MAA 201 2022-2023

Reduction of endomorphisms

Exercises - Week IV - (Ring) Ideals

Exercise 1. 1. Show by using Bezout theorem that (Z/nZ)∗ = {[a], a ∧ n = 1}

for any n ∈ N∗ .
2. We let φ(n) be the cardinality of (Z/nZ)∗ . Show that φ(pk ) = (p − 1)pk−1
for any prime number p and any k ∈ N∗ .
3. By using the Chinese remainder lemma, show that φ(pq) = φ(p)φ(q) for
p ∧ q = 1.
4. Conclude that φ(n) = i∈I (pi − 1)pki i −1 when n = i∈I pki i is the prime
Q Q
factorization of n.
5. By using Exercise 1 of Sheet 2, prove Euler-Fermat Theorem, i.e. for any a
with a ∧ n = 1,
aφ(n) ≡ 1 (mod n).

Solution:
1. See course.
2. any nonnegative integer a < pk with a∧p 6= 1 is a multiple of p, i.e. a = 0, p, 2p, · · · (pk−1 −
1)p. There are pk−1 numbers of this form. Therefore by the previous question, |(Z/nZ)∗ | =
|Z/nZ| − pk−1 = (p − 1)pk−1 .
3. for p∧q = 1, the Chinese remainder lemma says that the rings Z/pZ×Z/qZ and Z/pqZ are
isomorphic, therefore so are their group of invertible elements. Moreover (Z/pZ×Z/qZ)∗ =
(Z/pZ)∗ × (Z/qZ)∗ . Therefore φ(pq) = φ(p)φ(q).
4. Immediate.
5. By using Exercise 1 of Sheet 2, the order of [a] in the group (Z/nZ)∗ divides the order
φ(n) of (Z/nZ)∗ . Therefore [a]φ(n) = [1] which is equivalent to aφ(n) ≡ 1 (mod n) .

Exercise 2.
Solve in Z the equation
12x ≡ 15 (mod 21).

Solution: This is equivalent to solve 4x ≡ 5 (mod 7). Then 4 is invertible modulo 7 with inverse
2, therefore the solutions are the integers of the form 10 + 7k, k ∈ Z.

Exercise 3.
For d ∈ N we let Ad = {(x, y) ∈ Z2 , x − y ∈ dZ}.
 1. Show that Ad is a subring of Z2 .
2. Let A be a subring of Z2 . Show that H = {x ∈ Z, (x, 0) ∈ A} is a group.
3. Conclude that A = Ad for some d.

Solution:
1. Clearly Ad is a subgroup of Z2 . Let us check it is stable under ×. Let (x, y), (x0 , y 0 ) ∈ Ad
then we have xx0 − yy 0 = (x − y)x0 + y(x0 − y 0 ) ∈ dZ as x − y, x0 − y 0 belong to dZ.
2. If x, y ∈ H, then (x − y, 0) = (x, 0) − (y, 0) belongs to A as A is a subgroup of (Z2 , +).
3. As a subgroup of Z, there is d with H = dZ. We check that A = Ad : for any (x, y) ∈ A we
have (x − y, 0) = (x, y) − y(1, 1) ∈ A, as the unit (1, 1) in Z2 lies in a subring A. Therefore
x − y ∈ dZ. Conversely if x − y ∈ dZ then (x, y) = (x − y, 0) + y(1, 1) lies in A.

Exercise 4. 1. Given two ideals I and J of R, their sum is defined as

I + J = {i + j | i ∈ I, j ∈ J} ⊂ R.

Show that I + J is an ideal of R and that it is the smallest ideal containing

both I and J.
2. Given two ideals I and J of R, we define their product as
( n )
X
I ·J = ik jk | n > 1, ik ∈ I, jk ∈ J .
k=1

Show that I · J is an ideal of R and that it is contained both in I and J.

3. If I and J are two ideals of the ring of integers Z, what are the ideals I + J
and I · J ?

Solution:
1. Let y = i+j, y 0 = i0 +j 0 be two elements of I +J then x+y = i+j +i0 +j 0 = (i+i0 )+(j +j 0 )
is clearly in I + J. If x ∈ R then xy = x(i + j) = xi + xj and since xi ∈ I and xj ∈ J by
definition of an ideal we find that xy ∈ I + J.
Suppose that Z ⊂ R is an ideal containing I and J. Then for all i ∈ I and j ∈ J, i ∈ Z
and j ∈ Z thus i + j ∈ Z. This shows that I + J ⊂ Z.
0
P P 0 0 0
P y =
2. Let iP
k jk and y = ik jk be two elements of IJ and let x ∈ R. Then
P y+y =
0 0
( ik jk ) + ( ik jk ) is clearly an element of IJ. Furthermore we have xy = (xik )jk and
since I is an ideal xik is in I for each k thus xy ∈ IJ. We have thus shown that IJ is an
ideal of R.
P
Let us now show that IJ is contained in I and J. Let y = ik jk be an element of IJ.
Since I is an ideal and ik ∈ I for all k we find that ik jk is in I for all k thus y ∈ I.
Similarly y ∈ J.

Page 2
 3. Let nZ and mZ be ideals of Z with n, m > 0. Let’s show that nZ + mZ = gcd(n, m)Z.
Indeed if u = kn + k 0 m is an element of nZ + mZ then it is clearly divisible by gcd(n, m)
thus nZ+mZ ⊂ gcd(n, m)Z. Conversely by Bézout’s identity there exist x, y ∈ Z such that
xn + ym = gcd(n, m). Thus any u = k × gcd(n, m) ∈ gcd(n, m)Z is equal to kxn + kym
and is thus an element of nZ + mZ i.e. gcd(n, m)Z ⊂ nZ + mZ. By double inclusion this
indeed shows that gcd(n, m)Z = nZ + mZ.
It is obvious that nZmZ = (nm)Z.

Exercise 5.
Let R be a commutative ring and let I be an ideal of R. We say that I is a prime
ideal if I 6= R and if, for all x, y ∈ R such that x · y ∈ I, at least one of the
elements x and y lies in I. We say that I is a maximal ideal if I 6= R and if any
ideal J of R such that I ⊂ J and I 6= J is equal to R.
1. When is the ideal {0} prime ?
2. What are the prime and the maximal ideals the ring of integers Z ?
3. Let I be an ideal of R and x ∈ R \ I. Show that {y ∈ R, ∃i ∈ I and ∃a ∈
R s.t. y = i + ax} defines an ideal of R.
4. By using the previous question, show that a maximal ideal is a prime ideal.
5. Is the converse true ?

Solution:
1. 0 is prime if, and only if, there is no zero divisor in R, i.e. if xy = 0 then x = 0 or y = 0.
Note that this property is not always true. For instance, in Z/4Z, we have [2][2] = [0].
2. We know that the ideals of Z are the subsets of the form nZ with n ∈ N. We claim that
nZ is prime if and only if n = 0 or n is prime.
If n = 0 then nZ = {0} thus to show that 0Z is prime we must show that if ab = 0 with
a, b ∈ Z then a = 0 or b = 0 but this is obvious.
If n = p is prime, let’s show that pZ is prime. Note that ab ∈ pZ is equivalent to saying
that p|ab. By euclid’s lemma this implies that p|a or p|b i.e. that a ∈ pZ or b ∈ pZ which
is what we wanted to show.
If n = k1 k2 with k1, k2 > 1 (i.e. if n is not prime) then k1 k2 ∈ nZ but k1 and k2 are not
in nZ thus nZ is not a prime ideal.
We claim that the maximal ideals of Z are the ideals of the form pZ with p a prime
number. Indeed, nZ is maximal if, and only if, for all m > 1 such that mZ contains nZ,
we have mZ = Z or mZ = nZ. Hence, nZ is maximal if, and only if, for all m > 1 such
that mZ contains nZ, m = 1 or m = n. Which means, nZ is maximal if, and only if, for
all m > 1 such that m divides n, m = 1 or m = n. Therefore, nZ is maximal if, and only
if, n is prime.
3. Let I be a maximal ideal of A and let x, y ∈ A such that xy ∈ A. Suppose by contradiction
that x and y are not in I. Then I + (x) is an ideal which contains I and is stricly bigger
than I thus I + (x) = A. Thus there exist i ∈ I and a ∈ A such that 1 = i + ax. Similarly
there exists j ∈ I and b ∈ A such that 1 = j + by.

Page 3
 Thus
1 = (i + ax)(j + by) = ij + iby + jax + abxy
which is in I since xy ∈ I. Thus I = A which contradicts the fact that I is maximal.
4. The maximal ideals of Z are clearly the ideals of the form pZ with p a prime number. The
ideal {0} is a prime ideal of Z which is clearly not maximal.

Exercise 6.
For a prime number p, we let
m
Zp := {x = , (m, n) ∈ Z × N∗ with p ∧ n = 1}.
n
1. Check that Zp is a subring of (Q, +, ×).
2. For k ∈ N, we let
nm o
∗ k
Jpk = , (m, n) ∈ Z × N with p | m and p ∧ n = 1
n
Check Jpk is an ideal of Zp .
3. Show that any ideal of Zp is of the previous form.

Solution:
0
0 0 0
1. If x = m m mm
n and x = n with p ∧ n =1 and p ∧ n = 1, then xx = nn0 with nn ∧ p = 1.
Similarly the sum of x and x may be written as a fraction of integers with nn0 as a
0

denominator.
2. Jpk is just the principal ideal generated by pk ∈ Zp .
3. Let n m o
k = max l > 0, ∀x ∈ I, ∃(m, n) ∈ Z × N∗ , x = , pl | m, p ∧ n = 1
n
Clearly we have I ⊂ Jpk . To prove the equality it is enough to show pk ∈ I. By defintion
of k there is 0 6= x = m k
n ∈ I with p ∧ n = 1, p | m but p
k+1
- m, i.e. m = pk q with
n k
q ∧ p = 1. In particular we have y = q ∈ Zp , therefore xy = p ∈ I as I is an ideal.

Exercise 7.
Let R be a ring. We say that an element x ∈ R is nilpotent if there exists an
integer n > 0 such that xn = 0.
1. Assume R is commutative. Show that the set Nil(A) of nilpotent elements
of R is an ideal.
2. Show that, if x ∈ R is nilpotent, then 1 − x is invertible.
3. Show that if x, y ∈ R are such that x · y is nilpotent, then y · x is nilpotent.

Page 4
Solution:
Pn+m
1. Let x,
y ∈ Nil(A). Suppose that xn = 0 and y m = 0. Then (x + y)n+m = k=0 =
n+m k n+m−k n+m−k
k x y . Note that if k 6 n then n + m − k > m thus y = 0. If k > n then
xk = 0. Thus for all k ∈ {0, . . . , n + m} we have xk y n+m−k = 0 thus (x + y)(n+m) = 0 and
thus x + y is nilpotent.
If x is nilpotent with xn = 0 and a ∈ A then (ax)n = an xn = 0 thus ax is nilpotent. We
see that indeed Nil(A) is an ideal.
2. Suppose xn = 0 then 1 = 1 − xn = (1 − x)(1 + x + · · · + x(n−1) ) thus 1 − x is invertible.
3. Suppose (xy)n = 0 then (yx)(n+1) = y(xy)n x = 0 thus yx is nilpotent.

Page 5