Summary Group Theory
Summary Group Theory
Summary Group Theory
Matthieu Schaller
Group Theory Summary
The picture on the title page is a 2-dimensionnal projection graph of E8 , the largest
complex exceptionnal Lie group. It has been discovered in 1889 by Wilhelm Killing. It has
dimension 248 and rank 8.
1
Group Theory Summary
2
CONTENTS Group Theory Summary
Contents
Foreword 5
1 Notation Conventions 6
4 Representation of groups 14
4.1 Homomrphisms and Isomorphims . . . . . . . . . . . . . . . . . . . . . . . . . 14
4.2 Representations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4.3 Reductible and Irreducible Representations . . . . . . . . . . . . . . . . . . . 16
7 Permutation Groups 22
7.1 Permutations and Cycles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
7.2 Young Diagrams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
7.3 Young Tableaux and Sn irreps . . . . . . . . . . . . . . . . . . . . . . . . . . 24
10 Orthogonal Groups 34
10.1 O(n) groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
10.2 SO(n) groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
10.3 The SO(2) group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
10.4 The SO(3) group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3
CONTENTS Group Theory Summary
11 Unitary Groups 37
11.1 U (n) groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
11.2 The U (1) group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
11.3 SU (n) groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
11.4 The SU (2) group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
11.5 The SU (3) group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
11.6 SU (n) groups and Young Tableaux . . . . . . . . . . . . . . . . . . . . . . . . 38
4
CONTENTS Group Theory Summary
Foreword
Important remarks
This document has not been checked by the lecturer and I cannot ensure that everything is
true.
For obvious reasons of ecology, please print this document only if you really need it.
Only a few explanations are given and the theorem’s proofs are left out. It contains also
some elements of Lie Algebra and representation theory.
If you find a mistake, if you have some suggestions to improve this document or if you
want to correct my poor English, take contact with me. The last version of this document
can always be found on my website1 .
If this document has helped you or if you enjoyed (!) it, feel free to tell me.
Acknowledgements
Thanks to Fabien Margairaz for his help setting up the LATEX style of this document.
Matthieu Schaller
[email protected]
5
1 Notation Conventions Group Theory Summary
1 Notation Conventions
Sets
• X ∩ Y = {z : z ∈ X and z ∈ Y }
• X × Y = {(x, y) : x ∈ X and y ∈ Y }
Symbols used
• ∀: for all.
• ∃: There exists and ∃! there exists a unique.
• ∈: belongs to.
• ⊂: is included in (is a subset of).
• ⇒: implies.
• ⇔: if and only if.
• ≡: is equivalent to.
Complex numbers
• z: Complex conjugate. a + bi = a − bi
√
• |z|: Absolute value. |a + bi| = a2 + b2
Matrices
• Aij : A matrix.
• [A]ij : The element in row i and column j of the matrix A.
6
2 Abstract Group Theory Group Theory Summary
The composition law is not always the usual multiplication but the notation is the same.
By analogy, the notation gi2 means gi gi . The associativity property implies that the paren-
theses are superfluous.
Example 2.1. The set {0, 1, 2, 3} with the composition law being the addition modulo 4
operation is a group, usually denoted as Z4 .
Example 2.2. The set of all permutations of 3 elements is a group. Its composition law
corresponds to perform successive permutations. It is usually denoted as S3 . The 6 elements
of this group are:
123 123 123
e≡ , a≡ , b≡
123 213 132
123 123 123
c≡ , d≡ , f≡
321 312 231
The letters used here to describe these elements will be the same in the rest of this document.
Example 2.3. Z4 is an Abelian group and S3 is not. (It is the smallest non-Abelian group.)
7
2.3 Finite and infinite groups Group Theory Summary
These two theorems imply that one can talk form the inverse and the identity without
ambiguity.
Example 2.4. In Z4 , the identity element is the element 0.
Theorem 2.3 (Cancellation). In a group G, the left and right cancellation laws
hold.
gi gj = gi gk ⇒ gj = gk ∀gi , gj , gk ∈ G and
gj gi = gk gi ⇒ gj = gk ∀gi , gj , gk ∈ G
The period of a group is generally different from the set of elements in the group.
Example 2.6. In Z4 , |0| = 1, |1| = 4, |2| = 2, |3| = 4 and the different periods are:
{0, 2}, {0, 1, 2, 3}, {0, 3, 2, 1}.
Definition 2.6 (Cyclic group). A cyclic group is a group based on the set of
n different objects {g, g 2 , g 3 , ..., g n }, where g n = e. The element g is called the
generator of the group.
Cyclic groups of order n are denoted as Zn and all cyclic groups are Abelian.
All groups whose order are prime numbers are cyclic.
Example 2.7. Z2 is the smallest non-trivial cyclic group.
8
2.4 Multiplication Tables Group Theory Summary
0 1 2 3
0 0 1 2 3
1 1 2 3 0
2 2 3 0 1
3 3 0 1 2
A group is Abelian if and only if its multiplication table is symmetric about the main
diagonal.
This theorem implies that each group element appears once and only once on each row
and column of the multiplication table and hence, gives an easy way to compute multipli-
cation tables.
Example 2.9. The multiplication table of S3 is :
e a b c d f
e e a b c d f
a a e d f b c
b b f e d c a
c c d f e a b
d d c a b f e
f f b c a e d
9
2.5 Subgroups and Cosets Group Theory Summary
The unit element {e} and G itself form subgroups of G. These two subgroups are called
improper subgroups. The other one (if they exist) are called proper subgroups. The
identification of proper subgroups is one of the importantest part of group theory.
Example 2.11. The group S3 has 4 proper subgroups:
Theorem 2.5. Two cosets of a subgroup either contain exactly the same elements
or have no common elements at all.
This theorem sets an upper limit on the number of different cosets of a given subgroup.
Example 2.14. The subgroup H = {e, a} of S3 is of order 2, which is a divisor of 6, the order
of S3 .
10
2.6 Classes Group Theory Summary
2.6 Classes
Definition 2.12 (Conjugacy classes). The set of all elements that are conjugated
to each other is called a conjugacy class.
All elements in a conjugacy class have the same order. But the opposite is not true.
In an Abelian group, all elements are in their own conjugacy class.
Example 2.16. As it is Abelian, the group Z4 has 4 conjugacy classes:
Definition 2.14 (Product group). A group G is a product group if there are two
proper subgroups Hi and Hj such that :
All subgroups involved in a product group are self-conjugate. This implies that a simple
group can not be expressed as a product of smaller groups.
The order of a product group is given by: |Ha × Hb | = |Ha ||Hb |
Example 2.18. Z6 ∼
= Z3 ×Z2 (The precise signification of ∼
= will be given in the next section.)
11
2.8 Factor Groups Group Theory Summary
E A B C
E E A B C
A A E C B
B B C E A
C C B A E
{e, d, f }, {a, b, c}
E A
E E A
A A E
12
3 List of usual finite groups Group Theory Summary
Cyclic Groups A cyclic group is a group based on the set of n different objects
{g, g 2 , g 3 , ..., g n }, where g n = e. The element g is called the generator of the group.
These groups can be interpreted as the number 0, 1, ..., n − 1 with the addition modulo n as
composition law.
These groups are simple and their order is n.
They are usually designated as Zn , Z/nZ or Cn .
Klein four-Group With only four elements, it is the smallest non-cyclic group. It is the
product Z2 × Z2 . It corresponds to the symmetry of a rectangle in 2D.
It is usually designated as Z2 × Z2 or V .
Symmetric Groups They correspond to the sets of all bijection from X to X with the
composition of function as composition law. If X is finite (X = {1, ..., n}), the group is
called Sn and has order n!. These groups are not Abelian if n > 2.
Alterning Groups They correspond to the sets of all even-valued permutations of a finite
set X = {1, ..., n}. It has order n!/2. These groups are simple if n = 3 or if n ≥ 5.
They are usually denoted as An .
Dihedral Groups They correspond to the sets of symmetries of a regular polygon, in-
cluding reflexions and rotations. They are denoted as Dn where n is the number of sides of
the polygon. The order of the group is 2n.
Quaternion Group It is the group made of the 8 elements Q = {1, −1, i, −i, j, −j, k, −k}
with the composition law being:
• (−1)(−1) = 1
• (−1)a = a(−1) = −a ∀a ∈ Q
• i2 = j 2 = k 2 = ijk = −1
13
4 Representation of groups Group Theory Summary
4 Representation of groups
Group elements correspond to symmetry operations on spatial coordinates. They can be
represented as linear transformations with respect to a coordinate system. The resulting
matrices form a group which is equivalent to the group of symmetry operations. The matrices
form a representation of the group and each matrix is an element of the group.
Definition 4.1. Let G and G0 be two finite groups {e, g1 , g2 , ...} and {e0 , g10 , g20 , ...}.
Let ϕ be a mapping between the elements of G and G0 that preserves their compo-
sition rules.
If the two groups have the same order, then the mapping is called an isomorphism
and the two groups are said to be isomorphic. This is denoted as G ∼ = G0 .
If the orders are different, then the mapping is called an homomorphism and the
two groups are said to be homomorphic.
Example 4.1. Consider the group G = −1, 1 with the usual multiplication as composition
law and the group Z2 . Then the mapping
ϕ(1) = 0 ϕ(−1) = 1
establishes a one-to-one correspondance between G and Z2 . Thus, these two groups are
isomorphics.
E ≡ {e, d, f } A ≡ {a, b, c}
4.2 Representations
{e, a, b, c, d, f } → 1
14
4.2 Representations Group Theory Summary
{e, d, f } → 1 {a, b, c} → −1
Definition 4.3 (Similarity). Given a representation {De , Dg1 , Dg2 , ...} of a group
{e, g1 , g2 , ...} and a non-singular matrix S of the same dimension than the repre-
sentation the set
{SDe S −1 , SDg1 S −1 , SDg2 S −1 , ...}
is a similarity transformation of the representation.
Definition 4.4 (Direct Sum). Let {Dg } and {Dg0 } be two representations of dimen-
sions d and d0 of the same group. Then the direct sum of these two representations
is the set of block-diagonal matrices
De 0 Dg1 0 Dg2 0
, , , ...
0 De0 0 Dg0 1 0 Dg0 2
Example 4.6. The direct sum of the representations of S3 given in 4.3 and 4.4. Di0 = Di ⊕ 1
√ √
1 0 0 1
√ − 3 0 √1 3 0
1 1
De0 = 0 1 0 , Da0 = − 3 −1 0 , Db0 = 3 1 0 ,
2 2
0 0 1 0 0 2 0 0 2
√ √
−1 0 0 −1
√ − 3 0 −1
√ 3 0
1 1
Dc0 = 0 1 0 , Dd0 = 3 −1 0 , Df0 = − 3 −1 0
2 2
0 0 1 0 0 2 0 0 2
It is also a faithful representation of S3 .
15
4.3 Reductible and Irreducible Representations Group Theory Summary
Irreducible representations are often called irreps. For a given group, there are generally
more than one irreducible representation.
Example 4.7. The representations in example 4.3, 4.4 and 4.5 are irreducible but the one
from example 4.6 is redducible.
This theorem implies that one can deal with unitary matrices instead of general matrices
without loss of generality.
16
5 The Great Orthogonality Theorem Group Theory Summary
Lemma 5.1 (Schur’s first Lemma). A non-zero matrix that commutes with all of
the matrices of an irrep is a constant multiple of the unit matrix.
Lemma 5.2 (Schur’s second Lemma). Given two irreps {D1 , D2 , ..., D|G| } and
{D10 , D20 , ..., D|G|
0
} of a group G of dimensions d and d0 . Then if there is a matrix
M such that
M Di = Di0 M ∀i
then:
If d = d0 , either M = 0 or the two representations differ by a similarity.
If d 6= d0 , M = 0.
Schuhr’s lemmas provide restrictions on the form of matrices which commute with all of
the matrices of an irrep.
Theorem 5.1 (Great Orthogonality Theorem). Let {D1 , D2 , ..., D|G| } and
{D10 , D20 , ..., D|G|
0
} be two different irreps of a group G that have dimensionalities d
0
and d . Then
|G|
X
(Da )∗ij (Da0 )i0 j 0 = 0 ∀i, j, i0 , j 0
α=1
(k)
This theorem can be summarized in the following way. Let Dα be the αth matrix of
the kth irrep of a group G, then
|G|
X 0 |G|
(Da(k) )∗ij (Da(k ) )i0 j 0 = δi,i0 δj,j 0 δk,k0
α=1
d
Example 5.1. Consider the three irreps of S3 given in the examples 4.3, 4.4 and 4.5. Let
n o
(k) (k) (k) (k)
Dij = [De(k) ]ij , [Da(k) ]ij , [Db ]ij , [Dc(k) ]ij , [Dd ]ij , [Df ]ij
17
5 The Great Orthogonality Theorem Group Theory Summary
Example 5.2. The two one-dimensionnal irrep and the two-dimensionnal irrep of S3 given in
examples 4.3, 4.4 and 4.5 set in the previous equation give :
X
d2k = 12 + 12 + 22 = 6
k
which equals the order of the group. Hence, there are no other distinct irreducible represen-
tations of this group.
18
6 Characters and Character Tables Group Theory Summary
Example 6.1. For the representation of S3 given in example 4.4, the traces are given by:
The character of the αth class of the kth irreducible representation of a group is denoted
(in this document) as χkα .
Example 6.2. For the representation of S3 given in example 4.4, the characters are given by:
χ1 = 2, χ2 = 0, χ3 = −1
Theorem 6.1 (Orthogonality Theorem for Characters). The characters of the ir-
reducible representations of a group obey the relation
X 0
nα χkα χkα ∗ = |G|δk,k0
α
19
6.2 The decomposition Theorem Group Theory Summary
Theorem 6.3 (Decomposition Theorem). The character χa for the αth class of
any irrep can be written uniquely in terms of the corresponding characters of the
irreps of the group as X
χα = αk χkα
k
where
1 X
αk = nα χk∗
α χα
|G| α
and reducible if X
nα |χα |2 > |G|
α
20
6.4 Character Tables Group Theory Summary
This gives a diophantine equation which for small groups has only a few solutions and
thus implies that it is easy to determine the dimensions of the irreps.
Definition 6.3. A character table is a table which colums are labelled with the
different conjugacy classes and the rows are labelled with the irreducible represen-
tations (usually designated as Γi ). The elements in position i, j is the character of
the ith representation of the jth class.
To fill the different lines, the orthogonality theorem for characters is usefull. The “scalar
product” of two rows of the table gives zero if the rows are different, and |G| if they are the
same. The same relations holds for columns.
k0
X
nα χk∗
α χα = |G|δk,k0
α
where the first line is the trivial representation, the second line is the parity representation
and the third line is the 2 × 2 faithful matrix representation.
21
7 Permutation Groups Group Theory Summary
7 Permutation Groups
Permutation Groups play a central role in the theory of finite groups.
Definition 7.1. A cycle is a special type of permutation in which terms are just
cylced, so the first object becomes the second, the nth becomes the n + 1th and the
las object becomes the first. The notation used is
(12 . . . n)
to signify that
x1 → x2 , x2 → x3 , ..., xn → x1
Example 7.2. In S3 the diffrent permutations can be expressed in term of cycles as follow :
e ≡ (1)(2)(3)
a ≡ (12)(3)
b ≡ (1)(23)
c ≡ (13)(2)
d ≡ (231)
f ≡ (321)
Example 7.3. The cyclePstructure for the permutation a in S3 has the cycle-strucure, k1 =
n
1, k2 = 1, k3 = 0. And j=1 jkj = 1 · 1 + 2 · 1 + 3 · 0 = 3.
22
7.2 Young Diagrams Group Theory Summary
Lemma 7.2. Permutations are in the same class of Sn if and only if they have the
same cycle structure.
{e} : k1 = 3, k2 = 0, k3 = 0
{a, b, c} : k1 = 1, k2 = 1, k3 = 0
{d, f } : k1 = 0, k2 = 0, k3 = 1
Lemma 7.3. The number of elements nα in the αth class of Sn is given by:
n!
nα =
Πj kj !j kj
This gives a natural division of permutations into two families, the odd-permutations
and even-permutations which are equivalent to an odd or even number of two-cycles.
The even permutations form a group called the alterning group An . These two cosets also
form a group, a representation of Z2 . This implies Sn /An ∼
= Z2 .
23
7.3 Young Tableaux and Sn irreps Group Theory Summary
These diagrams correspond to the classes (1234), (123)(4), (12)(34),(12)(3)(4) and (1)(2)(3)(4)
respectively. As one cannot build other Young Tableaux with 4 boxes, these are the 5 classes
of Sn .
From these diagrams, one can find the cycle structure. For each column, the depth of
the colum corresponds to the length of the cycle. And the number of column of a given
depth ki corresponds to the number of ki -cycles in the group.
Example 7.7. The cycle structure from the second diagram in the previous exemple is k3 =
1, k1 = 1.
For the third diagram, k2 = 2 and for the fourth: k2 = 1, k1 = 2.
The number of elements in each class can then be found using the previously presented
formula or by considering the possible distinct cycles directly.
Example 7.8. From the diagrams in the previous example, one can find:
4! 4! 4! 4! 4!
n1 = = 6, n2 = = 8, n3 = = 3, n4 = = 6, n5 = =1
1!41 (1!31 )(1!11 ) (2!22 ) (1!21 )(2!12 ) 4!14
The sum of the ni is 24 as expected (|Sn | = n!).
Definition 7.3 (Hook factor of a box). The Hook factor of a box hi is the
number of boxes that a hook crosses in a given box i of a Young Tableau. The hook
is made of two lines: one dropping from the box and one other going to the right,
both starting from the box.
Example 7.9. The hook from the top-left box of the third Young Tableau of S3 is:
Definition 7.4 (Hook factor of a Young Tableau). The Hook factor of a Young
Tableau HY D is the product of all box’s hook factor of this Young Tableau.
n
Y
HY D = hi
i
24
7.3 Young Tableaux and Sn irreps Group Theory Summary
n!
dµ =
HY D
where HY D is the Hook factor of the Young diagram associated with the irrep µ.
25
8 Continous Groups and Lie Groups Group Theory Summary
R(~a)R(~b) = R(~c)
The continuity of the composition law implies that R(~a) and R(~a + δ~a) are both in the
group for an arbitrarly small value of δ.
Each group element can be represented as a point by associating it with its ~a d-dimensionnal
vector. The space in which the elements lie is called the group manifold.
Example 8.1. Rn with the usual vector-addition law is an (Abelian) n-parameter continuous
group.
Example 8.2. C∗ (complex numbers without 0) with the usual multiplication forms an
(Abelian) continuous group.
The set of fi is called the structure of the group and determine the whole group
in the same way as the multiplication table for discrete groups.
The properties of the composition law give some restrictons on the fi . The associativity
requires that
f~(~a, f~(~b, ~c)) = f~(f~(~a, ~b), ~c)
The existence of an identity element ~a requires that
f (a~0 , ~a) = f (~a, a~0 ) = ~a
And finally the existence of an inverse a~0 for each ~a implies that
f (a~0 , ~a) = f (~a, a~0 ) = a~0
Definition 8.3 (Lie Group). A continuous group in which all functions of the
structure are analytical is called a Lie Group.
26
8.2 Compact and Non-Compact Lie Groups Group Theory Summary
Example 8.3. Rn with the usual vector-addition law is an (Abelian) n-parameter Lie group.
Example 8.4. C∗ (complex numbers without 0) with the usual multiplication forms an
(Abelian) Lie group.
Example 8.5. The set of invertible matrices GL(n, R) of size n × n with the usual matrix
multiplication is a Lie group called the general linear group. It has dimension n2 .
Example 8.6. The set of invertible matrices SL(n, R) of size n × n and determinant 1, with
the usual matrix multiplication is a Lie group called the special linear group. It has also
dimension n2 .
In the two last examples, the matrices are mapping n−dimensionnal vectors into other
n−dimensionnal vectors. The space of vectors being mapped is called the representation
space and the dimension of this space is called the dimension of the representation.
This can take many different values for the same group. It is, in general, different from the
dimension of the group. The relationship between these two dimensions is different for each
kind of group.
Example 8.7. For GL(n, R), the fundamental representation is the one given previously: The
set of matrices n × n.
Definition 8.5 (Compact Lie Group). A Lie Group is compact if there exists a
faithful representation in which all the elements are alwas finite.
|Dij | < ∞
The complete definition is more complex but this one is sufficient for this course.
Example 8.8. The set of phases {exp(iθ)| − π < θ ≤ π} forms a compact group called U (1).
The importance of compact Lie Groups arises from the following thorem.
Theorem 8.1. (i) All compact Lie Groups have a finite dimensional unitary rep-
resentation.
∃{D(g)} such that D∗ (g)D(g) = Id
(ii) Finite dimensional representations of compact Lie groups are reducible, they
can be made block-diagonal.
(iii) Any irreducible representation of a compact Lie group is finite dimensional.
27
8.3 Matrix Generators Group Theory Summary
It is easy to classify all possible compact Lie Groups. And there is only one Abelian
compact Lie Group (U (1)). The one that are not product groups are called simple groups.
Products of simple groups with other non-Abelian simple groups are called semi-simple
groups.
There are three main families (SO(n), SU (n), Sp(2n)) of compact simple Lie Groups and
5 exceptionnal cases (E6 ,E7 ,E8 ,F4 and G2 ).
∂D ~
Dai ≡ D(0, 0, ..., ai , ..., 0) = D(0, 0, ..., 0) + (0)ai + O(a2i )
∂ai
This can be summarized as:
∂D(~a)
= Dai D(~a)
∂ai
By using the same process, one can define higher order derivatives.
This implies that one can define the Taylor expansion of D(~a) around ~0.
Lemma 8.1 (Taylor Expansion). Any element in the Lie Group (which represen-
tation has a dimension n) can be expressed as a Taylor Serie:
n n
X 1X 2 2
D(~a) = Id + ai Dai + a D + ···
i=0
2 i=0 i ai
∞
n X
X 1
= (ai )j
i=0 j=0
j!
= exp (a1 Da1 + a1 Da1 + · · · + an Dan )
to be checked
Unfortunately, the last equality cannot trivially be broken in the product of n exponentials.
[A, B] = AB − BA
where the product and substraction are the usual one for matrices.
28
8.4 Exponential of infinitesimal Generators Group Theory Summary
This reduces to the usual case exp(A) exp(B) = exp(A + B) when [A, B] = 0.
As one want exp (a1 Da1 + a1 Da1 + · · · + an Dan ) = exp ϕ1 Da1 exp ϕ2 Da2 · · · exp ϕn Dan ,
one apply the previous formula and get:
The only important parameters are the different commutators, which play an important
role in Lie Algebras. The commutators entirely define the Lie Group because they entirely
define the infinitesimal generators, which define all the elements by exponentiation.
29
9 Introduction to Lie Algebras Group Theory Summary
Definition 9.1 (Lie Algebra). A Lie Algebra A ia a vector space over a field
F together with a binary operation [·, ·] (called the Lie bracket) which obeys the
following rules:
Example 9.2. The Euclidian space R3 together with the usual cross product as Lie bracket
is a Lie Algebra.
Example 9.3. The vector space of unitary matrices n × n together with the Lie bracket
defined as
[x, y] = xy − yx
(where the multiplication and the substraction are the usual ones) is a Lie Algebra. The Lie
bracket defined in this example is called the commutator.
As this document only deal with the matrix representation of Lie Groups, the focus will
be set on parts of Lie Algebra that are relevent for these kinds of vector spaces.
9.2 Generators
Definition 9.2 (Generators). The generators of a Lie Algebra A are the set of
matrices {Ta } (basis vectors) of the vector space included in A.
Example 9.4. The generators of R3 are (1, 0, 0), (0, 1, 0), (0, 0, 1).
30
9.3 Subalgebras Group Theory Summary
Lemma 9.1 (Orthonormal generators). If the vector space included in the Lie
Algebra A has a scalar product h·, ·i, then the Lie Algebra has an orthonormal set
of generators.
hTa , Tb i = δa,b ∀a, b, ..., n = dim(A)
The orthonormal generators can be obtained from any generators by the Grahm-Schmidt
procedure.
9.3 Subalgebras
C = {A1 , A2 : [A1 , A2 ] = 0, A1 , A2 ∈ A}
For matrices, the Cartan Subalgebra is the largest number of elements of A which can
be diagonalised simultaneously. The standard choice for generators of A is to have the
largest number of diagonalised elements. Thus these generators are the basis of the Cartan
Subalgebra.
Example 9.7. For A = R3 the Cartan Subalgebra, is C = {(1, 0, 0)}.
Definition 9.5 (Rank of an Algebra). The rank of a Lie Algebra A is the dimen-
sion of the Cartan subalgebra of A.
31
9.4 Structure constants Group Theory Summary
and they totally define the Lie Algebra. The choice of structure constants for a given Lie
Algebra is not unique.
The Jacobi identity can be rewritten in term of the structure constants as:
dim(A)
X
f abe f ecg + f bce f eag + f cae f ebg = 0
e=1,g=1
Remark. The importance of this representation will be obvious later in the discussion of Lie
Groups.
U = exp(A), A∗ = −A
Pdim(A)
The generators {Ta } are the basis vectors of the Lie Algebra A = i=a ca Ta , and the
derivatives with respect to the parameters of the Lie Groups ca are given by:
dU
Da =
dca
This imply that the knowledge of the structure constnats is enough to caracterise a Lie
Group. They play the same role than multiplication table for finite groups.
32
9.6 Lie Algebras of product Lie Groups Group Theory Summary
Using the formula 8.2, whith P = exp(B) and Q = exp(C), one can find that
This implies that the generators of the Lie Algebra AG obey the following rules:
abc
[Ya , Yb ] = fH Yc ∈ AH ⊂ AG
abc
[Za , Zb ] = fK Yc ∈ AK ⊂ AG
[Ya , Zb ] = 0
The rest of this document will present the different simple Lie Groups.
33
10 Orthogonal Groups Group Theory Summary
10 Orthogonal Groups
Orthogonal groups play a central role in geometry.
Definition 10.1 (O(n)). The group O(n) is the compact Lie group of all real
orthogonal matrices n × n with the composition law being the usual matrix multi-
plication.
O(n) = {D ∈ GL(n, R) : DT D = DDT = Id}
This group is called the Orthogonal Group in n dimensions.
This definition implies that all matrices in O(n) have a determinant equal to ±1.
Example 10.1. The group O(1) = {−1; 1} is isomorphic to Z2 and not a Lie Group.
The O(n) groups correspond to the set of all linear transformations which preserve the
length of the vectors in Rn . This implies that the group is Abelian.
The dimension of the Lie Group O(n) is 21 n(n − 1) because the matrices need to be sym-
metric (modulo a change of sign).
O(n − 1) ⊂ O(n)
The group O(n) can be split in two sets according to the determinant of the matrices.
The subset with determinant −1 is not a group, because it doesn’t contain the identity.
Definition 10.2 (SO(n)). The SO(n) group is the subgroup of O(n) in which the
matrices have a determinant +1.
is in SO(2).
34
10.3 The SO(2) group Group Theory Summary
Z+ = Id
−Id if n odd
Z− =
diag(−1, +1, +1, ..., +1) if n even
35
10.4 The SO(3) group Group Theory Summary
The computation of the structure constants leads to f abc = εabc , where εabc is the Levi-
Civita tensor.
36
11 Unitary Groups Group Theory Summary
11 Unitary Groups
11.1 U (n) groups
Definition 11.1 (U (n)). The group U (n) is the compact Lie group of all complex
unitary matrices n × n with the composition law being the usual matrix multiplica-
tion.
U (n) = {D ∈ GL(n, C) : DT D = DDT = Id}
This group is called the Unitary Group in n dimensions.
This definition implies the matrices in U (n) obey the equation |detD| = 1.
The dimension of this group in n2 .
Definition 11.2. The U (1) group (or circle group) is the multiplicative group of
all complex numbers with absolute value 1.
U (1) = {z ∈ C : |z| = 1}
U (1) can be represented as a circle of radius 1 around the origin in the complex plane.
This circle is the group manifold. This is the same manifold than SO(2), hence:
U (1) ∼
= SO(2)
Definition 11.3 (SO(n)). The SU (n) group is the subgroup of U (n) in which the
matrices have a determinant +1.
ϕs : χ → χ : H → H 0 = ϕS H = SHS ∗ (11.2)
37
11.4 The SU (2) group Group Theory Summary
An obvious set of generators for this group is then the set of Pauli matrices.
0 1 0 −i 1 0
, σ2 = , σ3 =
1 0 i 0 0 −1
and the structure constants are f abc = εabc , the Levi-Civita tensor. These are the same
structure constants than SO(3) and hence, their Lie Algebras are identical. One can show
that
SU (2)/Z2 ∼ = SO(3)
38
12 Other Lie Groups Group Theory Summary
Definition 12.1 (Group Sp(2n)). The group Sp(2n) is the set of real matrices D
that satisfy M JM T = J, where J is the 2n × 2n matrix given in terms of n × n
blocks:
0 Id
J=
−Id 0
It is called the symplectic group of n-dimension.
This group is involved in classical Hamiltonian mechanics and its dimension is n(2n + 1) to be com-
pleted
12.2 Exceptional groups
These 5 groups are called exceptionnal because they don’t belong to an infinite serie (SO(n),
SU (n) or SP (2n)). to be com-
pleted
39
13 Group theory in Physics Group Theory Summary
40