Scribd
Upload
6 views2 pages

2008-10-17

The document discusses Lie groups and Lie algebras, defining a Lie group as a manifold that satisfies certain smoothness conditions. It explains the relationship between Lie groups and their corresponding Lie algebras, focusing on the properties of the Lie bracket and the Baker-Hausdorff theorem. Additionally, it covers various examples of Lie groups and their representations, as well as the concepts of ideals and semisimple Lie algebras.

Uploaded by

irfanu92
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
6 views2 pages

2008-10-17

The document discusses Lie groups and Lie algebras, defining a Lie group as a manifold that satisfies certain smoothness conditions. It explains the relationship between Lie groups and their corresponding Lie algebras, focusing on the properties of the Lie bracket and the Baker-Hausdorff theorem. Additionally, it covers various examples of Lie groups and their representations, as well as the concepts of ideals and semisimple Lie algebras.

Uploaded by

irfanu92
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

2008–10–17

A Lie group is a gruop that forms a manifold of finite dimension n, fulfilling the requirements

a) if g g ′ = g ′′, then the coordinates of g ′′ are smooth functions of the coordinates of g and g ′.

Lie algebra = powerful tool for studying Lie gropus. Amounts to studying Lie group close to
the identity element 1.

Definition: Lie algebra G of a Lie group G = tangent space of G at 1, with an extra structure,
see below. This means, a group element g close to identity 1 can be written g = 1 + ε A + O(ε2)
where ε is a small number. Also, if A ∈ G then

 N ∞
1 X 1 ν
exp A = eA = lim 1+ A = A ∈G
N →∞ N ν!
ν =0

(In all our examples g and A are n × n matrices, so pruducts of A’s are well defined. Mathem-
aticians define Lie algebras abstractly, have to do some more work here.)

Question: which Lie group elements can be obtained by exponentiation from G?

If A, B ∈ G ⇒ G ∋ eA eB = eC for some C ∈ G? Look first at group elements close to 1.

eεA eεB = e(ε)?

To order ε2?
  
1 1 2 2 1
1 + ε A + ε2 A2 1 + ε B + ε B = 1 + C + C2
2 2 2

1 2 2
ε A + 2 A B + B2 + 

1 + ε(A + B) +
2
(A+B)2 +AB −BA

1 2 1
ε [A, B] +  + ε(A + B) + O(ε2) + 
2
1 + ε(A + B) +
2 2

1
requires C(ε) = ε(A + B) + 2 ε2 [A, B] + O(ε3). Conclusion C ∈ G requires [A, B] ∈ G.

Mathematical definition of a Lie algebra G

G is a vector space over some number field F. In physics usually F = C or F = R (complex or


real numbers) with an additional structure: the Lie bracket.

A, B ∈ G ⇒ [A, B] ∈ G.

In our matrix examples [A, B] = A B − B A (commutator) but mathematicians define [A, B]


more abstractly. Need not be the commutator, but satisfies axioms. Is linear in A and in B and
is antisymmetric [A, B] = − [B , A]; satisfies the Jacobi identity.

Remark: The Poisson bracket satisfies these axioms. ⇒ There is a Lie algebra in Hamiltonian
dynamics.

So, if A, B ∈ G then, by definition of G

eεA eεB = eC(ε)

1
for some C(ε) = g at least to order ε2. Then according to a theorem, Baker-Hausdorff theorem,
C(ε) ∈ G and can be written as an infinite sum of multicommutators of A and B. Explicit
expression, first obtained by Dynkin. Our example groups:

F dim G G G
1
R n (n − 1) O(n) A is n × n real antisymmetric, A + AT = 0
2
R n2 U (n) A is n × n complex antihermitian, A + A † = 0
2
R n −1 SU(n) in addition, tr(A) = 0
R n2 GL(n) A is n × n real arbitrary
2
R n −1 SL(n) A in addition traceless

In ll these examples elements of G G are n × n matrices, at as linear transformations in some n-


dimensional vector space. One says that one has an n dimensional representation of G or G and
the space of vectors ϕ is representation module = V . A group can have many different repres-
entations.
Adjoint representation. Choose basis for G {Aa , a = 1,  , dim G }. Then [Aa , Ab] = c f a bc Ac,
P
that is, commutation realtions are summarised by n3 constants fa b c. Interpretation: G is repres-
ented on itself, V = G. Aa ∈ G transforms Ab ∈ V to fa b cAc ∈ V . Aa is represented by a matrix
(Ma)cb = adjoint representation matrix.
Note: [Aa ′ [Aa ′, Ab]] = [Aa ′, fa b cAc] = fa b cfa ′ cdAd = Ad(Ma ′)dc(Ma)cb.
In order to check that this is really a representation of G one must check that the commutator
of Aa and Aa ′ is represented by the commutator of the matrices M ; i.e.,

[[Aa ′, Aa], Ab] = [Aa ′, [Aa , Ab]] − [Aa[Aa ′, Ab]]

but this is an identity.


Cartan-Killing form of a Lie algebra. Def by

1 1X
ga b = − tr(Ma Mb) = − fac d fbd c
2 2
c,d

For albelian Lie albegra fa b c = 0 and ga b = 0. On the other extreme among Lie algebras there
are semisimple Lie algebras for which gab is nonsingular, and works as metric on G. Is used to
a b
raise and lower indices, g ab = g −1 . The analog in Lie algebra theory of the concept of
invariant subgroups in group theory is called ideal .
Definition: H ⊂ G is called an ideal of G if

1. H is itself a Lie algebra.

2. A ∈ G , B ∈ H then [A, B] ∈ H.

A Lie algebra G which has no ideals except {0} and G it is called simple.
Another property of semisimple Lie algebra. A Lie algebra which has no albelian ideal is called
semisimple.
Same method as used on SU(2) in quantum mechanics can be generalized to arbitrary simple
Lie algebras, all such, and all unitary irreducible representations, of them are classified:
1) Simple Lie algebras: An , Bn , Cn , Dn , n > 1 with some muntiple counting.
E6, E7, E8, F4, G2.
2) All their irreducible representations, see Pope!

You might also like