MAGNETIZATION IN CRYSTALS — DILLON 



399 



M^ftfr? 



c. 



a. 



Figure 8. — Possible magnetization distributions in a crystal in which the magnetization is 

 constrained to lie perpendicular to the surface. 



strain. In such a specimen the domain structure is fantastically 

 fragile. Merely touching the crystal with a single bristle from a fine 

 camel's-hair brush will completely alter the structure obseived. 

 Stretching or compressing or bending any of these samples will radi- 

 cally affect the magnetization. If a simple stress is applied, the effects 

 can often be easily miderstood. 



Consider a crystal plate which suffers from a uniform surface com- 

 pression as just discussed, such that its easy direction of magnetiza- 

 tion is perpendicular to the faces. Wliy is its magnetization not 

 merely all in one direction in one domain the size of the crystal as 

 in figure 8 (a) ? The answer lies in the minimization of what is called 

 demagnetizing energy. If a sample has the magnetization as in figure 

 8{a), there is a magnetic field extending out into space around the 

 sample. It requires energy to set up a magnetic field in space — so- 

 called field energy, and it depends on the strength of the field and the 

 volume of space. If the magnetization breaks up into a ribbon pat- 

 tern as in figure 8(&) , the intensity of the fields involved will be about 

 the same, but they will fill a much smaller volume of space. Thus 

 by breaking up into ribbon domains, the field energy part of the total 

 magnetic energy has been drastically reduced. 



The reader might ask why the process does not continue, say, to 

 the condition (c), where the ribbons are even smaller and the field 

 energy is even lower. It turns out that the domain walls have energy, 



