In most cases, STO integrals are evaluated assuming that the orbitals are directed towards each other, using spherical harmonics; however, in GTO-based software Cartesian GTOs are used, with the orbitals being aligned along the x, y or z axis.
Is there any way to evaluate STO integrals using Cartesian STOs, or convert between Cartesian and spherical STOs? For example, if I had a pz-type STO at (0,0,0) and another one at (1,1,1), is there any way to decompose the resultant expression in terms of spherical STO integrals?