2215 Solutions Problem Sheet 11

Ring, Subrings, Ideals, Intregal Domains, Ring Homomorphisms, First

Isomorphism Theorem.
1. Let R, S be rings. Verify that the direct product, that is,
R S = {(a, b) : a R, b S}
with operations
(a, b) + (c, d) = (a + c, b + d),
(a, b)(c, d) = (ac, bd)
is a ring. Show that R S is commutative if and only if R and S are commutative
and R S has an identity if and only if R and S have an identity.
Solution:
First, we know that R S is an abelian group under +.
Clearly, R S is closed under ., and, since R and S both satisfy the associative law
under ., we have
(a, b)((c, d)(e, f )) = (a, b)(ce, df ) = (a(ce), b(df )) = ((ac)e, (bd)f )
= (ac, bd)(e, f ) = ((a, b)(c, d))(e, f ) for all (a, b), (c, d), (e, f ) R S.
To complete the argument that R S is a ring, we need to show that the distributive
laws hold. For any (a, b), (c, d), (e, f ) R S,
(a, b)((c, d) + (e, f )) = (a, b)(c + e, d + f ) = (a(c + e), b(d + f ))
= (ac + ae, bd + bf ) = (ac, bd) + (ae, bf ) = (a, b)(c, d) + (a, b)(e, f ),
using the fact that R and S satisfy the distributive laws.
If R and S are commutative, then for any (a, b), (c, d) R S we have that
(a, b)(c, d) = (ac, bd) = (ca, db) = (c, d)(a, b),
so that R S is commutative. If R and S have identities 1R and 1S respectively,
then for any (a, b) R S,
(1R , 1S )(a, b) = (1R a, 1S b) = (a, b) = (a1R , b1S ) = (a, b)(1R , 1S ),
so that (1R , 1S ) is the identity of R S.
1

Esther Vergara Diaz, [email protected], see also http://www.maths.tcd.ie/~evd

Conversely, suppose that R S is commutative. Then for any a, b R we have that

(a, 0S )(b, 0S ) = (b, 0S )(a, 0S ),
which gives that (ab, 0S ) = (ba, 0S ) and ab = ba. Similarly, S is commutative.
Finally, suppose that R S has identity, say (u, v). Then for any a R we have that
(u, v)(a, 0S ) = (a, 0S ) = (a, 0S )(u, v),
whence
(ua, 0S ) = (a, 0S ) = (au, 0S ),
giving that ua = a = au so that u is the identity of R. Similarly, S has identity v.
2. Let R be a ring with identity, and let I be an ideal of R. Show that R = I if and
only if 1 I.
Solution:
If R = I then clearly 1 I.
Conversely, suppose that 1 I. Then for any r R,
r = 1.r I
since I is an ideal. Hence R = I.
3. F Let I, J be ideals of R. Show that I + J is an ideal, where
I + J = {x + y : x I, y J}.
Hence explain why if M is an ideal and a R (fixed) with R commutative ring, then
{m + ar : r R}
is an ideal. Prove that I J is an ideal if and only if I J or J I.
Solution:
We have that 0 I and 0 J, so that 0 = 0 + 0 I + J. Suppose now that
x, y I + J and r R. By definition of I + J we have that x = a + b, y = a0 + b0
where a, a0 I, b, b0 J. Then
x + y = (a + b) + (a0 + b0 ) = (a + a0 ) + (b + b0 ) I + J,
since a + a0 I and b + b0 J. Similarly,
x = (a + b) = (a) + (b) I + J
and
xr = (a + b)r = ar + br I + J,
2

so that I + J is an ideal as required.

If M is an ideal, then let be s S = {m + ar : r R}, then there s = m + ar
for some r R then r R we have r(m + ar) = rm + rar and using that R is
commutative we get rar = a
rr (and rr R) so we obtain that s S. To prove that
(m + ar)
r S is easier since m
r M and ar
r aR which is therefore an ideal.
If I J (J I) then I J = J (I J = I) so that in either case, I J is an ideal.
Conversely, suppose that I J is an ideal and I 6 J. Pick a I \ J. Let b be any
element of J. Then a, b I J, which is an ideal, so that a + b I J. Hence
a + b I or a + b J. If a + b = c J, then a = c b J, a contradiction. Thus
a + b = c I so that b = c a I. Consequently, J I.
4. Let R be a ring, and let I, J be ideals of R. Prove that I J is an ideal of R.
Solution:
0 I and 0 J since I, J are ideals, so 0 I J.
If a, b I J, then a, b I and a, b J, so a b I and a b J (I, J ideals), so
a b I J.
If a I J and r R, then a I and a J, so ar I and ar J (I, J ideals), so
ar I J.
Hence I J is an ideal of R.

5. F Let D Z. Show that Z[ D] is an integral domain where

Z[ D] = {a + b D : a, b Z}.
(We usually assume that D is square-free, that is, D 6= 1 and D is not divisible by
the square
of any prime. This ensures the uniqueness of a and b in the expression

a + b D.)
Solution: To show that it is subring we need to see that the set is not empty
(actually
it contains
0) andit is closed under addition and multiplication. Suppose

a + D, c + d D are in Z[ D]. Then:

(a + b D) + (c + d D) = a + c + (b + d) D Z[ D]

(a + b D)(c + d D) = ac + bdD + (ad + bc) D Z[ D],

using the distributive laws and since x, y Z then x + y, xy Z.

Need
to prove there are not zero divisors. Suppose a, b 6= 0 and consider (a+b
D)(c+
d D) = 0 then either c 6= 0 or d 6= 0 (if we want to find zero divisors c + d D =6= 0
by definition). Then

ac + bdD + (ad + bc) D = 0 ac + bdD = (ad + bc) D

3

taking squares in the latest equation:

(ac + bdD)2 = (ad + bc)2 D
and therefore if ad + bc 6= 0 then
D=

(ac + bdD)2
(ac + bdD)2

contradicting the fact that D is square free.

Now if ad + bc = 0 then ad = bc and from ac + bdD + (ad + bc) D = 0 we

must have as well that ac + bdD = 0. If we multiply by d this lattest equation
we get adc + bd2 D = 0 and the replacing ad by bc we have bc2 + bd2 D = 0 or
b(c2 D d2 ) = 0.
2
If b 6= 0 then c2 D = d2 and if c 6= 0 we obtain D = d2 contradicting the fact that D is
c
square free. If c = 0 then d2 = 0 which implies c + d D = 0 which is a contradiction
since we were supposing either c or d were different to zero.

Now if b = 0 (this implies that a 6= 0 otherwise a + b D = 0) then (ac + bdD)2 =

(ad + bc)2 D becomes a2 c2 = a2 d2 D or c2 = d2 D which implies that either c = d = 0
or D is not square free.

So we conclude there cannot be any zero divisors.

The unity
of this ring is 1 Z and the zero of the ring is 0 Z. And furthermore
since Z[ D] R, it follows that it is a commutative ring.

6. Explain why the ring of Q is not isomorphic to Q[ 2].

Solution: If f : Q Q[ 2] was a ring isomorphism then f (1) = 1 and therefore

f (2) =
f (1 + 1) = f (1) + f (1) = 2. Furthermore if f is an isomorphism so it is
f 1 : Q[ 2] Q and f 1 (2) = 2 ().
Since f 1 is a ring isomorphism we have:

f 1 (2) = f 1 ( 2 2) = f 1 ( 2)f 1 ( 2) = (f 1 ( 2))2

Now using ()

2 = f 1 (2) = (f 1 ( 2))2

but f 1 ( 2) Q, say f 1 ( 2) = y. And therefore y 2 Q. Hence we have found a

rational number y such that y 2 = 2, and we know that it is impossible in Q to find
such number.
7. F Show that Q[i] = {a + ib : a, b Q} is a subring of C and Q[i] is field.
Solution: It is clearly not empty since it contains 1. It is closed under the addition
and multiplication because if a + ib, c + id Q[i] then:
a + ib + c + id = a + c + i(c + d) Q[i]
4

and
(a + ib)(c + id) = ac bd + i(ad + cb) Q[i],
since x, y Q then x + y, xy Q.
The negative of a + ib is ia ib.
This proves that Q[i] is a subring of the ring C.
Furthermore, because C is a commutative ring, so it is Q[i].
To prove that it is a field, we need to see that there is a unity, it is an integral domain
and every number has an inverse.
The unity is 1 Q and the zero of the ring is 0 Q.
To prove that there are no zero divisors. Suppose a + ib 6= 0 and (a + ib)(c + id) = 0
with either c 6= 0 or d 6= 0. Then
(a + ib)(c + id) = ac bd + i(ad + cb) = 0
then the real part of ac bd + i(ad + cb) and the imaginary part of ac bd + i(ad + cb)
must be zero:
ac bd = 0 and ad + cb = 0
or
ac = bd
and
ad = cd
multiplying the second by c we obtain acd = c2 b and using the first equation we
know that we can replace ac for bd in acd = c2 d obtaining
bd2 = c2 b
or
b(c2 + d2 ) = 0.
Now suppose that b 6= 0 then c = d = 0 and that means that c + id = 0 is not a zero
divisor.
If b = 0 then a + ib = a 6= 0 and a(c + id) = 0 or ac + iad = 0. If the real part ac
and the imaginary part ad must be zero. Which means that c = d = 0 so we obtain
again that c + id is not a zero divisor.
The inverse of a number a + ib 6= 0 is
(a + ib)1 =

a ib
a
b
1
= + 2 = + 2 + (i + 2 ) Q[i]
a + ib
a b
a b
a b

8. Find the units of Z[i]. (Hint: If z Z[i] is a unit, then z 1 must be in Z[i]. What
can you deduce about |z|?)
Solution:
If a + bi Z[i], then a2 + b2 = |a + ib|2 Z. If z is a unit, there exist w in Z[i] such
that zw = 1. We know that it does happens because Z[i] is a subring of C we need
to have w = z 1 . This implies that |z 1 |2 Z.
Now 1 = |zz 1 | = |z||z 1 |, so |z|12 = |z 1 |2 is a positive integer. Hence |z|2 = 1, that is
a2 + b2 = 1. Since a, b Z, the only possibilities are a = 1, b = 0 and a = 0, b = 1,
that is, the only possible units are 1, i, 1, i. Clearly, these numbers all are units
in Z[i], so U (Z[i]) = {1, i, 1, i}.
9. Find the units and the zero-divisors of Z15 .
Solution: Let x Z15 . We know that x is a unit if and only if hcf(x, 15) = 1 and x
is a zero-divisor if and only if hcf(x, 15) > 1. Hence
U (Z15 ) = {1, 2, 4, 7, 8, 11, 13, 14}
and the zero-divisors are
{3, 5, 6, 9, 10, 12}.
10. An ideal P of a ring R is prime if
ab P a P or b P.
Show that an ideal I = nZ of Z (n 2) is prime if and only if n is prime.
Solution:
Let I = nZ be an ideal of Z, n 2.
Suppose that n = ab where 1 < a < n and 1 < b < n then ab I but neither a nor b
lies in I (since x I n|x). Thus I in this case is not prime.
Conversely, if n = p is prime, then
ab I p|ab p|a or p|b a I or b I,
so that in this case I is prime.
11. F Let P be an ideal of a ring R. Show that P is a prime ideal if and only if the
factor ring R/P has no zero divisors.
Solution: Suppose that P is a prime ideal. Let a + P, b + P R/P and suppose
that
(a + P )(b + P ) = 0 + P,
that is,
ab + P = P
6

by definition of the multiplication of classes and so ab P . But P is prime, so that

either a P or b P , whence a + P = P or b + P = P . Thus R/P has no zero
divisors.
Conversely, suppose that R/P has no zero divisors. Let a, b R be such that ab P .
In the factor ring R/P we have
(a + P )(b + P ) = ab + P = P.
But R/P has no zero divisors, so that a + P = P or b + P = P . Thus either a P
or b P and P is prime.
12. Let I, J be ideals of a ring R. Explain why I J is an ideal of R.
13. Show that for a, b Z ,
aZ bZ = lcm(a, b)Z,
where lcm(a, b) is the least common multiple of a and b. Write down 10Z 25Z and
10Z + 25Z.
Solution:
Let c aZ bZ. Then c aZ and c bZ. Hence c = au = bv for some u, v Z, so
that c is a common multiple of a and b. Consequently c is a multiple of lcm(a, b) = m,
so that c mZ and aZ bZ mZ.
Conversely, let c mZ, so that c = mt say. Write m as m = ax = by, which is
possible as m = lcm(a, b). Then c = axt aZ and c = byt bZ so that c aZ bZ
and mZ aZ bZ. This completes the proof that mZ = aZ bZ.
With a = 10 and b = 25 we have that
10Z + 25Z = hcf(10, 25)Z = 5Z, 10Z 25Z = lcm(10, 25)Z = 50Z.
14. Let D be a finite integral domain. Prove that D is a field.
[Hint: Let D = D \ {0} be the set of nonzero elements so that the identity is a
member of D . Let
D = {a1 , . . . , an }.
For any ai , consider ai D = {ai a1 , . . . , ai an }. Explain why ai D D and |ai D | = n.
Now deduce that ai is a unit.]
Solution:
Let D be a finite integral domain, and say D = {a1 , . . . , an } (one of the ai s will be
1). For ai D , consider ai D = {ai a1 , . . . , ai an }. Each element of ai D is a product
of non-zero elements of D, and so, since D is an integral domain, every element of
ai D is non-zero. Hence ai D D .
If ai aj = ai ak , then since D is an integral domain and ai is non-zero, aj = ak (use the
previous question). Hence the elements in the list ai a1 , . . . , ai an are distinct, that is,
7

|ai D | = n. Thus ai D = D , and since 1 D , we have 1 = ai aj for some j, that

is, ai is a unit of D. Hence, every non-zero element of D is a unit, so D is a field.
15. Prove that if R is a commutative ring and a, b R and n is positive then (a + b)n
can be computed by the binomial theorem.
16. Let D be an integral domain with characteristic D = p. Show that for all a, b D,
(a + b)p = ap + bp (you may assume the binomial theorem holds for D). Now show
that : D D given by (a) = ap is an endomorphism of D. (i.e. a homomorphism
from D to D).
Solution:
Let a, b D. Using the binomial theorem we have
 
i=p p
p
(a + b) = i=0
ai bpi .
i
We have

 
p(p 1) . . . (p i + 1)
p
=
=w
i
i!

where w N. Hence p | i! w, so that if 0 < i < p, p | w. As ch D = p we deduce that

(a + b)p = ap + bp .
For any a, b D, bearing in mind that D is commutative,
(ab) = (ab)p = ap bp = (a)(b)
and
(a + b) = (a + b)p = ap + bp = (a) + (b),
so that is an endomorphism of D.
.
17. Let R be a commutative ring with 1. For any a R , we say that a divides b R,
written a | b, if b = ac for some c R. Show that aR bR iff a = b = 0 or b | a.
Let D be an integral domain, and let a, b D be non-zero elements. Prove that
aD = bD if and only if a = bu for some unit u of D.
Solution:
If a = b = 0, clearly aR = bR = {0}. If b | a, then a = bc for some c R. Hence
a bR and as bR is an ideal, aR bR.
Conversely, if aR bR, then either b = 0, in which case a = b = 0, or b 6= 0. Then
as a aR we have a bR so that a = bc for some c R and b | a as required.
If aD = bD, then a = a1 aD, so a bD and so a = bu for some u D. Similarly,
b = av for some v D and so a = bu = (av)u = a(vu). Now D is an integral domain
8

and a 6= 0, so we can cancel a to get 1 = vu. Hence v and u are units, and we have
a = bu for some unit u of D.
Conversely, if a = bu for some unit u of D, then b | a, so aD bD. Since u is a unit,
u1 D and b = au1 , so a | b and hence bD aD.
Thus aD = bD.
18. Use the Fundamental Theorem of Homomorphisms for Rings to show that for n
N, n 2, Z/nZ
= Zn .
Solution:
Let : Z Zn be defined by (z) = [z]. Then for all z, t Z,
(z + t) = [z + t] = [z] [t] = (z) (t)
and similarly for . Clearly (1) = [1], so that is a ring homomorphism. Now r
Ker if and only if
(r) = [r] = [0],
so that Ker = nZ. By the Fundamental Theorem of Homs for rings, Z/nZ
= Zn .
19. Let T be a set and let RT be the set of all functions from T to R. Then RT is a
commutative ring, where f + g and f g are defined by:
(f + g)(t) = f (t) + g(t), (f g)(t) = f (t)g(t) for all t T.
For each a R, let ca RT be the constant function on a, i.e. ca (t) = a for all
t T . Show that c0 and c1 are the zero and identity of RT , respectively.
Let r be a fixed element of T and let S be any subring of RT . Define r : S R
by r (f ) = f (r). Verify that r is an onto homomorphism. Deduce that in the case
where R is a field,
I = {f R : f (r) = 0}
is a maximal ideal of R.
r is called an evaluation map.
Solution:
For any f RT we have
(c0 + f )(t) = c0 (t) + f (t) = 0 + f (t) = f (t),
and
(c1 f )(t) = c1 (t)f (t) = 1 f (t) = f (t),
for any t T . Hence c0 + f = f and c1 f = f , so that c0 is the zero and c1 the
identity of RT .
9

Let f, g S. We have
r (f + g) = (f + g)(r) = f (r) + g(r) = r (f ) + r (g),
and similarly
r (f g) = r (f )r (g).
Since also r (c1 ) = c1 (r) = 1, we have that r is a ring homomorphism. Now r is
onto, as for any a R,
r (ca ) = ca (r) = a.
For any f S, we have that f Ker r exactly when r (f ) = 0, i.e. f (r) = 0. Hence
I = Ker r and so I is an ideal of S. Note that I 6= S, since c1
/ Ker r = I. Further,
by the Fundamental Theorem of Homs., S/I
R.
If
R
is
a
field, so must be S/I,
=
hence I is a maximal ideal of S as require.

10