Derivation-invariant not implies characteristic: Difference between revisions

From Groupprops
(Created page with '{{Lie subring property non-implication| stronger = derivation-invariant Lie subring| weaker = characteristic subring of a Lie ring}} ==Definition== There can exist a [[Lie ring…')
 
Line 31: Line 31:
In other words:
In other words:


<math>d(x,y) = (d_{11}(x) + d_{21}(y), d_{12}(x) + d_{22}(y))</math>.
<math>\! d(x,y) = (d_{11}(x) + d_{21}(y), d_{12}(x) + d_{22}(y))</math>.


The derivation condition states that:
The derivation condition states that:


<math>d[(x,y),(x',y')] = [d(x,y),(x',y')] + [(x,y),d(x',y')]</math>.
<math>\! d[(x,y),(x',y')] = [d(x,y),(x',y')] + [(x,y),d(x',y')]</math>.


This gives:
This gives:


<math>(d_{11}([x,x']) + d_{21}([y,y']), d_{12}([x,x']) + d_{22}([y,y'])) = ([d_{11}(x),x'] + [d_{21}(y),x']+ [x,d_{11}(x')] + [x,d_{12}(y')], [d_{12}(x),y'] + [d_{22}(y),y']) + [y,d_{21}(x')] + [y,d_{22}(y')])</math>.
<math>\! (d_{11}([x,x']) + d_{21}([y,y']), d_{12}([x,x']) + d_{22}([y,y'])) = ([d_{11}(x),x'] + [d_{21}(y),x']+ [x,d_{11}(x')] + [x,d_{12}(y')], [d_{12}(x),y'] + [d_{22}(y),y']) + [y,d_{21}(x')] + [y,d_{22}(y')])</math>.


We thus get:
We thus get:


<math>d_{11}([x,x']) + d_{21}([y,y'])  = [d_{11}(x),x'] + [d_{21}(y),x']+ [x,d_{11}(x')] + [x,d_{12}(y')]</math>
<math>\! d_{11}([x,x']) + d_{21}([y,y'])  = [d_{11}(x),x'] + [d_{21}(y),x']+ [x,d_{11}(x')] + [x,d_{12}(y')]</math>


and:
and:


<math>d_{12}([x,x']) + d_{22}([y,y']) = [d_{12}(x),y'] + [d_{22}(y),y']) + [y,d_{21}(x')] + [y,d_{22}(y')]</math>.
<math>\! d_{12}([x,x']) + d_{22}([y,y']) = [d_{12}(x),y'] + [d_{22}(y),y']) + [y,d_{21}(x')] + [y,d_{22}(y')]</math>.


Setting <math>y = y' = 0</math> gives that <math>d_{11}</math> is a derivation. Setting <math>x = x' = 0</math> gives that <math>d_{22}</math> is a derivation. Plugging these back in, we get:
Setting <math>y = y' = 0</math> gives that <math>d_{11}</math> is a derivation. Setting <math>x = x' = 0</math> gives that <math>d_{22}</math> is a derivation. Plugging these back in, we get:


<math>d_{21}([y,y'])  = [d_{21}(y),x']+ [x,d_{12}(y')]</math>
<math>\! d_{21}([y,y'])  = [d_{21}(y),x']+ [x,d_{12}(y')]</math>


and:
and:


<math>d_{12}([x,x']) = [d_{12}(x),y'] + [y,d_{21}(x')]</math>.
<math>\! d_{12}([x,x']) = [d_{12}(x),y'] + [y,d_{21}(x')]</math>.


Setting <math>y' = 0</math> in the first equation gives that <math>[d_{21}(y),x'] = 0</math> for all <math>x',y</math>, implying that <math>d_{21}</math> takes values in the center of <math>A_1</math>. Similarly, setting <math>x' = 0</math> in the second equation gives that <math>d_{12}(x)</math> is in the center of <math>A_2</math>. In particular, this implies that:
Setting <math>y' = 0</math> in the first equation gives that <math>[d_{21}(y),x'] = 0</math> for all <math>x',y</math>, implying that <math>d_{21}</math> takes values in the center of <math>A_1</math>. Similarly, setting <math>x' = 0</math> in the second equation gives that <math>d_{12}(x)</math> is in the center of <math>A_2</math>. In particular, this implies that:


<math>d(x,0) = (d_{11}(x), d_{12}(x))</math>
<math>\! d(x,0) = (d_{11}(x), d_{12}(x))</math>


takes values in <math>A_1 \oplus Z(A_2) = A_1 + Z(L) = S</math>.
takes values in <math>A_1 \oplus Z(A_2) = A_1 + Z(L) = S</math>.


Thus, <math>d(A_1) \subseteq S</math>. Since <math>Z(L)</math> is derivation-invariant by fact (1), <math>d(Z(L)) \subseteq Z(L)</math>, so <math>d(S) = d(A_1) + d(Z(L)) \subseteq S + Z(L) = S</math>. Thus, <math>S</math> is a derivation-invariant subring of <math>L</math>.
Thus, <math>d(A_1) \subseteq S</math>. Since <math>Z(L)</math> is derivation-invariant by fact (1), <math>d(Z(L)) \subseteq Z(L)</math>, so <math>d(S) = d(A_1) + d(Z(L)) \subseteq S + Z(L) = S</math>. Thus, <math>S</math> is a derivation-invariant subring of <math>L</math>.

Revision as of 01:51, 18 July 2009

This article gives the statement and possibly, proof, of a non-implication relation between two Lie subring properties. That is, it states that every Lie subring satisfying the first Lie subring property (i.e., derivation-invariant Lie subring) need not satisfy the second Lie subring property (i.e., characteristic subring of a Lie ring)
View a complete list of Lie subring property non-implications | View a complete list of Lie subring property implications
Get more facts about derivation-invariant Lie subring|Get more facts about characteristic subring of a Lie ring

Definition

There can exist a Lie ring with a subring such that is a derivation-invariant Lie subring of , such that is not a characteristic subring of .

Related facts

Facts used

  1. Center is derivation-invariant

Proof

Suppose is a non-abelian Lie ring, are isomorphic copies of , and is the direct sum . Define . Then, is derivation-invariant but not characteristic.

Proof that the subring is not characteristic

Consider the coordinate exchange automorphism that interchanges and . Under this automorphism, goes to . Since is non-abelian, it is not contianedin , so the image of is not equal to it.

Proof that the subring is derivation-invariant

Consider a derivation . There exist four abelian group endomorphisms that describe , namely:

.

In other words:

.

The derivation condition states that:

.

This gives:

.

We thus get:

and:

.

Setting gives that is a derivation. Setting gives that is a derivation. Plugging these back in, we get:

and:

.

Setting in the first equation gives that for all , implying that takes values in the center of . Similarly, setting in the second equation gives that is in the center of . In particular, this implies that:

takes values in .

Thus, . Since is derivation-invariant by fact (1), , so . Thus, is a derivation-invariant subring of .