Hello Mejdeddine,
for non-periodic calculations, you need to set "PERIODIC NONE" in &CELL subsection too, to avoid pair interactions. Also, the cell need to be grosser than the system density. I guess at large distances the system will be Na⁺ and Cl⁻... for that the cell need to be, in the bond direction, the bond distance + twice the ionic radii of the biggest ion to ensure the density is zero at the borders, this is ~3.5 A due to the ionic radii of Cl⁻ is 1.7 A.
Using WAVELET solver needs a cubic box, so, if you want to compute the system at 10.0 A distance, the box will be 17.0A and cubic. With a cell of 10.0 A, you will have problems of non-zero density at 4.0 A distance, I guess.
Otherwise, the EPSs values must be equal for SCF and OUT_SCF, and the DEFAULT could be tightest (I use order of -6 and -12 respectively). And, the number of inner cycles, is useful less steps to use the outer cycles and improve the convergence.
Changing those things I get:
DIST (A) ENERGY(Ha)
1.0 -60.135198879741978
1.5 -61.855322129770521
2.0 -62.197077891951885
2.5 -62.227919856240661
3.0 -62.207241440696443
3.5 -62.182709877371700
4.0 -62.162755189608077
4.5 -62.148301846426023
5.0 -62.138459966222577
5.5 -62.131738730461414
6.0 -62.126565992226809
6.5 -62.122559283798338
7.0 -62.119778885873323
7.5 -62.118151081860589
8.0 -62.117267918233409
8.5 -62.116553100338109
9.0 -62.115268838231749
9.5 -62.114008578635961
10.0 -62.113128520139142
The minima is near 2.5 A, but I guess 2.6 A is closer... It is a coarse grid for the potential profile.
Also, the sentence "tend to 1.5 eV even at a large distance.", according to what I understand this value is arbitrary. At calculations this energy will be the energy of isolated Cl⁻ plus isolated Na⁺... But is a pseudopotential calculation, changing the pseudo or functional, the value will change. The important value is the energy depth of the minima with respect to the large converged distance. The large distance would be a coulomb r⁻¹ behavior.
I attach the input that I use to obtain list values.
Regards - Lucas Lodeiro