Scribd
Upload
28 views2 pages

Cartan'scriteria

The document discusses Cartan's criteria for solvability and semi-simplicity in Lie algebras, presenting theorems and proofs related to the properties of these algebraic structures. It establishes conditions under which a Lie algebra is solvable or semi-simple, including the implications of the Killing form. The document also explains the significance of semi-simplicity and provides a framework for understanding the structure of Lie algebras through ideals and their relationships.

Uploaded by

army forever india
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
28 views2 pages

Cartan'scriteria

The document discusses Cartan's criteria for solvability and semi-simplicity in Lie algebras, presenting theorems and proofs related to the properties of these algebraic structures. It establishes conditions under which a Lie algebra is solvable or semi-simple, including the implications of the Killing form. The document also explains the significance of semi-simplicity and provides a framework for understanding the structure of Lie algebras through ideals and their relationships.

Uploaded by

army forever india
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Cartan’s criteria

Sunaina Pati
July 10, 2024

§1 Cartan’s criteria for solvability


Theorem 1.1. Suppose L is a complex solvable lie algebra. Suppose L is a lie algebra. Then

L is solvable ⇐⇒ ∀x, y, z ∈ L, k ([x, y ], z ) = 0.

Proof. Suppose L is solvable. Then consider ad : L → gl (L). Note that im(ad) is a


quotient and homomorphism of L, hence solvable. So there exists a basis of L such
that ∀w ∈ L, adw is upper triangular matrix. Hence [adx , ady ] is upper triangular
matrix. Hence [adx , ady ]adz is is upper triangular matrix. So T r ([adx , ady ]adz ) = 0 =⇒
k ([x, y ], z ) = 0.
Suppose ∀x, y, z ∈ L, we have k ([x, y ], z ) = 0. We will show that L0 is nilpotent. Note
that w = [x, y ] ∈ L0 =⇒ k (w, z ) = 0. We know that by jordan decomposition,
adw = (adw )d + (adw )n = Wd + W − n. But note that tr (W · Wd ) = tr (Wd · Wd ) =
|λ1 |2 + · · · + |λm |2 and λi is eigenvalue of W . But tr (W · Wd ) = tr (ad[x,y ] · adwd ) = 0.
So the eigenvalues are 0. So L0 is nilpotent. So L is solvable.

§2 Cartan’s criteria for semi simplicity


Definition 2.1. W ⊥ = {x ∈ W |k (x, w ) = 0∀w ∈ W } where k is the killing form and
W is subspace of lie algebra L.

Theorem 2.2. If I is ideal of L then I ⊥ is ideal of L.

Proof. Let x ∈ L and j ∈ I ⊥ . We want to show [j, x] ∈ I ⊥ . We want to show

tr (ad[j,x] · adi ) = 0∀i ∈ I.

But we know that


tr ([a, b]c) = tr (a[b, c]).
So
tr (ad[j,x] · adi ) = tr (adj · ad[x,i] ) but [x, i] ∈ I =⇒ tr (adj · ad[x,i] ) = 0.

Theorem 2.3. Suppose L is a lie algebra over C. Then L is semi-simple ⇐⇒ k is non-


degenerate. That is if x 6= 0 =⇒ k (x, x) 6= 0.

1
Sunaina Pati (July 10, 2024) Cartan’s criteria

Proof. Suppose L is semi-simple.So rad L is 0. So there is no solvable ideal of L. Note that


L ⊂ L is an ideal. Now take x, y, z ∈ L⊥ =⇒ [x, y ] ∈ (L⊥ )0 . Note that k ([x, y ], z ) = 0
as [x, y ] ∈ L. So L⊥ is solvable (by the criterion) and ideal. So L⊥ = 0. Hence k is
non-degenarate.
Other way: Say L is not simple =⇒ it has solvable ideals. Take 0 6= I ⊂ L a solvable
ideal =⇒ ∃N ∈ N such that I (N ) = 0 and I (N −1) 6= 0. Let A = I (N −1) . Now ∀y ∈ I
we have
ada adx ada (y ) = [a, [x, [a, y ]]] = 0 =⇒ (ada adx )2 = 0
=⇒ ada · adx is nilpotent =⇒ tr (ada · adx ) = 0 =⇒ k (a, x) = 0 =⇒ k (a, a) = 0.

§3 Why is it called semi-simple


Now, we will understand why it is called semi-simple!

Theorem 3.1. Let L is a lie algebra L overL C. Then L is semisimple =⇒ ∃ simple ideals
L1 , . . . , Ln ⊂ L such that L = L1 · · · Ln .

Proof. We do induction on dim L. Base case: dim L = 1 =⇒ L is simple. Say for all
lie algebras of dim < n, the statement holds and dim L = m. Let I ⊂ L be ideal of L
and minimal dimension. If I = L then L is simple. If I 6= L then we have the following
claim.

Claim 3.2.
I ⊥ , I, I ⊥ are semi-simple .
M
L=I

Proof. Note that

x 6= 0 ∈ I ∩ I ⊥ =⇒ k (x, x) = 0 =⇒ k is degenerate =⇒ x = 0.

Note that I, I ⊥ commute.

[x, w ] ∈ I, I ⊥ =⇒ [x, w ] = 0∀x, w ∈ I, I ⊥ .


Note that L = I + I ⊥ as I → I → C is isomorphism. So V → I → C is surjective and
kernel is I ⊥ . And dimensions follow! They are semi-simple, because suppose J ⊂ I is a
solvable ideal. Then

[J, I ⊥ ] ⊂ [I, I ⊥ ] = 0 =⇒ J ⊂ I ⊥ and solvable.


Not possible. So both are semi-simple.

Now use induction on I ⊥ .


Other direction: Suppose M M
L = L1 ··· Ln .
Let I = radL. Let Ik = [I, Lk ]. Note that Ik ⊂ Lk a solvable ideal. So Ik = 0. So
M M M
[I, L] = [I, L1 . . . Ln ] ⊂ I1 ··· In = 0
M M
=⇒ I ⊂ Z (L) ⊂ Z (L1 ) ··· Z (Ln ) = 0.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy