Mathematics / Pure Mathematics Abstract Algebra: Groups, Rings & Fields 100% Free Open Access
Chapter 5 • Theory & Derivations

Unit 5: Group Homomorphisms, Isomorphism Theorems & Automorphisms

Exhaustive exploration of structure-preserving maps in abstract algebra: group homomorphisms, kernels, images, the First, Second (Diamond), and Third Isomorphism Theorems with complete formal proofs, Cayley's representation theorem, the automorphism group Aut(G), inner automorphisms Inn(G), and characteristic subgroups.

§5.1 Group Homomorphisms, Kernels & Fundamental Properties

1. Definition of a Group Homomorphism

Let $(G, \cdot)$ and $(H, *)$ be two groups.

Definition 5.1 (Homomorphism): A function $\phi: G \to H$ is called a group homomorphism if it preserves the group operation:

$$\phi(a \cdot b) = \phi(a) * \phi(b), \quad \forall a, b \in G$$
  • If $\phi$ is injective (one-to-one), it is called a monomorphism (or embedding).
  • If $\phi$ is surjective (onto), it is called an epimorphism.
  • If $\phi$ is bijective (both injective and surjective), it is called an isomorphism, denoted $G \cong H$.
  • An isomorphism from $G$ to itself ($\phi: G \to G$) is called an automorphism.
  • A homomorphism from $G$ to itself is called an endomorphism.

2. Fundamental Properties of Homomorphisms

Proposition 5.1 (Elementary Properties): Let $\phi: G \to H$ be a homomorphism. Then:

  1. $\phi(e_G) = e_H$.
  2. $\phi(g^{-1}) = (\phi(g))^{-1}$ for all $g \in G$.
  3. $\phi(g^n) = (\phi(g))^n$ for all $g \in G$ and $n \in \mathbb{Z}$.
  4. If $|g|$ is finite, then $|\phi(g)|$ divides $|g|$.
  5. If $K \le G$, then $\phi(K) \le H$.
  6. If $L \le H$, then the preimage $\phi^{-1}(L) \le G$.
Proof of 1, 2, and 4:
  • Proof of 1: $\phi(e_G) = \phi(e_G \cdot e_G) = \phi(e_G) * \phi(e_G)$.

Multiplying both sides by $(\phi(e_G))^{-1}$ in $H$:

$$e_H = \phi(e_G) \quad \blacksquare$$
  • Proof of 2: $\phi(g) * \phi(g^{-1}) = \phi(g \cdot g^{-1}) = \phi(e_G) = e_H$.

By uniqueness of inverses in $H$, $\phi(g^{-1}) = (\phi(g))^{-1}$. $\blacksquare$

  • Proof of 4: Let $n = |g|$. Then $g^n = e_G$.

Applying $\phi$: $(\phi(g))^n = \phi(g^n) = \phi(e_G) = e_H$. By Proposition 2.3, this implies $|\phi(g)|$ divides $n = |g|$. $\blacksquare$


3. Kernel and Image

Definition 5.2 (Kernel and Image): Let $\phi: G \to H$ be a group homomorphism.

  1. The kernel of $\phi$ is:
$$\ker(\phi) = \{ g \in G : \phi(g) = e_H \} = \phi^{-1}(\{e_H\})$$
  1. The image of $\phi$ is:
$$\text{im}(\phi) = \phi(G) = \{ \phi(g) : g \in G \} \subseteq H$$

Theorem 5.1 (Kernel is a Normal Subgroup): Let $\phi: G \to H$ be a group homomorphism. Then:

  1. $\ker(\phi) \trianglelefteq G$.
  2. $\text{im}(\phi) \le H$.
  3. $\phi$ is injective if and only if $\ker(\phi) = \{e_G\}$.
Complete Proof:

1. Normality of $\ker(\phi)$:

  • Non-emptiness: $\phi(e_G) = e_H \implies e_G \in \ker(\phi)$.
  • Subgroup test: Let $x, y \in \ker(\phi)$. Then:
$$\phi(x y^{-1}) = \phi(x) * \phi(y^{-1}) = \phi(x) * (\phi(y))^{-1} = e_H * e_H^{-1} = e_H$$

Thus $x y^{-1} \in \ker(\phi)$, so $\ker(\phi) \le G$.

  • Normality: For any $g \in G$ and $k \in \ker(\phi)$:
$$\phi(g k g^{-1}) = \phi(g) * \phi(k) * \phi(g^{-1}) = \phi(g) * e_H * (\phi(g))^{-1} = \phi(g) * (\phi(g))^{-1} = e_H$$

Hence $g k g^{-1} \in \ker(\phi)$ for all $g \in G$. Therefore, $\ker(\phi) \trianglelefteq G$. $\blacksquare$

2. Subgroup property of $\text{im}(\phi)$:

Let $u, v \in \text{im}(\phi)$. Then $u = \phi(a)$ and $v = \phi(b)$ for some $a, b \in G$.

$$u v^{-1} = \phi(a) * (\phi(b))^{-1} = \phi(a) * \phi(b^{-1}) = \phi(a b^{-1}) \in \text{im}(\phi)$$

Thus $\text{im}(\phi) \le H$. $\blacksquare$

3. Injectivity Criterion:

  • $(\implies)$ If $\phi$ is injective and $\phi(x) = e_H = \phi(e_G)$, then injectivity implies $x = e_G$. Thus $\ker(\phi) = \{e_G\}$.
  • $(\impliedby)$ Suppose $\ker(\phi) = \{e_G\}$. If $\phi(a) = \phi(b)$, then:
$$\phi(a b^{-1}) = \phi(a) * (\phi(b))^{-1} = \phi(a) * (\phi(a))^{-1} = e_H$$

Thus $a b^{-1} \in \ker(\phi) = \{e_G\}$, meaning $a b^{-1} = e_G \implies a = b$. Therefore $\phi$ is injective. $\blacksquare$

§5.2 The First Isomorphism Theorem for Groups

1. Statement of the First Isomorphism Theorem

The First Isomorphism Theorem (often called the Fundamental Theorem of Homomorphisms) reveals that every homomorphism $\phi: G \to H$ can be factored into a canonical projection onto a quotient group followed by an isomorphism onto its image.

Theorem 5.2 (First Isomorphism Theorem): Let $\phi: G \to H$ be a group homomorphism with kernel $K = \ker(\phi)$. Then the quotient group $G/K$ is isomorphic to the image $\text{im}(\phi)$:

$$G/\ker(\phi) \cong \text{im}(\phi)$$

Specifically, the mapping:

$$\Phi: G/K \to \text{im}(\phi), \quad \Phi(gK) = \phi(g)$$

is a well-defined group isomorphism.

``` G ------------ \phi ------------> H | ^ | \pi | v | \Phi G/K ------------ \cong -----------> im(\phi) ```


2. Complete Rigorous Proof

We must verify four assertions:

  1. $\Phi$ is well-defined (independent of coset representative).
  2. $\Phi$ is a homomorphism.
  3. $\Phi$ is injective.
  4. $\Phi$ is surjective.
Step 1: Well-Definedness

Suppose $g_1 K = g_2 K$ for $g_1, g_2 \in G$. Then $g_1^{-1} g_2 \in K = \ker(\phi)$. This means $\phi(g_1^{-1} g_2) = e_H$. Since $\phi$ is a homomorphism:

$$\phi(g_1)^{-1} \phi(g_2) = e_H \implies \phi(g_1) = \phi(g_2)$$

Therefore, $\Phi(g_1 K) = \phi(g_1) = \phi(g_2) = \Phi(g_2 K)$. The value of $\Phi(gK)$ is completely independent of the representative $g$.

Step 2: Homomorphism Property

For any cosets $aK, bK \in G/K$:

$$\Phi((aK)(bK)) = \Phi((ab)K) = \phi(ab)$$

Since $\phi$ is a homomorphism:

$$\phi(ab) = \phi(a) \phi(b) = \Phi(aK) \Phi(bK)$$

Thus $\Phi$ preserves the group operation.

Step 3: Injectivity

We examine the kernel of $\Phi$:

$$\ker(\Phi) = \{ gK \in G/K : \Phi(gK) = e_H \}$$

Now $\Phi(gK) = e_H \iff \phi(g) = e_H \iff g \in \ker(\phi) = K$. By coset equality (Lemma 4.1), $g \in K \iff gK = K = e_{G/K}$. Therefore:

$$\ker(\Phi) = \{ K \} = \{ e_{G/K} \}$$

By Theorem 5.1, $\ker(\Phi) = \{ e_{G/K} \}$ implies $\Phi$ is injective.

Step 4: Surjectivity

Let $y \in \text{im}(\phi)$. By definition of the image, there exists some $g \in G$ such that $\phi(g) = y$. Then for the coset $gK \in G/K$, we have:

$$\Phi(gK) = \phi(g) = y$$

Thus every element of $\text{im}(\phi)$ is hit by $\Phi$. $\Phi$ is surjective.

Conclusion: $\Phi$ is a bijective homomorphism, hence an isomorphism:

$$G/\ker(\phi) \cong \text{im}(\phi) \quad \blacksquare$$

§5.3 The Second (Diamond) and Third Isomorphism Theorems

1. The Second Isomorphism Theorem (Diamond Isomorphism Theorem)

The Second Isomorphism Theorem describes the relationship between the intersection and product of a subgroup and a normal subgroup.

Theorem 5.3 (Second Isomorphism Theorem): Let $G$ be a group, $H \le G$ a subgroup, and $N \trianglelefteq G$ a normal subgroup. Then:

  1. $HN = \{ hn : h \in H, n \in N \} \le G$.
  2. $N \trianglelefteq HN$.
  3. $H \cap N \trianglelefteq H$.
  4. The quotient groups are isomorphic:
$$\frac{HN}{N} \cong \frac{H}{H \cap N}$$

``` HN / \ H N \ / H \cap N ```

Complete Proof:

1. $HN \le G$: Let $h_1 n_1, h_2 n_2 \in HN$. Then:

$$(h_1 n_1)(h_2 n_2)^{-1} = h_1 n_1 n_2^{-1} h_2^{-1} = h_1 (n_1 n_2^{-1}) h_2^{-1}$$

Since $N \trianglelefteq G$, $h_2^{-1} N h_2 = N$, so there exists $n_3 \in N$ such that $(n_1 n_2^{-1}) h_2^{-1} = h_2^{-1} n_3$. Then:

$$(h_1 n_1)(h_2 n_2)^{-1} = h_1 h_2^{-1} n_3 = (h_1 h_2^{-1}) n_3 \in HN$$

By the one-step subgroup test, $HN \le G$.

2. $N \trianglelefteq HN$: For any $hn \in HN$ and $m \in N$:

$$(hn) m (hn)^{-1} = h n m n^{-1} h^{-1} = h (n m n^{-1}) h^{-1}$$

Since $N \trianglelefteq G$, $n m n^{-1} \in N$, and $h (n m n^{-1}) h^{-1} \in N$. Thus $N \trianglelefteq HN$.

3. Application of First Isomorphism Theorem:

Define the map:

$$\theta: H \to \frac{HN}{N}, \quad \theta(h) = hN$$
  • Homomorphism: $\theta(h_1 h_2) = (h_1 h_2) N = (h_1 N)(h_2 N) = \theta(h_1) \theta(h_2)$.
  • Surjectivity: Any element of $HN/N$ is a coset of the form $(hn)N = h(nN) = hN$ (since $n \in N \implies nN = N$).

Thus $\theta(h) = hN = (hn)N$, so $\theta$ is surjective!

  • Kernel:
$$\ker(\theta) = \{ h \in H : \theta(h) = N \} = \{ h \in H : hN = N \} = \{ h \in H : h \in N \} = H \cap N$$

By Theorem 5.1, the kernel of any homomorphism is normal in the domain, so $H \cap N \trianglelefteq H$. Applying the First Isomorphism Theorem (Theorem 5.2) to $\theta$:

$$\frac{H}{\ker(\theta)} \cong \text{im}(\theta) \implies \frac{H}{H \cap N} \cong \frac{HN}{N} \quad \blacksquare$$

2. The Third Isomorphism Theorem

Theorem 5.4 (Third Isomorphism Theorem): Let $G$ be a group, and let $N$ and $K$ be normal subgroups of $G$ such that $N \le K \trianglelefteq G$. Then:

  1. $K/N \trianglelefteq G/N$.
  2. The quotient of quotients satisfies:
$$\frac{G/N}{K/N} \cong \frac{G}{K}$$
Complete Proof:

Define the map $\psi: G/N \to G/K$ by:

$$\psi(gN) = gK, \quad \forall g \in G$$
  • Well-definedness: If $g_1 N = g_2 N$, then $g_1^{-1} g_2 \in N$. Since $N \subseteq K$, $g_1^{-1} g_2 \in K$, which implies $g_1 K = g_2 K$.
  • Homomorphism: $\psi((aN)(bN)) = \psi((ab)N) = (ab)K = (aK)(bK) = \psi(aN) \psi(bN)$.
  • Surjectivity: For any coset $gK \in G/K$, $\psi(gN) = gK$.
  • Kernel:
$$\ker(\psi) = \{ gN \in G/N : \psi(gN) = K \} = \{ gN \in G/N : gK = K \} = \{ gN \in G/N : g \in K \} = K/N$$

Applying the First Isomorphism Theorem to $\psi$:

$$\frac{G/N}{\ker(\psi)} \cong \text{im}(\psi) \implies \frac{G/N}{K/N} \cong \frac{G}{K} \quad \blacksquare$$

§5.4 Cayley's Representation Theorem

1. The Concrete Universality of Symmetric Groups

Can every abstract group—no matter how exotic—be viewed as a group of permutations of some set? Arthur Cayley answered this affirmatively in 1854.

Theorem 5.5 (Cayley's Theorem): Every group $G$ is isomorphic to a subgroup of the symmetric group $\text{Sym}(G)$. In particular, if $G$ is a finite group of order $n$, then $G$ is isomorphic to a subgroup of $S_n$.


2. Complete Constructive Proof

For each element $g \in G$, define the left-multiplication map:

$$L_g: G \to G, \quad L_g(x) = gx, \quad \forall x \in G$$
Step 1: $L_g$ is a Permutation of $G$

We show $L_g \in \text{Sym}(G)$ by showing it is bijective:

  • Injective: If $L_g(x) = L_g(y)$, then $gx = gy$. Multiplying by $g^{-1}$ on the left gives $x = y$.
  • Surjective: For any $y \in G$, $L_g(g^{-1}y) = g(g^{-1}y) = y$.

Thus $L_g: G \to G$ is a bijection, meaning $L_g \in \text{Sym}(G)$.

Step 2: The Left Regular Representation

Now define the mapping:

$$\phi: G \to \text{Sym}(G), \quad \phi(g) = L_g$$

We prove that $\phi$ is a group homomorphism: For any $a, b \in G$ and any $x \in G$:

$$[L_a \circ L_b](x) = L_a(L_b(x)) = L_a(bx) = a(bx) = (ab)x = L_{ab}(x)$$

Since this holds for all $x \in G$:

$$L_a \circ L_b = L_{ab} \implies \phi(a) \circ \phi(b) = \phi(ab)$$

Therefore, $\phi$ is a homomorphism.

Step 3: $\phi$ is Injective

We examine the kernel of $\phi$:

$$\ker(\phi) = \{ g \in G : \phi(g) = \text{id}_G \} = \{ g \in G : L_g = \text{id}_G \}$$

If $L_g = \text{id}_G$, then $L_g(x) = x$ for all $x \in G$. Evaluating at $x = e$:

$$L_g(e) = e \implies ge = e \implies g = e$$

Thus $\ker(\phi) = \{e\}$. By Theorem 5.1, $\phi$ is injective.

Step 4: Applying First Isomorphism Theorem

By the First Isomorphism Theorem (Theorem 5.2):

$$G \cong G / \{e\} = G / \ker(\phi) \cong \text{im}(\phi) \le \text{Sym}(G)$$

Therefore, $G$ is isomorphic to the subgroup $\text{im}(\phi) = \{ L_g : g \in G \} \le \text{Sym}(G)$. $\blacksquare$

§5.5 Automorphisms, Inner Automorphisms & Characteristic Subgroups

1. The Automorphism Group $\text{Aut}(G)$

An automorphism of $G$ is an isomorphism $\alpha: G \to G$.

Theorem 5.6 (The Group $\text{Aut}(G)$): Under functional composition $\circ$, the set of all automorphisms of $G$, denoted $\text{Aut}(G)$, forms a group.

  • Identity: $\text{id}_G(x) = x$.
  • Inverse: The functional inverse $\alpha^{-1}$ of an isomorphism is an isomorphism.
  • Associativity: Inherited from composition of functions.

2. Inner Automorphisms $\text{Inn}(G)$

For any fixed element $g \in G$, define the conjugation map by $g$:

$$\gamma_g: G \to G, \quad \gamma_g(x) = g x g^{-1}, \quad \forall x \in G$$

Lemma 5.2: $\gamma_g$ is an automorphism of $G$, called the inner automorphism induced by $g$.

Proof:
  • Homomorphism: $\gamma_g(xy) = g(xy)g^{-1} = (gxg^{-1})(gyg^{-1}) = \gamma_g(x) \gamma_g(y)$.
  • Bijective: $\gamma_g \circ \gamma_{g^{-1}} = \gamma_{g^{-1}} \circ \gamma_g = \text{id}_G$. Thus $\gamma_g \in \text{Aut}(G)$. $\blacksquare$

The set of all inner automorphisms is denoted $\text{Inn}(G) = \{ \gamma_g : g \in G \}$.

Theorem 5.7 ($\text{Inn}(G) \cong G / Z(G)$):

  1. $\text{Inn}(G)$ is a normal subgroup of $\text{Aut}(G)$ ($\text{Inn}(G) \trianglelefteq \text{Aut}(G)$).
  2. $\text{Inn}(G) \cong G / Z(G)$, where $Z(G)$ is the center of $G$.
Complete Proof:

Define the map $\Gamma: G \to \text{Aut}(G)$ by $\Gamma(g) = \gamma_g$.

1. Homomorphism: For any $a, b \in G$ and $x \in G$:

$$\gamma_{ab}(x) = (ab)x(ab)^{-1} = a(bxb^{-1})a^{-1} = \gamma_a(\gamma_b(x)) = (\gamma_a \circ \gamma_b)(x)$$

Thus $\Gamma(ab) = \Gamma(a) \circ \Gamma(b)$.

2. Kernel of $\Gamma$:

$$\begin{aligned} \ker(\Gamma) &= \{ g \in G : \Gamma(g) = \text{id}_G \} = \{ g \in G : \gamma_g(x) = x, \; \forall x \in G \} \\ &= \{ g \in G : gxg^{-1} = x, \; \forall x \in G \} = \{ g \in G : gx = xg, \; \forall x \in G \} \\ &= Z(G) \end{aligned}$$

3. Application of First Isomorphism Theorem:

The image of $\Gamma$ is by definition $\text{Inn}(G)$. By Theorem 5.2:

$$G / \ker(\Gamma) \cong \text{im}(\Gamma) \implies G / Z(G) \cong \text{Inn}(G) \quad \blacksquare$$

3. Outer Automorphisms & Characteristic Subgroups

Definition 5.3 (Outer Automorphism Group): The quotient group:

$$\text{Out}(G) = \frac{\text{Aut}(G)}{\text{Inn}(G)}$$

is called the outer automorphism group of $G$. Elements of $\text{Out}(G)$ are cosets of $\text{Inn}(G)$ called outer automorphism classes.

Definition 5.4 (Characteristic Subgroup): A subgroup $H \le G$ is called characteristic in $G$ (denoted $H \text{ char } G$) if for every $\alpha \in \text{Aut}(G)$:

$$\alpha(H) = H$$

Key Distinctions:

  • Normal: Invariant under inner automorphisms ($\gamma_g(N) = N$ for all $g \in G$).
  • Characteristic: Invariant under all automorphisms.
  • Every characteristic subgroup is normal ($H \text{ char } G \implies H \trianglelefteq G$).
  • Transitivity: While normality is NOT transitive in general ($K \trianglelefteq H \trianglelefteq G \not\implies K \trianglelefteq G$), characteristic subgroups ARE transitive:
$$K \text{ char } H \quad \text{and} \quad H \trianglelefteq G \implies K \trianglelefteq G \quad \blacksquare$$
Tiered Solved Examination Problems & Rigorous Derivations

Step-by-step rigorous derivations with unskipped proofs, categorized into Foundational Concepts, Advanced Structural Analysis, and Honors / Proof Challenge tiers.

Foundational Example 5.1: The Determinant Homomorphism and the Structure of $GL_n(\mathbb{R}) / SL_n(\mathbb{R})$
  1. Let $GL_n(\mathbb{R})$ be the general linear group of invertible $n \times n$ real matrices under matrix multiplication, and let $\mathbb{R}^ = \mathbb{R} \setminus \{0\}$ be the multiplicative group of non-zero real numbers. Prove that the determinant mapping $\det: GL_n(\mathbb{R}) \to \mathbb{R}^$ is an epimorphism (surjective homomorphism). 2. Determine its kernel $\ker(\det)$ and use the First Isomorphism Theorem to deduce the quotient group $GL_n(\mathbb{R}) / SL_n(\mathbb{R})$. 3. Determine the center $Z(GL_n(\mathbb{R}))$.

Step 1: Homomorphism and Surjectivity of $\det$

Let $A, B \in GL_n(\mathbb{R})$. By the multiplicative property of determinants:

$$\det(A B) = \det(A) \cdot \det(B)$$

Since $A, B$ are invertible, $\det(A) \ne 0$ and $\det(B) \ne 0$, so $\det(A), \det(B) \in \mathbb{R}^*$. Thus $\det: GL_n(\mathbb{R}) \to \mathbb{R}^*$ is a group homomorphism.

To show surjectivity, let $c \in \mathbb{R}^*$ be any non-zero real scalar. Consider the diagonal matrix:

$$D_c = \begin{pmatrix} c & 0 & \cdots & 0 \\ 0 & 1 & \cdots & 0 \\ \vdots & \vdots & \ddots & \vdots \\ 0 & 0 & \cdots & 1 \end{pmatrix} \in M_n(\mathbb{R})$$

Its determinant is $\det(D_c) = c \cdot 1 \cdots 1 = c \ne 0$. Thus $D_c \in GL_n(\mathbb{R})$ and $\det(D_c) = c$. Hence $\det$ is surjective (an epimorphism).


Step 2: Kernel and Application of First Isomorphism Theorem

The kernel of $\det$ is:

$$\ker(\det) = \{ A \in GL_n(\mathbb{R}) : \det(A) = 1_{\mathbb{R}^*} = 1 \}$$

By definition, the set of all $n \times n$ real matrices with determinant $1$ is the special linear group:

$$\ker(\det) = SL_n(\mathbb{R})$$

By Theorem 5.1, $SL_n(\mathbb{R}) \trianglelefteq GL_n(\mathbb{R})$. Applying the First Isomorphism Theorem (Theorem 5.2):

$$\frac{GL_n(\mathbb{R})}{\ker(\det)} \cong \text{im}(\det) \implies \frac{GL_n(\mathbb{R})}{SL_n(\mathbb{R})} \cong \mathbb{R}^* \quad \blacksquare$$

Step 3: Center of $GL_n(\mathbb{R})$

An element $A \in GL_n(\mathbb{R})$ is in $Z(GL_n(\mathbb{R}))$ if and only if $A M = M A$ for all $M \in GL_n(\mathbb{R})$. Let $E_{ij}$ denote the matrix with $1$ in entry $(i, j)$ and $0$ elsewhere. For $i \ne j$, the elementary shear matrix $I + E_{ij} \in GL_n(\mathbb{R})$. The condition $A(I + E_{ij}) = (I + E_{ij})A$ simplifies to:

$$A E_{ij} = E_{ij} A$$

Computing both sides:

  • The $k$-th row of $A E_{ij}$ has entry $(k, j)$ equal to $A_{ki}$, and 0 elsewhere.
  • The $k$-th row of $E_{ij} A$ has entry $(i, l)$ equal to $A_{jl}$ when $k=i$, and 0 for $k \ne i$.

Setting $k \ne i$ forces $A_{ki} = 0$ for all $k \ne i$, so $A$ must be diagonal. Setting $k = i$ and $l = j$ forces $A_{ii} = A_{jj}$ for all $i, j$. Thus all diagonal entries of $A$ are equal: $A_{11} = A_{22} = \dots = A_{nn} = \lambda \ne 0$. Therefore, the center consists solely of non-zero scalar multiples of the identity:

$$Z(GL_n(\mathbb{R})) = \{ \lambda I_n : \lambda \in \mathbb{R}^* \} \cong \mathbb{R}^* \quad \blacksquare$$
Advanced Example 5.2: Automorphism Groups of Cyclic Groups and the Klein Four-Group
  1. Prove that for any positive integer $n$, the automorphism group of the cyclic group $\mathbb{Z}_n$ is isomorphic to the group of units $U(n) = (\mathbb{Z}_n)^\times$: $\text{Aut}(\mathbb{Z}_n) \cong U(n)$. 2. Let $V_4 = \mathbb{Z}_2 \times \mathbb{Z}_2$ be the Klein four-group. Prove that $\text{Aut}(V_4) \cong S_3 \cong GL_2(\mathbb{F}_2)$.

Part 1: $\text{Aut}(\mathbb{Z}_n) \cong U(n)$

Let $G = \mathbb{Z}_n = \langle 1 \rangle$ under addition modulo $n$. Any endomorphism $\alpha: \mathbb{Z}_n \to \mathbb{Z}_n$ is completely determined by $\alpha(1)$:

$$\alpha(k) = \alpha(\underbrace{1 + \dots + 1}_{k \text{ times}}) = k \cdot \alpha(1)$$

Let $a = \alpha(1) \in \mathbb{Z}_n$. For $\alpha$ to be an automorphism, it must be surjective (and since $\mathbb{Z}_n$ is finite, this is equivalent to being bijective). A linear map $k \mapsto ka$ is surjective on $\mathbb{Z}_n$ if and only if $a$ generates $\mathbb{Z}_n$. By Theorem 2.4, $\langle a \rangle = \mathbb{Z}_n$ if and only if $\gcd(a, n) = 1$. Thus each $a \in U(n) = (\mathbb{Z}_n)^\times$ defines a unique automorphism $\alpha_a(k) = ka \pmod n$.

Define the map $\Psi: \text{Aut}(\mathbb{Z}_n) \to U(n)$ by:

$$\Psi(\alpha) = \alpha(1) \pmod n$$
  • Bijective: As shown above, every $a \in U(n)$ yields a unique automorphism $\alpha_a$, and $\Psi(\alpha_a) = a$.
  • Homomorphism: For $\alpha_a, \alpha_b \in \text{Aut}(\mathbb{Z}_n)$:
$$\Psi(\alpha_a \circ \alpha_b) = (\alpha_a \circ \alpha_b)(1) = \alpha_a(\alpha_b(1)) = \alpha_a(b) = b \cdot a \equiv a b \pmod n$$

Thus $\Psi(\alpha_a \circ \alpha_b) = \Psi(\alpha_a) \cdot \Psi(\alpha_b)$. Therefore, $\Psi$ is an isomorphism:

$$\text{Aut}(\mathbb{Z}_n) \cong U(n) \quad \blacksquare$$

Part 2: $\text{Aut}(V_4) \cong S_3 \cong GL_2(\mathbb{F}_2)$

Let $V_4 = \{ (0,0), (1,0), (0,1), (1,1) \}$ under component-wise addition modulo 2. Every non-zero element has order 2:

$$e = (0,0), \quad a = (1,0), \quad b = (0,1), \quad c = (1,1)$$

Notice that $a + b = c, \; b + c = a, \; a + c = b$.

Method 1 (Permutation of non-identity elements): Any automorphism $\sigma \in \text{Aut}(V_4)$ must fix $e = (0,0)$: $\sigma(e) = e$. Therefore, $\sigma$ must permute the 3 non-identity elements $X = \{a, b, c\}$. This defines a homomorphism:

$$\rho: \text{Aut}(V_4) \to \text{Sym}(X) \cong S_3, \quad \rho(\sigma) = \sigma|_X$$
  • Injectivity of $\rho$: If $\rho(\sigma) = \text{id}_X$, then $\sigma$ fixes $a, b, c$ and fixes $e$, so $\sigma = \text{id}_{V_4}$.

Thus $\ker(\rho) = \{\text{id}_{V_4}\}$, meaning $\rho$ is injective.

  • Surjectivity of $\rho$: Any permutation of $\{a, b, c\}$ preserves the group operation!

For instance, consider the transposition $(a \; b)$: $\sigma(a) = b$, $\sigma(b) = a$, and $\sigma(c) = \sigma(a+b) = \sigma(a) + \sigma(b) = b + a = c$. This is a valid automorphism! Similarly, the 3-cycle $(a \; b \; c)$ gives $\sigma(a)=b, \sigma(b)=c, \sigma(c)=a$, with $\sigma(a+b) = b+c = a = \sigma(c)$. Since the transpositions generate $S_3$, every permutation in $S_3$ is realized by an automorphism of $V_4$. Thus $|\text{Aut}(V_4)| = |S_3| = 6$, and:

$$\text{Aut}(V_4) \cong S_3$$

Method 2 (Vector Space Approach): $V_4$ is canonically isomorphic to the 2-dimensional vector space $\mathbb{F}_2^2$ over the field with 2 elements $\mathbb{F}_2 = \{0, 1\}$. Since addition of vectors is the group operation, and scalar multiplication by $0$ and $1$ is trivial, any group automorphism of $V_4$ is automatically $\mathbb{F}_2$-linear! Thus:

$$\text{Aut}(V_4) \cong GL_2(\mathbb{F}_2)$$

The order of $GL_2(\mathbb{F}_2)$ is:

$$|GL_2(\mathbb{F}_2)| = (2^2 - 1)(2^2 - 2) = (3)(2) = 6$$

The only non-abelian group of order 6 is $S_3$. Therefore:

$$\text{Aut}(V_4) \cong S_3 \cong GL_2(\mathbb{F}_2) \quad \blacksquare$$
Honors / Proof Challenge Example 5.3: Complete Classification of Groups of Order $2p$ ($p$ Odd Prime)

Let $p$ be an odd prime number ($p \ge 3$), and let $G$ be a group of order $2p$. 1. Prove that $G$ contains a normal subgroup $N$ of order $p$, and that $N$ is cyclic. 2. Prove that $G$ contains an element $s$ of order 2. 3. By studying the conjugation action of $s$ on $N$ via $\text{Aut}(N)$, prove that $G$ is isomorphic to either the cyclic group $\mathbb{Z}_{2p}$ or the dihedral group $D_{2p}$.

Step 1: Existence and Normality of Cyclic Subgroup of Order $p$

We first show that $G$ contains an element of order $p$. By Corollary 4.2.1, the possible orders of elements in $G$ are divisors of $2p$: $\{1, 2, p, 2p\}$. If $G$ contains an element of order $2p$, then $G \cong \mathbb{Z}_{2p}$ is cyclic, and $N = \langle g^2 \rangle$ has order $p$. Suppose $G$ does not contain an element of order $2p$. Can all non-identity elements have order 2? If every $g \ne e$ has order 2, then for all $x, y \in G$:

$$x y = (x y)^{-1} = y^{-1} x^{-1} = y x$$

So $G$ would be abelian. In an abelian group where every non-identity element has order 2, $G \cong (\mathbb{Z}_2)^k$, so $|G| = 2^k$. However, $|G| = 2p$ where $p \ge 3$ is odd, which is never a power of 2! Hence $G$ must contain an element $r$ of order $p$.

Let $N = \langle r \rangle = \{1, r, r^2, \dots, r^{p-1}\}$. Since $|r| = p$, $|N| = p$, and $N$ is cyclic: $N \cong \mathbb{Z}_p$. The index of $N$ in $G$ is:

$$[G : N] = \frac{|G|}{|N|} = \frac{2p}{p} = 2$$

By Proposition 4.1, any subgroup of index 2 is normal:

$$N \trianglelefteq G$$

Step 2: Existence of an Element of Order 2

Since $|N| = p$ and $|G| = 2p$, there are $p$ elements in the complement $G \setminus N$. Let $x \in G \setminus N$. Since $N \trianglelefteq G$, the quotient group $G/N$ has order 2, so:

$$(xN)^2 = x^2 N = N \implies x^2 \in N$$

Thus $x^2 = r^k$ for some $k \in \{0, 1, \dots, p-1\}$. The order of $x$ must divide $2p$, so $|x| \in \{2, p, 2p\}$.

  • If $|x| = 2$, we are done: set $s = x$.
  • If $|x| = p$, then $\langle x \rangle$ is a subgroup of order $p$ distinct from $N$ (since $x \notin N$).

Then $N \cap \langle x \rangle = \{e\}$, so $|N \langle x \rangle| = \frac{p \cdot p}{1} = p^2 > 2p$, a contradiction! (Since $p \ge 3$). Thus no element outside $N$ can have order $p$.

  • If $|x| = 2p$, then $x^p$ has order 2: set $s = x^p \in G \setminus N$.

Therefore, there always exists an element $s \in G$ with $|s| = 2$. Note that since $s$ has order 2 and $|N| = p$ is odd, $s \notin N$.


Step 3: Action by Conjugation and Final Classification

Since $N \trianglelefteq G$ and $s \in G$, conjugation by $s$ is an automorphism of $N$:

$$\gamma_s: N \to N, \quad \gamma_s(r) = s r s^{-1} = s r s \quad (\text{since } s = s^{-1})$$

Since $N = \langle r \rangle$, $s r s \in N$, so:

$$s r s = r^k \quad \text{for some } k \in \{1, 2, \dots, p-1\}$$

Now compute $s^2 r s^{-2}$:

$$r = s^2 r s^{-2} = s (s r s) s = s r^k s = (s r s)^k = (r^k)^k = r^{k^2}$$

Therefore:

$$r^{k^2 - 1} = e \implies k^2 \equiv 1 \pmod p$$

Since $p$ is prime, the congruence $k^2 \equiv 1 \pmod p$ factors as:

$$(k - 1)(k + 1) \equiv 0 \pmod p \implies p \mid (k-1) \quad \text{or} \quad p \mid (k+1)$$

This yields exactly two solutions modulo $p$:

$$k \equiv 1 \pmod p \quad \text{or} \quad k \equiv -1 \equiv p-1 \pmod p$$
Case 1: $k \equiv 1 \pmod p$

Then $s r s = r \implies s r = r s$. Here $s$ and $r$ commute! Since $|r| = p$ and $|s| = 2$ with $\gcd(p, 2) = 1$, the element $g = rs$ has order:

$$|g| = |rs| = \text{lcm}(|r|, |s|) = \text{lcm}(p, 2) = 2p$$

Therefore, $G = \langle rs \rangle$ is cyclic:

$$G \cong \mathbb{Z}_{2p}$$
Case 2: $k \equiv -1 \pmod p$

Then $s r s = r^{-1} \implies s r s = r^{-1}$. Together with $r^p = 1$ and $s^2 = 1$, the group presentation of $G$ is:

$$G = \langle r, s \mid r^p = 1, \; s^2 = 1, \; s r s = r^{-1} \rangle$$

This is precisely the presentation of the dihedral group of order $2p$:

$$G \cong D_{2p}$$

Conclusion: Every group of order $2p$ ($p$ odd prime) is isomorphic to either $\mathbb{Z}_{2p}$ or $D_{2p}$. $\blacksquare$