Algebra 557: Week 1

Throughout this course by a ring A we will understand a commutative ring

with 1.

1 Rings and Modules.

1.1 Maximal and Prime Ideals, Chinese Remainder The-

orem.
We recall the definitions of ideals, ring homomorphisms, unit (invertible) ele-
ments, proper ideals, maximal ideals and prime ideals.

Theorem 1. If I is a proper ideal of a ring A then there exists at least one max-
imal ideal containing I.

Definition 2. An ideal P ⊂ A for which the quotient A/P is an integral domain

is called a prime ideal. Equivalently, an ideal P is prime if it satisfies
• P  A,
• x, y  P ⇒ xy  P.

An easy consequence of the above definition is

Theorem 3. If I , J are ideals of A and P a prime ideal, then I⊂ P , J ⊂P ⇒

IJ ⊂ P.

Definition 4. A subset S ⊂ A is called a multiplicative subset if it satisfies

• x, y ∈ S ⇒ x y ∈ S, and
• 1 ∈ S.

Theorem 5. Let S be a multiplicative subset and I ⊂ A an ideal disjoint from S;

then there exists a prime ideal P ⊃ I which is disjoint from S.

Proof. By Zorn’s lemma there exists an ideal P which is maximal amongst the
ideals containing I and disjoint from S. We claim that P is prime. Suppose that
x, y ∈P . Then, P + x A, P + y A both meet S, and hence their product (P +
x A)(P + yA) also meets S since S is multiplicative. But (P + xA)(P + y A) =
P + xyA, which implies that xy ∈ P , proving that P is prime. 
√
Definition 6. The radical of an ideal I (written I ) is defined by
√
I = {a ∈ A |an ∈ I for some n > 0}.
√ \
Theorem 7. I = P.
P ⊃I ,P prime

1
2 Section 1

n
Proof. If P is a prime ideal
√ √ containing I , and a ∈ I ⊂ P , then a ∈2 P .3 Thus,
I ⊂ P . Conversely, if x ∈ I , then the multiplicative set Sx = {1, x, x , x ,  } is
disjoint from I. By previous theorem there exists a prime √ ideal P containing I
T
but disjoint from S. In particular, x ∈P , implying that I = P ⊃I ,P prime P . 
p
Definition 8. (0) is called the nilradical of A (denoted nil(A)). Clearly,
\
nil(A) = P.
P prime ideal of A

If nil(A) = 0 we call the ring A to be reduced. Otherwise, we denote by Ared the

reduced ring A/nil(A).

Definition 9. A ring A having only one maximal ideal m is called a local ring.
The field k = A/m is the called the residue field (and we denote such a local ring
by the triple (A, m, k). A ring having a finite number of maximal ideals will be
called semi-local.

Definition 10. The Jacobson radical of A (denoted rad(A)) is defined as the

ideal of A which is the intersection of all its maximal ideals.

The Jacobson radical can be characterized by the following property.

Theorem 11. An element x ∈ rad(A) if and only if 1 + ax is a unit for each a ∈

A.

Proof. Suppose x ∈ rad(A) and a ∈ A. Let m be the maximal ideal containing 1 +

ax. Then, x ∈ m. But then 1 ∈ m. Hence, (1 + ax) = (1).
Conversely, suppose that 1 + ax is invertible for each a ∈ A, and let m be a
maximal ideal. Then if x ∈m, we will have 1 ∈ m + (x), and thus 1 = m − ax, for
some m ∈ m, and a ∈ A. This would imply that m is a unit. 

1.2 Chinese Remaindering

In general for any two ideals I , J , the product I J is contained in I ∩ J but not
necessarily equal to it. Equality holds if I + J = A (in which case we say I , J are
co-prime).

Lemma 12. If I + J = 1 then IJ = I ∩ J .

Proof. In this case I ∩ J = (I ∩ J )(I + J) ⊂ IJ ⊂ I ∩ J. 

Lemma 13. If I , J are coprime and I , K are co-prime then I is co-prime to JK.

Proof. (1) = (I + J)(I + K) = I + JK . 

By induction we obtain

Theorem 14. If I1, I2,  , In are co-prime in pairs then

I1I2 In = I1 ∩ I2 ∩  ∩ In.
Rings and Modules. 3

Remark 15. In particular, if A is a semi-local ring and m1,  , mn are its max-
imal ideals then
rad(A) = m1 ∩  ∩ mn = m1 mn.

We also have

Theorem 16. (Chinese Remainder) If I1, I2,  , In are co-prime in pairs then
A/I1 In @ A/I1 ×  × A/In.

1.3 Modules.
1.3.1 Basic Definitions.

Definition 17. An A-module M is an abelian group under addition, along with a

left-multiplication by elements of A satisfying:
• a(x + y) = ax + by
• (ab)x = a(bx)
• (a + b)x = ax + bx
• 1x = x.

Definition 18. Let N , N ′ be sub-modules of M. Then we denote by

(N : N ′)A = {a ∈ A| aN ′ ⊂ N }
(note that (N : N ′)A is an ideal of A).
Similarly, given an ideal I ⊂ A, we will denote
(N: I)M = {m ∈ M | Im ⊂ N }
which is a sub-module of M (called the colon-quotient).
Finally, the annhilator of a module M (denoted by ann(M )) is the ideal (0:
M )A .
We call a module M faithful if ann(M ) = 0.

1.3.2 Determinant trick, Nakayama lemma and applications.

Theorem 19. Let M be a finitely generated A-module generated by n elements

and let φ ∈ HomA(M , M ). Let I be an ideal of A with φ(M ) ⊂ I M . Then there
exists a relation of the form
φn + a1 φn−1 +  + an−1 φ + an = 0,
with ai ∈ I i , 1 ≤ i ≤ n (both sides considered as elements of HomA(M , M )).

Proof. Cramer’s rule. 

Theorem 20. (Nakayama’s Lemma) Let M be a finitely generated A-module and

I ⊂ A an ideal of A, such that M = I M . Then, there exists a ∈ A, with a ≡
1 mod I, such that aM = 0. In particular, if I ⊂ rad(A), then M = 0.
4 Section 1

Proof. Take in the previous theorem φ = IdM . Then the obtained relation has
the form
1 + a1 +  + an = 0
and choose for a = 1 + a1 +  + an. 

Theorem 21. Let M be an A-module and N a sub-module such that M /N is

finitely generated. Let I ⊂ rad(A) be an ideal such that M = N + IM. Then, M =
N.

Proof. Let M ′ = M /N . Then IM ′ = M ′, and by previous theorem M ′ = 0. 

Theorem 22. If A is a local ring and M a f.g. A-module, then every minimal
generating set of M has the same number of elements.

Proof. Choose a basis u¯1,  , ūn ∈ M /mM of the k-vector space M /mM , and lift
vector ūi to an element ui ∈ M . We claim that u1,  , un generate M . Let
each P
N = 1 ≤i≤n Aui. Then
M = N + mM .
Applying Nakayama (Theorem 20), we have that M = N .
Conversely, any minimal generating set of M is of the above form. If u1,  ,
un ∈ M is a minimal generating set, then clearly ū1,  , ūn ∈ M /mM span M /mM .
We claim that if fact they form a basis. Indeed if any proper subset of these vec-
tors spanned, then by the above argument their inverse images would generate M
contradicting the minimality of the generating set.


Theorem 23. Let M be a f.g. A module and f ∈ HomA(M , M ) a surjective

homomorphism. Then f is injective as well.

Proof. Consider the A[X]-module structure on M given by X m = f (m) for all

m ∈ M . Then, (X) M = M , and hence by Nakayama (Theorem 20) there exists
F ∈ A[X] such that (1 + XF )M = 0. If u ∈ ker f , then we have that
0 = (1 + XF )u = u + F (X(u)) = u + 0 = u.