16 Homomorphisms, Isomorphisms

1

Question

Let φ:G→H be a Homomorphism (or Group Morphism).
a. Prove that φ(xn)=φ(x)n for all n∈Z+
b. Do (a) for n=−1 and deduce that φ(xn)=φ(x)n for all n∈Z.

Proof

a. We'll do this by induction. Clearly n=1 works here. Let the inductive hypothesis be that φ(xn−1)=φ(x)n−1 and any lower powers. Now notice that since φ is a homomorphism then:

φ(xn)=φ(xn−1⋅x)=φ(x)n−1φ(x)=φ(x)n

thus completing the inductive step.

b. We'd do an induction on n∈Z+ and proving:

φ(x−n)=φ(x)−n

To do this, first consider the base case n=−1:

φ(x−1)=?

Notice we want to prove that the left is φ(x)'s inverse, which we can verify:

φ(x−1)φ(x)=φ(x−1⋅x)=φ(1)=1

since φ is a Homomorphism (the first step, the other order works the same way) and the last identity step holds true since for any x∈G (using Cancellation Theorem):

φ(x⋅1)=φ(x)φ(1)=φ(x)⇒φ(1)=1

Continuing on, you'd make the inductive step just like we did for (a).

☐

2

Question

If φ:G→H is an Isomorphism, prove that |φ(x)|=|x| for all x∈G. Deduce that any two isomorphic groups have the same number of elements of order n for each n∈Z+. Is the result true if φ is only assumed to be a homomorphism?

Proof

Let x∈G. If we let |x|=n then:

xn=1

Now notice using 16 Homomorphisms, Isomorphisms#1 (since being an isomorphism implies it's a homomorphism) that:

φ(x)n=φ(xn)=φ(1)=1

Thus |φ(x)|=n=|x| as desired.

Showing that G,H have the same number of elements of order n for each n∈Z+, let n∈Z+. Say G only has kn elements of order n. Then φ is a bijection that shows that there are at least kn elements of the same order n (use the first part). But notice that if there were any more elements, then we can consider those extra elements s1,… and apply φ−1 to them and get an element φ−1(s1)∈G we missed (ie: |s1|=n which is a contradiction), so then we must have exactly kn elements for that order n.

Notice that this doesn't work for homomorphisms (where that φ−1 existence may not be guarunteed). For example, the map h:Z→Z3Z
where h(u)=umod3 is a homomorphism, but not an isomorphism. Furthermore, Z has no elements of order say 2 but Z3Z has one (namely 1―).

☐

5

Question

Prove that the additive groups R and Q are not isomorphic.

Proof

Clearly we need to use the Irrationality of sqrt(2), so we want to try to make an isomorphism such that f(2)=2 to break it. Namely, assume for contradiction that such an isomorphism f:R→Q exists.
Then:

f(x+y)=f(x)+f(y)

and f−1 exists. Consider f(x) where x is an integer. Then:

f(x)=f(∑i=1x1)=∑i=1xf(1)=xf(1)=x

So then f maps each integer to itself. Similarly for rational numbers ab∈Q:

b⋅f(ab)=∑i=1bf(ab)=f(∑i=1bab)=f(a)=a

So:

f(ab)=ab

So if we consider f(x) when x∈I then clearly if f mapped x to itself then that would contradict x∉Q. But if it didn't map to itself then it would have to map to a rational number that's already been mapped (which contradicts f being a bijection).

Thus we have a contradiction, so f cannot exist, so R and Q are not isomorphic.

☐

11

Question

Let A,B be Groups. Prove that A×B≅B×A.

Proof

For clarity say A has the operation ⋅ and B has the operation ⋆.

Define the map φ:A×B→B×A such that for all a∈A,b∈B:

φ(a,b)=(b,a)

We'll first show that φ is a valid Homomorphism (or Group Morphism). Namely, let c∈A,d∈B. Then:

φ((a,b)(c,d))=φ(a⋅b,c⋆d)=(c⋆d,a⋅b)=(c,a)(d,b)=φ(a,c)φ(b,d)

To show that φ is a bijection. I claim that φ2 is it's inverse:

φ2(a,b)=φ(b,a)=(a,b)

Thus φ2=I and therefore φ is its own inverse, showing that φ is a bijection.

☐

13

Question

Let G and H be groups and let φ:G→H be a Homomorphism (or Group Morphism). Prove that the image of φ, namely φ(G), is a Subgroup of H. Prove that if φ is injective then G≅φ(G).

Proof

Using the Subgroup#The Subgroup Criterion conditions, we need to show that ∀h,k∈φ(G)(hk,h−1∈φ(G)) in order for φ(G) to be a subgroup of H (as already φ(G)⊆H).

Let h,k∈φ(G) be arbitrary. Then notice that by φ being a homomorphism. Then by being in the image then ∃gh,gk∈G where φ(gh)=h and φ(gk)=k. Notice that by φ being a homomorphism:

φ(ghgk)=φ(gh)φ(gk)=hk

So then we've determined that hk∈φ(G) as desired. Similarly, since h∈φ(G) then ∃gh∈G where φ(gh)=h. Then I claim that h−1=φ(gh−1)=φ(gh)−1. Verifying:

hh−1=φ(gh)φ(gh−1)=φ(ghgh−1)=φ(1)=e

And the other direction is checked similarly. Notice then that h−1∈φ(G) as desired.

Therefore, we've shown that φ(G)≤H.

For the second part of the proof, suppose φ is injective. We know that φ is already a homomorphism, so if we can show that the restricted operator φ⋆:G→φ(G) such that φ⋆:g↦φ(g) is surjective, then we are done (since φ⋆ is a restricted operator of an injective function, then it itself is already injective). But notice that if we consider any h∈φ(G) then ∃g∈G where φ⋆(g)=h, showing surjectivity! Thus then G≅φ(G).

☐

14

Question

Let G,H be groups and let φ:G→H be a Homomorphism (or Group Morphism). Define the kernel of φ to be {g∈G|φ(g)=1H}. Prove that the kernel of φ is a Subgroup of G. Prove that φ is injective iff the kernel of φ is the identity subgroup of G (the group with only the identity element).

Proof

  1. To show it's a subgroup, we'll show closure of the restricted operator on ker⁡(φ) for the binary operator, as well as inverses.

For the former, let h,k∈ker⁡(φ). Then h,k∈G and φ(h),φ(k)=1H. By φ being homomorphic then:

φ(hk)=φ(h)φ(k)=1H1H=1H

Thus then since hk∈G then hk∈ker⁡(φ), showing closure under the binary operation.

For the latter, let h∈ker⁡(φ), so like before then φ(h)=1H where h∈G. Then notice that h−1∈G exists and:

φ(h−1)=φ(h)−1=1H−1=1H

thus then h−1∈ker⁡(φ).

(→): Suppose φ is injective. To show that ker⁡(φ)={1H}, assume that it wasn't, that ∃h≠1H∈ker⁡(φ). Then notice that both h,1H∈ker⁡(φ), so then φ(h)=φ(1H)=1H. But notice then that since φ is injective that that implies that h=1H, which is a contradiction! Thus then ker⁡(φ)={1H}.

(←): Suppose ker⁡(φ)={1H}. Let a,b∈ker⁡(φ) be arbitrary, and suppose φ(a)=φ(b). Clearly a=b=1H so φ is injective.

☐

16

Question

Let A,B be groups and let G be their direct product, A×B. Prove that the maps π1:G→A and π2:G→B defined by π1:(a,b)↦a and π2:(a,b)↦b are Homomorphism (or Group Morphism)s and find their Kernels.

Proof

Let (a,b),(c,d)∈A×B be arbitrary. Notice that:

π1((a,b)(c,d))=π1((ac,bd))=ac=π1(a,b)π1(c,d)π2((a,b)(c,d))=π2((ac,bd))=bd=π2(a,b)π2(c,d)

thus both are homomorphisms. For their kernels we want to find {g∈G|π1(g)=1A} first. Namely {(1A,b)|b∈B} fits that bill, so then ker⁡(π1)={(1A,b)|b∈B}. A similar arguments shows that ker⁡(π2)={(a,1B)|a∈A}.

☐

17

Question

Let G be any Group. Prove that the map from G to itself defined by g↦g−1 is a Homomorphism (or Group Morphism) iff G is an Abelian Group.

Proof

(⇒): Suppose φ (defined above) is a homomorphism. Let a,b∈G be arbitrary. Then:

ab=(b−1a−1)−1=φ(b−1a−1)=φ(b)−1φ(a)−1=(b−1)−1(a−1)−1=ba

(⇐): Suppose G is abelian, and let a,b∈G. Now notice:

φ(ab)=(ab)−1=b−1a−1=φ(b)φ(a)=φ(a)φ(b)G is abelian

☐

18

Question

Let G be any Group. Prove that the map from G to itself defined by g↦g2 is a homomorphism iff G is an Abelian Group.

Proof

(⇒): Suppose φ (defined above) is a homomorphism, and let a,b∈G. Then:

φ(ab)=φ(a)φ(b)(ab)2=a2b2abab=a2b2ba=ab(Left and Right Cancellation)

(⇐): Suppose G is an Abelian Group. Then:

φ(ab)=(ab)2=abab=aabb=a2b2=φ(a)φ(b)

☐

20

Question

Let G be a group and let Aut(G) be the set of all Isomorphisms from G→G. Prove that Aut(G) is a Group under function composition (called the automorphism group of G and the elements of Aut(G) are called automorphisms of G).

Proof

First, let's establish that there's closure under these operations, which is easy to verify. If φ1,φ2∈Aut(G) then φ1∘φ2∈Aut(G) since it still maps G→G, it's a composition of bijections (so φ1∘φ2 is a bijection), and is a homomorphism if we let g,j∈G and show that:

(φ1∘φ2)(gj)=φ1(φ2(gj))=φ1(φ2(g)φ2(j))=φ1(φ2(g))φ1(φ2(j))=(φ1∘φ2)(g)(φ1∘φ2)(j)

For the individual properties:

  1. By function composition applying from right to left, then (φ1φ2)φ3=φ1(φ2φ3).
  2. The identity element here is the identity map φ0:g↦g. Namely, let φ∈Aut(G). Then:
(φ0φ)=φ=(φφ0)
  1. For each φ∈Aut(G), by being an isomorphism and thus a bijection, then φ−1 is the inverse element. Namely:
φφ−1=φ0=φ−1φ

☐