Scribd
Upload
50 views2 pages

Homework #8 Solutions

This document contains solutions to homework problems from a course on integral domains and polynomial rings. The problems cover topics like units, irreducibility, maximal and prime ideals, unique factorization domains, and Euclidean domains. Specific examples are worked through in domains like Z, Z[i], and polynomial rings over fields.

Uploaded by

Sarah Seunarine
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
50 views2 pages

Homework #8 Solutions

This document contains solutions to homework problems from a course on integral domains and polynomial rings. The problems cover topics like units, irreducibility, maximal and prime ideals, unique factorization domains, and Euclidean domains. Specific examples are worked through in domains like Z, Z[i], and polynomial rings over fields.

Uploaded by

Sarah Seunarine
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Homework #8 Solutions

p 333, #6 Let D be an integral domain and a, b, c ∈ D


(i) Reflexivity. Since a = 1a and 1 is a unit, a ∼ a.
(ii) Symmetry. If a ≡ b then a = ub for some unit u ∈ D. But then b = u−1 a so that b ∼ a,
since u−1 is also a unit in D.
(iii) Transitivity. If a ≡ b and b ≡ c then there exist units u, v ∈ D so that a = ub and
b = vc. But then a = ub = u(vc) = (uv)c. Since the set of units in D is closed under
multiplication, uv is also a unit and hence a ∼ c.

p 333, #8 Let u be a unit in D. Then d(1) ≤ d(1u) = d(u) and d(1) = d(uu−1 ) ≥ d(u)
so that d(u) = d(1). Now suppose that d(u) = d(1). Use the division algorithm to write
1 = qu + r for some q, r ∈ D with r = 0 or d(r) < d(u) = d(1). Since d(1) ≤ d(1r) = d(r),
the latter case cannot occur so we conclude that r = 0, i.e. 1 = qu for some q ∈ D. That is,
u is a unit in D.

p 333, #10 It should be pointed out that the problem is incorrectly stated in the text. One
must assume at the beginning that p is nonzero. We do so below.
Let p ∈ D be irreducible and let I ⊂ D be an ideal with hpiI. Since D is a PID, I = hai
for some a ∈ D. Then hpi ⊂ hai implies that p = ab for some b ∈ R. As p is irreducible,
either a is a unit, in which case I = hai = D, or b is a unit, in which case I = hai = hpi.
This proves that hpi is maximal.
Now suppose that hpi is maximal. Since a maximal ideals are always proper, p is not a
unit in D. Suppose that p = ab for some a, b ∈ D. Then p ∈ hai so that hpi ⊂ hai. The
maximality of hpi implies that hpi = hai or that hai = D. In the first case it follows that
a ∈ hpi so that a = kp for some k ∈ D. But then p = ab = (kp)b = p(kb) and cancelation in
D implies that kb = 1, i.e. b is a unit. In the second case, a is a unit since 1 ∈ hai implies
that 1 = ka for some k ∈ D. So, we have shown that if p = ab in D then either a or b is a
unit, and hence p is irreducible.

p 333, #12 Let I ⊂ D be a proper ideal. If I is maximal, there is nothing to show. So


suppose that I is not maximal. Then there is a proper ideal I2 6= I so that I ⊂ I2 . If I2 is
maximal we are finished. If not, then we may find a proper ideal I3 6= I2 so that I2 ⊂ I3 .
Continue to construct ideals in this way: if In is not maximal then choose a proper ideal
In+1 6= In so that In ⊂ In+1 . If none of the ideals In is ever maximal then we obtain an
infinite ascending chain of ideals I ⊂ I1 ⊂ I2 ⊂ I3 ⊂ in which every containment is proper.
However, we know that no such a chain can exist in a PID. It follows that at some point one
of the In will be maximal and since I ⊂ I1 ⊂ I2 ⊂ · · · ⊂ In , this finishes the proof.
p 334, #14 In Z[i] we have N (1 − i) = 1 + 1 = 2, which is prime. Therefore 1 − i is
irreducible.

√ √
p 334, #18 In Z[ 6], N (7) = 49. So 7 is not a unit √ and if 7 = xy in Z[ 6] for some
nonunits x and y, then N (x) = ±7. Writing x = a + b 6 for some a, b ∈ Z this would mean
that a2 − 6b2 = ±7. Going mod 7 we obtain a2 − 6b2 = 0 in Z7 or a2 = 6b2 . If b 6= 0 in
Z7 then this yields (a/b)2 = 6, which is impossible since 6 is not a square in Z7 . Therefore
a = b = 0 in Z7 , i.e. both a and b are divisible by 7. But then both a2 and b2 are divisible by
72 , which implies that 2 2
√ 49 divides a − 6b = ±7, an impossibility. This contradiction means
that if 7 = xy in Z[ 6] then x or y is a unit, i.e. 7 is irreducible.

√
p 334, #20 According to Example 1, Z[ −3] has √ irreducible elements that are not prime.
Since every irreducible
√ in a UFD is also prime, Z[ −3] is not a UFD. Since every PID is
also a UFD, Z[ −3] is not a PID, either.

√
p 334, #22 √ In Z[ 5] we have N (2) = 4, so 2 is not a unit. If 2 = xy with neither x nor y a
unit in Z[ 5] then √ it must be the case that N (x) = ±2. Then we would have integers a, b so
that ±2 = N (a + b 5) = a2 − 5b2√ , which implies that a2 (mod 5) = 2 or 3, neither of which
is possible.
√ Hence, if 2 = xy in Z[ 5] then√ x or y √ is a unit, which means that√ 2 is irreducible
√
in Z[ 5]. Notice that 2 · 2√= 4 = (1 + √5)(−1 + 5), so that 2 divides (1 + √5)(−1 + 5),
but 2 divides neither√ 1 + 5 nor −1√+ 5, proving that 2 is not prime in Z[ 5].
Similarly,
√ in Z[ 5]√we have N (1+ 5) = −4, which proves that −4 is not a unit. Moreover,
if 1 + 5 = xy in Z[ 5] with neither x nor y a unit then, √ as above, N (x) = ±2, which we
have already argued
√ is impossible.
√ It follows that
√ 1 + 5 is irreducible. Again √ noting that
√
2 · 2 = 4 = (1 +√ 5)(−1 + √ 5), we see that 1 + 5 divides
√ 2 · 2. But for any a + b 5 ∈ Z[ 5]
we have (a + b 5)(1 + 5) = (a + 5b) + (a + b) 5, which can never equal √ two since the
system a + 5b = 2, a + b = 0 has no solution √ in integers. Therefore, 1 + 5 does not divide
2, showing that the former is not prime in Z[ 5].

p 334, #28 We know that x + iy ∈ Z[i] is a unit if and only if 1 = N (x + iy) = x2 + y 2 .


Since x and y are both integers this can only occur if (x2 , y 2 ) = (1, 0) or (x2 , y 2 ) = (0, 1),
which means that x + iy is one of the four elements ±1, ±i.

p 334, #30 This is not a contradiction because the irreducible factors in question are
associates, which is all we are guaranteed in a UFD. In particular we have 3(3x + 2) =
9x + 6 = 4x + 1 and 2(x + 4) = 2x + 8 = 2x + 3 over Z5 , and both 3 and 2 are units in Z5 .

p 335, #34 A subdomain of a Euclidean domain need not be Euclidean. For example, the
ring Z[x] is not a PID and therefore is not Euclidean, however it is a subdomain of Q[x]
which we know to be a Euclidean domain.

You might also like