Composite field (mathematics)

From testwiki
Jump to navigation Jump to search

A composite field or compositum of fields is an object of study in field theory. Let K be a field, and let E1, E2 be subfields of K. Then the (internal) composite[1] of E1 and E2 is the field defined as the intersection of all subfields of K containing both E1 and E2. The composite is commonly denoted E1E2.

Properties

Equivalently to intersections we can define the composite E1E2 to be the smallest subfield[2] of K that contains both E1 and E2. While for the definition via intersection, well-definedness hinges only on the property that intersections of fields are themselves fields, here two auxiliary assertions are included. Firstly, that there exist minimal subfields of K that include E1 and E2 and secondly, that such a minimal subfield is unique and therefore justly called the smallest.

It also can be defined using field of fractions

E1E2=E1(E2)=E2(E1),

where F(S) is the set of all F-rational expressions in finitely many elements of S.[3]

Let LE1E2 be a common subfield and E1/L a Galois extension then E1E2/E2 and E1/(E1E2) are both also Galois and there is an isomorphism given by restriction

Gal(E1E2/E2)Gal(E1/(E1E2)),σσ|E1.

For finite field extension this can be explicitly found in Milne[4] and for infinite extensions this follows since infinite Galois extensions are precisely those extensions that are unions of an (infinite) set of finite Galois extensions.[5]

If additionally E2/L is a Galois extension then E1E2/L and (E1E2)/L are both also Galois and the map

ψ:Gal(E1E2/L)Gal(E1/L)×Gal(E2/L),σ(σ|E1,σ|E2)

is a group homomorphism which is an isomorphism onto the subgroup

H={(σ1,σ2):σ1|E1E2=σ2|E1E2}=Gal(E1/L)×Gal((E1E2)/L)Gal(E2/L)Gal(E1/L)×Gal(E2/L).

See Milne.[6]

Both properties are particularly useful for L=E1E2 and their statements simplify accordingly in this special case. In particular ψ is always an isomorphism in this case.

External composite

When E1 and E2 are not regarded as subfields of a common field then the (external) composite is defined using the tensor product of fields.[7] Note that some care has to be taken for the choice of the common subfield over which this tensor product is performed, otherwise the tensor product might come out to be only an algebra which is not a field.

Generalizations

If ={Ei:iI} is a set of subfields of a fixed field K indexed by the set I, the generalized composite field[8] can be defined via the intersection

iIEi=FK s.t. iI:EiFF.

Notes

Template:Reflist

References


Template:Abstract-algebra-stub

  1. Roman, p. 42.
  2. Roman, p. 42.
  3. Template:Cite web
  4. Milne, p. 40; take into account the preliminary definition of Galois as finite on p. 37
  5. Milne, p. 93 and 99
  6. Milne, p. 41 and 93
  7. Template:SpringerEOM
  8. Roman, p. 42.