Section III.3.5

Irreducible representations of finite groups

[10/20%]
Show
LaTeX
Problem III.3.5.1

Obtain relation (4) by writing out explicitly the expression tij=(Γi,χj)Gt_{i j}=\left(\Gamma_{i}, \chi_{j}\right)_{G} for the coefficients of the expansion Γi=∑jtijχj\Gamma_{i}=\sum_{j} t_{i j} \chi_{j} of the basis central function Γi\Gamma_{i} (see (7) from § 4) in terms of the irreducible characters.

?
Problem III.3.5.2

Verify (and recall the isomorphism between the vector space VV and its dual space V∗V^{*} of linear functions) that the map τ:A⟶A^^\tau : A \longrightarrow \hat{\hat{A}}, defined by the condition

aτ(χ)=χ(a) a^{\tau }(\chi )=\chi (a)

gives an isomorphism of the abelian group AA onto A^^\hat{\hat{A}}.

This exercise, together with Theorem 4, establishes part of the so-called duality law for finite abelian groups. An analogous, but much deeper, duality law for topological abelian groups, leading to important consequences, was established in the 1930s by L.S. Pontryagin.

?
Problem III.3.5.3

Prove that if a finite abelian group AA admits a faithful complex irreducible representation, then AA is a cyclic group.

?
Problem III.3.5.4

Let AA be a finite abelian group, BB a subgroup of it. Prove that any character of the group BB extends to a character of the group AA, and the number of such extensions equals the index (A:B)(A: B).

?
Problem III.3.5.5

Justify the phrase before the closing parentheses in part V) of item 4.

Recall what this refers to. Let Φ\Phi be the natural representation of the group S4S_{4} on the space V=⟨e1,e2,e3,e4⟩V=\langle e_{1}, e_{2}, e_{3}, e_{4}\rangle with the action Φ(g)ei=eg(i)\Phi (g) e_{i}=e_{g(i)}; then Φ=Φ(1)∔Ψ\Phi =\Phi^{(1)} \dotplus \Psi, where Φ(1)\Phi^{(1)} is the unit representation and Ψ\Psi is an irreducible three-dimensional representation acting on the space W=⟨e1−e4,e2−e4,e3−e4⟩W=\langle e_{1}-e_{4}, e_{2}-e_{4}, e_{3}-e_{4}\rangle, with character χΨ(g)=N(g)−1\chi_{\Psi }(g)=N(g)-1, where N(g)N(g) is the number of points of {1,2,3,4}\left\{ 1,2,3,4\right\} fixed by gg. In the character table of the group S4S_{4} the character χ5=χΨ\chi_{5}=\chi_{\Psi } takes the values 3,−1,1,0,−13, -1, 1, 0, -1 on the classes ee, (1 2)(3 4)(1\, 2)(3\, 4), (1 2)(1\, 2), (1 2 3)(1\, 2\, 3), (1 2 3 4)(1\, 2\, 3\, 4) respectively. Part V) (irreducible representations of the symmetric group S4S_{4}) ends as follows: "The representation Φ(5)=Ψ\Phi^{(5)}=\Psi with character χ5\chi_{5} is connected with the 2-transitivity of the group S4S_{4}. It is also equivalent to the representation corresponding to all symmetry transformations of the tetrahedron Δ4\Delta_{4} (rotations plus reflections; it is exactly these transformations that are important in describing the vibrations of the phosphorus molecule (Problem 2 of § 2, Chapter 1 in Part I of this book))".

?
Problem III.3.5.6

What is the average value 1∣G∣∑gχ(g)\frac{1}{\left|G\right|} \sum_{g} \chi (g) of the values of a complex character χ\chi on the elements of a finite group GG?

?
Problem III.3.5.7

Collect, from various places, the tables relating to the group A5A_{5} into a consolidated character table

60115201212
A5A_{5}ee(12)(34)(12)(34)(123)(123)(12345)(12345)(12354)(12354)
χ1\chi _{1}11111
χ2\chi _{2}3-10(1+5)/2(1+\sqrt{5}) / 2(1+5)/2(1+\sqrt{5}) / 2
χ3\chi _{3}3-10(1−5)/2(1-\sqrt{5}) / 2(1−5)/2(1-\sqrt{5}) / 2
χ4\chi _{4}401-1-1
χ5\chi _{5}∗*∗*∗*∗*∗*

Give a description of the irreducible representations with characters χ1,χ2,χ3,χ4\chi_{1}, \chi_{2}, \chi_{3}, \chi_{4}. Fill in the last row of the table, using the second orthogonality relation (4) for characters.

?
Problem III.3.5.8

Let P=⟨AiBjCk;0⩽i,j,k⩽p−1⟩P= \left\langle A^{i} B^{j} C^{k} ; 0 \leqslant i, j, k \leqslant p-1\right\rangle be the group of order p3p^{3} considered in Exercise 3 from §3\S 3 Ch. 1;V=⟨e0,e1,…,ep−1⟩C1 ; V= \left\langle e_{0}, e_{1}, \ldots , e_{p-1}\right\rangle_{\mathbb {C}} a complex vector space of dimension p;εp ; \varepsilon a primitive pp-th root of unity ;A,B,C; \mathcal{A}, \mathcal{B}, \mathcal{C} linear operators on VV defined by the relations

Aei=ei+1,Bkei=ε−kiei,Ckei=εkei,0⩽i⩽p−1 \mathcal{A} e_{i}=e_{i+1}, \quad \mathcal{B}^{k} e_{i}=\varepsilon ^{-k i} e_{i}, \quad \mathcal{C}^{k} e_{i}=\varepsilon ^{k} e_{i}, \quad 0 \leqslant i \leqslant p-1

(the subscripts of the basis elements are taken modulo pp).

Show that the map

Φ(k):A↦A,B↦Bk,C↦Ck \Phi ^{(k)}: A \mapsto \mathcal{A}, B \mapsto \mathcal{B}^{k}, C \mapsto \mathcal{C}^{k}

gives a linear irreducible representation of the group PP. The representations Φ(1),…,Φ(p−1)\Phi^{(1)}, \ldots , \Phi^{(p-1)} are pairwise inequivalent and, together with the p2p^{2} one-dimensional representations (p2p^{2} is the index of the commutator subgroup P′=⟨C⟩P^{\prime }=\langle C\rangle in PP), exhaust all the irreducible complex representations of the group PP.

?
Problem III.3.5.9

Fill in with calculations the following argument. Let

Dn=⟨a,b∣an=e,b2=e,bab−1=an−1⟩ D_{n}= \left\langle a, b \mid a^{n}=e, b^{2}=e, b a b^{-1}=a^{n-1}\right\rangle

be the dihedral group of order 2n2 n, whose properties (including the description of the conjugacy classes) are given in Example 1 from item 5§45 \S 4 Ch. 1. Since ⟨a⟩◃Dn\langle a\rangle \triangleleft D_{n}, the maps a↦1,b↦1a \mapsto 1, b \mapsto 1 and a↦1,b↦−1a \mapsto 1, b \mapsto -1 define two one-dimensional representations. Let ε\varepsilon be a primitive nn-th root of unity. Then the map

Φ(j):a↦∥εj00ε−j∥ \Phi ^{(j)}: a \mapsto \left\Vert \begin{smallmatrix} \varepsilon ^{j} & 0 \\ 0 & \varepsilon ^{-j} \end{smallmatrix}\right\Vert

will define a representation of degree 2. The representation Φ(j)\Phi^{(j)} is irreducible for j=1,2,…,[(n−1)/2]j=1,2, \ldots ,[(n-1) / 2] ([α][\alpha ] is the integer part of the real number α)\alpha ). For n=2mn=2 m, the representation Φ(m)\Phi^{(m)} splits into a direct sum of two one-dimensional representations: a↦−1,b↦1a \mapsto -1, b \mapsto 1 and a↦−1,b↦−1a \mapsto -1, b \mapsto -1. This agrees with the fact that the commutator subgroup D2m′D_{2 m}^{\prime } has index 4 in D2mD_{2 m} and D2m/D2m′≅Z2×Z2D_{2 m} / D_{2 m}^{\prime } \cong Z_{2} \times Z_{2}. All the indicated representations are irreducible and constitute the complete set of irreducible complex representations of the dihedral group. Find a real realization of the representations Φ(j)\Phi^{(j)}. Give explicitly the isomorphism (equivalence) Φ(j)≈Φ(k),k>m,j⩽m\Phi^{(j)} \approx \Phi^{(k)}, k>m, j \leqslant m.

?
Problem III.3.5.10

Crystallographic groups (relating to Problem 2 from § 2 Ch. 1 in [ВА I]). Let EE be an nn-dimensional Euclidean space and VV the vector space associated with it, with Euclidean inner product. To every motion dd of the space EE there corresponds an orthogonal linear transformation dˉ⊂O(n)\bar{d} \subset O(n), in such a way that d1d2‾=dˉ1dˉ2\overline{d_{1} d_{2}}=\bar{d}_{1} \bar{d}_{2}. A group DD of motions of the space is called a crystallographic group if the DD-orbit of an arbitrary point is discrete (has no limit points) and there exists a compact set M⊂EM \subset E for which D(M)=⋃d∈Dd(M)=ED(M)=\bigcup_{d \in D} d(M)=E. The Schoenflies--Bieberbach theorem holds, according to which a crystallographic group DD contains nn independent affine translations generating a normal subgroup LL in DD, and Dˉ≅D/L\bar{D} \cong D / L is a finite group (the point crystallographic group). In all, there are 32 geometrically distinct point crystallographic groups for n=3n=3. Among them there must obviously be groups containing reflections (improper motions). From the crystallographic conditions it follows that every proper rotation from Dˉ\bar{D} is represented by a matrix similar to

A=∥cos⁡θ−sin⁡θ0sin⁡θcos⁡θ0001∥ A=\left\Vert \begin{smallmatrix} \cos \theta & -\sin \theta & 0 \\ \sin \theta & \cos \theta & 0 \\ 0 & 0 & 1 \end{smallmatrix}\right\Vert

tr⁡A=1+2cos⁡θ∈Z\operatorname {tr} A=1+2 \cos \theta \in \mathbb {Z}. Relying on Theorem 2 from § 3 and on the noted consideration, show that for n=3n=3 the point crystallographic groups without reflections are precisely the cyclic groups C1,C2,C3,C4,C6C_{1}, C_{2}, C_{3}, C_{4}, C_{6}, the dihedral groups D2,D3,D4,D6D_{2}, D_{3}, D_{4}, D_{6}, the tetrahedron group T\mathbf{T}, and the group of the cube (octahedron) O\mathbf{O}.

?