The aim of this article is to give a practical way for the use of real spherical functions in another frame than the frame in which they have been defined. For instance, we calculate physical properties from a local frame and use them in the general frame, deduced one from the other by Eulerian rotations on the coordinates. The power of the method is in the use of Cartesian coordinates and in the definition of a scalar product between these functions to set up the complete vectorial space generated by these representations of the group 0;.
I. AimsIn the early days of quantum chemistry state of art were ab initio CI calculations on 1s and 2s 2p functions for a few atoms. In the field of catalysis, as well as in solid state calculations for magnetism or spin-dependent properties, one is forced to consider the 3d "chemistry" of Fe, Co, Ni, Cu, and in the VIII metal group, Ni, Pd, and Pt by also considering relativistic effects. The nd multiplets give rise to many difficult problems that can be solved by, say, Wigner-Racah algebra and Clebsch-Gordan coupling coefficients [l-41, and the mf configurations for rare earths are even more complex [5-61. For binuclear properties one can use a kind of Slater-Koster parameter [6-91 but when we need only a one function property to study them from a local point of view these methods are inadequate. Rare earth calculations are now of great interest not only in the uranide [5] group but also for material science components in new high-technology innovative materials: catalysis, superconductivity, magnets and maglevs, postcombustion catalysts, optical glasses, solar cells, optic fibers, electric bulbs, ceramic oxygen sensors, and, color centers in cathodic displays [lo]. All these applications need a convenient description of nd-nf real functions in order to understand the specificity of these atomic properties.