yahoo!?
 Tired of spam?  Yahoo! Mail has the best spam protection around
 http://mail.yahoo.com