Hi Hyunbok Lee,|
I don't exactly know the difference, I guess that it should be related to the fact that in some cases the whole set of basis functions are not independent (i.e. some of them are just a linear combination of the rest). Therefore, gaussian makes the distinction between the total number of basis functions ("basis functions") and the number of those linearly independent ("independent basis functions"). Normally, I have the same number in both sets (i.e. all basis functions are linearly independent). Obviously, when there is a difference, both sets will give rise to the same results so the smaller one ("independent basis functions") is taken for computational efficiency.
The problem seem to strive in how gaussview gets the number of functions to be read, as it apparently get it from "Number of basis functions" while in the chk file the information refers to the independent set. For example, in your case, if you go to Line Number 1737, as the error says, you'll see the number of Alpha Orbital Energies equals the "Number of independent basis functions" (recall that within the Rothaan-Hall approach, one obtains the same number of molecular orbitals as basis functions where used) not the "Number of basis functions". As the program was expecting a larger number of orbital energies than the actual one, it sorts out that error.
Well, this is just my appreciation (so error-prone) and still did not answer why one can get such linear dependencies in the basis set. But hope it gives you an idea to start with.
El 18/07/12 08:50, Hyunbok Lee mutebeat*|*gmail.com escribió:
Javier CEREZO BASTIDA
Universidad de Murcia