UT theoretical chemistry code forum

Dear all:

I have two question:

1)Is it credible to find a transition state with negative activation energy ?

2)Is it possibale to optmize a smaller lattice parameter than the experoment one using GGA? for example, the experimental lattice parameter of Ni is 3.524 Angstrom, but my VASP optimised a 3.521Angstrom with considering the spin effect(3.508 without considering spin effect). I also found Eichler obtained 3.53Angstrom. we all use VASP code and PW91.

thanks

1) I can't see a situation that would result in a saddle point with an energy less than the initial minimum. By definition, the initial state should be reached by minimizing from the saddle point.

That said, within the general transition state theory, you can choose a valid transition state (an N-1 dimensional surface separating reactants from products) with a free energy that is less than a surface near the initial state.

2) Although there can be trends in how GGA compares with experiment (lower barriers, smaller band gaps), I don't think you can say anything for certain. The DFT lattice constant for a material could be larger or smaller than the experimental value.