THE PHYSICAL BASIS OF FERROMAGNETISM 



21 



45 shell which then becomes half full as shown by the line "Cu" as 

 well as by {g) of Fig. 10. The diagram does not show changes in the 

 relative levels of the 3c^+ and Zd— bands that occur in going from 

 one element to another; when both 3d bands are filled, as in copper, 

 these levels are the same. The numbers of electrons and "holes" in 

 metals near iron in the periodic table are given in Table I. A more 



TABLE I 



Number of Electrons and Vacancies (Holes) in Various Shells 

 IN Metal Atoms Near Iron in the Periodic Table 



accurate determination of the form of the 3d and 45 bands for copper 

 is given in Fig. 12, due to Slater.^ 



An especially simple and interesting illustration of the atom-model 

 described is afforded by the alloys of nickel and copper. The substitu- 

 tion of one copper for one nickel atom in the lattice is equivalent to 

 adding one electron to the alloy. This electron seeks the place of 

 lowest energy in the alloy and finds it in the 3c?-shell of a nickel atom 

 rather than in the copper atom to which it originally belonged. This 

 lowers the magnetic saturation of the alloy by one Bohr unit, since 

 the added electron in the 3d— band just neutralizes the moment of one 

 in the M-\- band. Addition of more copper to nickel decreases the 

 average moment until the empty spaces in the 3d— band are just full; 

 this occurs when 60 per cent of the atoms are copper, and then the 

 magnetic saturation at 0° K will be just zero. This is the explanation 

 of the experimental results ^ shown in Fig. 13. There are shown also 

 the saturation moments for other alloys of nickel; it is evident that 

 zinc with two 45 electrons fills up the 3d band twice as fast as copper, 

 aluminum three times as fast, silicon and tin four times and antimony 

 five, in good accord with theory. In each of these cases the added 



» J. C. Slater, Phys. Rev., 49, 537-545 (1936). 



9V. Marian, Ann. de Physique (11), 7, 459-527 (1937). Some of the data for 

 the other alloys shown in Fig. 12 are taken from C. Sadron, Ann. de Physique, 17, 

 371-452 (1932). The interpretation of these results is due to E. C. Stoner, ref. 6. 



