THE PHYSICAL BASIS OF FERROMAGNETISM 9 



energy per atom is 



kd = 1.4 X 10->3 erg= 0.09 electron-volt 



and the internal field 



Hi = 7,000,000 oersteds. 



Although this field is much stronger than any so far produced in the 

 laboratory, the energy involved is small compared to that which 

 controls chemical binding. For example, the energy of ionization of 

 the helium atom is about 25 electron volts. Another way of showing 

 that the magnetic forces are small compared to the electrostatic 

 forces holding atoms together, is to compare the Curie temperature 

 with the temperature of vaporization. 



The calculation of magnetic forces by theory is thus extremely 

 difficult, because they are but small additions to the electrostatic 

 forces which themselves cannot usually be calculated with much 

 precision. 



Ewing's Theory 



Ewing 2 was one of the first to attempt to explain ferromagnetic 

 phenomena in terms of the forces between atoms. His theory will be 

 described briefly here, since many physicists today, when thinking 

 about magnetic phenomena, still go back to Ewing's ideas of fifty 

 years ago. He assumed with Weber that each atom was a permanent 

 magnet free to turn in any direction about its center. The orientations 

 of the various magnets with respect to the field and to each other 

 were supposed to be due entirely to the mutual magnetic forces. 

 The I, H curve and hysteresis loop were calculated for a linear group 

 of such magnets and were determined experimentally using models 

 having as many as 130 magnets arranged at the points of a plane 

 square lattice. 



The calculations for a linear chain show that as the field is gradually 

 increased in magnitude from zero there is at first a slow continuous 

 rotation of the magnets, then a sudden change in orientation and 

 finally a further continuous rotation until the magnets lie parallel to 

 the field. The I, H curves calculated for such a group of magnets 

 resemble in general form the actual curves of iron : they show a perme- 

 ability first increasing then decreasing, and saturation and hysteresis. 



A magnetization curve and a hysteresis loop obtained ^ with a 



model of 130 magnets in square array, are shown in Fig. 6. Experi- 



"J. A. Ewing summarized in "Magnetic Induction in Iron and Other Metals," 

 The Electrician, London, 3d ed. (1900). 



3 J. A. Ewing and H. G. Klaassen, Phil. Trans. Roy. Soc, 184A, 985-1039 (1893). 



