Problem 5: Prove or disprove: Zp × Zp is a cyclic group.
Statement: The statement is False.
Proof: Let G = Zp × Zp . The order of the group is |G| = p · p = p2 . For G to be cyclic, it must contain
a generator element g of order p2 . Let (a, b) be an arbitrary element in G. The order of an element in a
direct product is the least common multiple of the orders of its components:
|(a, b)| = lcm(|a|, |b|)
Since a, b ∈ Zp , the order of a and b is either 1 (if the element is 0) or p (since Zp is a group of prime
order). Therefore, lcm(|a|, |b|) is either 1 or p. Since the maximum order of any element in G is p, and
p < p2 (for any prime p), no element can generate the group. Thus, Zp × Zp is not cyclic.
Solution 6
Problem: Let |G| = pn where p is a prime. Does G contain an element of order p?
Answer: Yes.
Proof: We invoke Cauchy’s Theorem, which states that if a prime p divides the order of a finite group
G, then G contains an element of order p. Since |G| = pn and n ≥ 1, p clearly divides |G|. Therefore,
an element of order p exists.
(Alternative Elementary Proof without Cauchy’s Theorem): Let x ∈ G be a non-identity element. By
Lagrange’s Theorem, the order of x divides |G|, so |x| = pk for some 1 ≤ k ≤ n. If k = 1, then x has
k−1
order p. If k > 1, consider the element y = xp . The order of y is:
k−1 k
(xp )p = xp = e
k−1
Since xp ̸= e (as the order of x is exactly pk ), y is an element of order p.
Solution 7
Theorem: Let G be a finite group such that |G| = pn where p is a prime and n ≥ 1. Then G contains
an element of order p.
Proof: We proceed by induction on n. For n = 1, |G| = p. By Lagrange’s Theorem, any non-identity
element x ∈ G generates a subgroup ⟨x⟩. Since |⟨x⟩| > 1 and |⟨x⟩| divides p, |⟨x⟩| = p. Thus x has order
p.
Assume the result holds for all groups of order pk with k < n. Consider the class equation for G:
m
X
|G| = |Z(G)| + [G : CG (xi )]
i=1
where xi are representatives of the distinct conjugacy classes with more than one element.
1. For each i, [G : CG (xi )] > 1 and [G : CG (xi )] divides |G| = pn . Thus, p | [G : CG (xi )].
P
2. Since p | |G| and p | [G : CG (xi )], it follows that p | |Z(G)|.
1
3. Since e ∈ Z(G), |Z(G)| ≥ p.
Case 1: There exists a proper subgroup H < G such that p | |H|. By the inductive hypothesis, H
contains an element of order p. Since H ⊆ G, this element is in G.
Case 2: No proper subgroup of G has order divisible by p. This case is actually impossible by the result
of step 3 (Z(G) is a subgroup), but we can conclude directly from Z(G): Pick z ∈ Z(G) such that z ̸= e.
k−1
By Lagrange, |z| = pk for some 1 ≤ k ≤ n. Let y = z p . Then
k−1 k
y p = (z p )p = z p = e
k−1
Since z has order pk , z p ̸= e. Thus, y is an element of order p. ■
Solution 8
Problem: Let H and K be subgroups of a group G.
(a) Prove that the intersection xH ∩ yK of two cosets of H and K is either empty or else is
a coset of the subgroup H ∩ K.
Proof: Let S = xH ∩ yK. Case 1: S = ∅. The statement holds. Case 2: S ̸= ∅. Let z ∈ S. Then
z ∈ xH and z ∈ yK. Since two cosets are either disjoint or identical, z ∈ xH =⇒ xH = zH. Similarly,
z ∈ yK =⇒ yK = zK. Substituting these back into the intersection:
S = zH ∩ zK
S = {zh | h ∈ H} ∩ {zk | k ∈ K}
S = {g ∈ G | g ∈ zH and g ∈ zK}
An element g is in zH ∩ zK if and only if z −1 g ∈ H and z −1 g ∈ K, which implies z −1 g ∈ H ∩ K. Thus,
g ∈ z(H ∩ K). Therefore, S = z(H ∩ K), which is a left coset of the subgroup H ∩ K.
(b) Prove that if H and K have finite index in G then H ∩ K also has finite index in G.
Proof: Let [G : H] and [G : K] be finite. We define a map ψ : G/(H ∩ K) → G/H × G/K given by:
ψ(g(H ∩ K)) = (gH, gK)
1. Well-defined: If g1 (H ∩ K) = g2 (H ∩ K), then g2−1 g1 ∈ H ∩ K. This implies g2−1 g1 ∈ H (so
g1 H = g2 H) and g2−1 g1 ∈ K (so g1 K = g2 K). The image is independent of the representative.
2. Injective: If ψ(g1 (H ∩ K)) = ψ(g2 (H ∩ K)), then (g1 H, g1 K) = (g2 H, g2 K). This implies g1 H =
g2 H =⇒ g2−1 g1 ∈ H and g1 K = g2 K =⇒ g2−1 g1 ∈ K. Thus g2−1 g1 ∈ H ∩ K, so the cosets are
equal. Since ψ is an injection, the number of cosets in G/(H ∩ K) is less than or equal to the size
of the product set:
[G : H ∩ K] ≤ [G : H] · [G : K]
Since [G : H] and [G : K] are finite, their product is finite. Thus, [G : H ∩ K] is finite.
2
Solution 9
Problem: Determine whether or not G is isomorphic to the product group H × K in the following cases.
(a) G = R× , H = {±1}, K = {positive real numbers}.
Answer: Yes.
Proof: Define a map ϕ : H × K → G by ϕ(h, k) = h · k.
1. Homomorphism: Since G is abelian, ϕ((h1 , k1 )(h2 , k2 )) = ϕ(h1 h2 , k1 k2 ) = (h1 h2 )(k1 k2 ) =
(h1 k1 )(h2 k2 ) = ϕ(h1 , k1 )ϕ(h2 , k2 ).
2. Bijection: Any non-zero real number x ∈ G can be uniquely written as x = sgn(x) · |x|.
• Here, sgn(x) ∈ {1, −1} = H.
• |x| ∈ R>0 = K. Since the representation is unique, ϕ is a bijection. Conclusion: G ∼
= H ×K.
(b) G = C× , H = S 1 (unit circle), K = R>0 (positive reals).
Answer: Yes.
Proof: Define ϕ : H × K → G by ϕ(z, r) = z · r.
1. Homomorphism: Standard complex multiplication commutes, so this is a homomorphism.
2. Bijection: Any non-zero complex number w ∈ C× has a unique polar decomposition w = reiθ .
• Let r = |w| ∈ K.
• Let z = w/|w|. Since |z| = |w|/|w| = 1, z ∈ H. Thus w = z · r uniquely. Conclusion:
G∼= H × K.
(c) G = {Upper triangular invertible 2 × 2}, H = {Diagonal invertible}, K = {Unitriangular}.
Answer: No.
Proof: For G to be isomorphic to the direct product H ×K, elements of H must commute with elements
of K (since elements in different
factors of a direct
product
commute: (h, 1)(1, k) = (h, k) = (1, k)(h, 1)).
2 0 1 1
Let’s test this: Let h = ∈ H and k = ∈ K.
0 1 0 1
2 0 1 1 2 2
hk = =
0 1 0 1 0 1
1 1 2 0 2 1
kh = =
0 1 0 1 0 1
Since hk ̸= kh, thegroupG is not abelian. However, H is abelian (diagonal matrices) and K is abelian
1 x
(matrices of form add the x term). The direct product of two abelian groups is abelian. Since
0 1
G is non-abelian, it cannot be isomorphic to H × K. (Note: G is actually a semi-direct product K ⋊ H).
Solution 10
Problem: Classify groups of order 6.
(a) Case 1: G contains an element of order 6. If there exists an element x ∈ G with |x| = 6, then
G is generated by x. Classification: G ∼
= C6 ∼
= Z6 (Cyclic Group).
(b) Case 2: G contains an element of order 3 but none of order 6. Let x ∈ G be an element of
order 3. Let H = ⟨x⟩ = {1, x, x2 }. Since [G : H] = 6/3 = 2, H is a normal subgroup of G. Let y ∈ G \ H.
3
The cosets are H and yH. Thus G = H ∪ yH. Consider the element y 2 . Since G/H is a group of order
2, y 2 ∈ H.
• If y 2 = x or y 2 = x2 , then the order of y would be lcm(2, 3) = 6, which contradicts the assumption.
• Therefore, y 2 = 1. So y has order 2. Since H is normal, yxy −1 ∈ H. It must be an element of
order 3, so yxy −1 ∈ {x, x2 }.
• Subcase 1: yxy −1 = x. Then yx = xy. G is abelian. The element xy has order lcm(3, 2) = 6.
Contradiction.
• Subcase 2: yxy −1 = x2 = x−1 . This gives the presentation: ⟨x, y | x3 = 1, y 2 = 1, yx = x−1 y⟩.
Classification: This is the definition of the Dihedral Group D3 (isomorphic to S3 ).
(c) Case 3: All elements of G have order 1 or 2. If every non-identity element has order 2, then
for any a, b ∈ G: ab = (ab)−1 = b−1 a−1 = ba. Thus, G must be abelian. If G is a finite abelian group
where every element has order 2, G is a vector space over the field F2 . Thus |G| = 2k for some integer k.
However, |G| = 6, which is not a power of 2. Classification: This case yields no groups (The empty
set).
Summary: The only groups of order 6 are Z6 (cyclic) and S3 (non-abelian).