Math 446 - Homework # 3

1. Prove the following:

(a) Given a, b ∈ Z with b 6= 0, there exist x, y ∈ Z with gcd(x, y) = 1
a x
and = .
b y
Solution: Let d = gcd(a, b). Let x = a/d and y = b/d. Then
from class, we know that gcd(x, y) = 1. And we also have that
a/b = (a/d)/(b/d) = x/y.
(b) If p is a prime and a is a positive integer and p|an , then pn |an .
Solution: Suppose that p is a prime and p divides an = a · a · · · a.
Recall that when a prime divides a product of integers then it
must divide at least one of the integers contained in the product.
Hence p|a. Therefore, pk = a for some integer k. Hence, an =
(pk)n = pn k n . Therefore pn |an .
√
(c) 5 5 is irrational.
√ √
Solution: Suppose that 5 5 is rational. Then 5 5 = a/b where
a, b ∈ Z. We may always cancel common divisors in a fraction,
hence we may assume that gcd(a, b) = 1.
√
Taking the fifth power of both sides of 5 5 = a/b gives 5 = a5 /b5 .
Hence a5 = 5b5 . Therefore 5 divides the product a5 = a · a · a · a · a.
Recall that when a prime divides a product of integers then it must
divide at least one of the integers contained in the product. Since
5 is prime we must have that 5 divides a. Therefore a = 5k where
k is an integer. Substituting this expression into a5 = 5b5 yields
55 k 5 = 5b5 . Hence 5(53 k 5 ) = b5 . Therefore 5 divides b5 . Since 5
is prime we must have that 5|b. But then 5 would be a common
divisor of a and b and hence gcd(a, b) ≥ 5. This contradicts our
assumption that gcd(a, b) = 1.
√
Therefore 5 5 is irrational.
√
(d) If p is a prime, then p is irrational.
√ √
Solution: Suppose that p is rational. Then p = a/b where
a, b ∈ Z. We may always cancel common divisors in a fraction,
hence we may assume that gcd(a, b) = 1.
√
Squaring both sides of p = a/b and then multiplying through by
b2 gives us that pb2 = a2 . Hence p|a2 . Recall that when a prime
 divides a product of integers then it must divide at least one of
the integers in the product. Since p is a prime, p must divide a.
Therefore, a = pk for some integer k. Substituting this back into
pb2 = a2 gives us that pb2 = p2 k 2 . Dividing by p gives us b2 = pk 2 .
Thus p|b2 . Again, since p is a prime, we must have that p|b.
From the above arguments we see that p|a and p|b. Hence gcd(a, b) ≥
p. However, we also have that gcd(a, b) = 1. This gives us a con-
tradiction.

2. (a) Suppose that a, b, c are integers with a 6= 0 and b 6= 0. If a|c, b|c,

and gcd(a, b) = 1, then ab|c.
Solution 1: Since a|c and b|c we have that c = at and c = br
where r, t ∈ Z. Therefore at = br. Thus a|br. Since gcd(a, b) = 1
and a|br we have that a|r. Thus r = ak where k ∈ Z. Thus,
c = br = bak = (ab)k. Hence ab|c.
Solution 2: Since a|c and b|c we have that c = at and c = br
where r, t ∈ Z. Since gcd(a, b) = 1, there exist integers x and y
with ax + by = 1. Multiplying this by c we get that acx + bcy = c.
Now substitute c = br into the first term and c = at into the
second term to get that c = acx+bcy = abrx+baty = (ab)(rx+ty).
Therefore ab|c.
√
(b) Prove that 6 is irrational.
√
Solution: Suppose that 6 was rational.
√ We show that this leads
to a contradiction. We may write 6 = x/y where x and y are
integers with y 6= 0 and gcd(x, y) = 1. Squaring this equation and
cross-multiplying we get that 6y 2 = x2 or 2 · 3 · y 2 = x2 . Therefore,
2 divides x2 = x · x. Since 2 is prime we must have that 2 divides
x. Similarly, 3 divides x2 = x · x. And since 3 is prime we must
have that 3 divides x. Since 2|x and 3|x and gcd(2, 3) = 1, by the
first part of this problem, we have that 6 = 2 · 3 must divide x. So
x = 6u where u is a non-zero integer. Subbing this into 6y 2 = x2
gives us that 6y 2 = 62 u2 . Thus y 2 = 6u2 . Following the same
reasoning as above, this forces that 6 must divide y. Therefore,
6 is a common divisor of x and y which contradicts the fact that
gcd(x, y) = 1.

3. Prove that log10 (2) is an irrational number.

 Solution: Suppose that log10 (2) was rational. Then log10 (2) = a/b
where a and b are positive integers (we may assume they are positive
since log10 (2) is positive). In particular, b 6= 0. We have that 10a/b = 2
by the definition of the logarithm. Hence 10a = 2b . Therefore 2a 5a = 2b .
Since prime factorizations are unique (by the fundamental theorem of
arithmetic) we must have that a = 0 since there are no factors of 5
on the right-hand side of 2a 5a = 2b . Hence 20 50 = 2b . This gives
2b = 1. But this implies that b = 0 which is not true. Hence log10 (2) is
irrational.

4. We say that an integer n ≥ 2 is a perfect square if n = m2 for some

integer m ≥ 2. Prove that n is a perfect square if and only if the prime
factorization of n = pk11 pk22 · · · pkr r has even exponents (that is, all the ki
are even).
Solution: Suppose that n is a perfect square. Therefore n = m2 where
m is a positive integer. By the fundamental theorem of arithmetic
m = q1e1 q2e2 · · · qrer where qi are primes and ej are positive integers. We
see that
n = m2 = (q1e1 q2e2 · · · qrer )2 = q12e1 q22e2 · · · qr2er .
Therefore every prime in the prime factorization of n is raised to an
even exponent.

Conversely suppose that every prime in the prime factorization of n

is raised to an even exponent. Then n = p2k 1 2k2 2kr
1 p2 · · · pr where pi are
k1 k2 kr
primes and kj are positive integers. Let m = p1 p2 · · · pr . Then m is
an integer and n = m2 . Hence n is a perfect square.

5. (a) Let a and b be positive integers. Prove that gcd(a, b) > 1 if and
only if there is a prime p satisfying p|a and p|b.
Solution:
Suppose that d = gcd(a, b) > 1. Since d is positive integer with
d ≥ 2, by the fundamental theorem of arithmetic, there is at least
one prime p with p|d. Since p|d and d|a we must have that p|a.
Since p|d and d|b we must have that p|b. Hence p|a and p|b.

Conversely suppose that there is a prime p with p|a and p|b. Then
gcd(a, b) ≥ p > 1.
 (b) Let a, b, and n be positive integers. Prove that if gcd(a, b) > 1 if
and only if gcd(an , bn ) > 1.
Solution: Suppose that d = gcd(a, b) > 1. So a = dk and b = dm
where k and m are integers. Thus an = dn k n and bn = dn mn . So
d|an and d|bn . Hence gcd(an , bn ) ≥ d > 1.

Conversely, suppose that gcd(an , bn ) > 1. Then by exercise (5a),

there exists a prime q with q|an and q|bn . Since q divides the
product an = a · a · · · a and q is prime, we must have that q|a.
Since q divides the product bn = b · b · · · b and q is prime, we must
have that q|b. Hence q|a and q|b. Thus gcd(a, b) ≥ q > 1.

6. Suppose that x and y are positive integers where 4|xy but 4 - x. Prove
that 2|y.
Solution: Since 4|xy we have that 4s = xy for some integer s. Hence
2(2s) = xy. Thus 2|xy. Since 2 is prime we have that either 2|x or 2|y.
We break this into cases.
case 1: If 2|y then we are done.
case 2: Suppose that 2|x. Then x = 2k where k is some integer. Since
4 - x we must have that k is odd. Hence 2 - k. Substituting x = 2k
into 4s = xy gives 4s = 2ky. Hence 2s = ky. Therefore 2|ky. Since 2
is prime we must have either 2|k or 2|y. But 2 - k. Therefore, 2|y.

7. Let a and b be positive integers. Suppose that 5 occurs in the prime

factorization of a exactly four times and 5 occurs in the prime factor-
ization of b exactly two times. How many times does 5 occur in the
prime factorization of a + b?
Solution: By assumption a = 54 s and b = 52 t where s and t are
positive integers and 5 - s and 5 - t. Note that a + b = 52 (25s + t). We
want to show that 5 does not divide 25s + t. If 5 did divide 25s + t then
5k = 25s + t for some integer k. This would imply that 5(k − 5s) = t,
which gives that 5 divides t. But we know that is not true.
Therefore a + b = 52 (25s + t) where 5 does not divide 25s + t. Hence
5 occurs twice in the prime factorization of a + b.

8. A positive integer n ≥ 2 is called squarefree if it is not divisible by

any perfect square. For example, 12 is not squarefree because 4 = 22
is a perfect square and 4|12. However, 10 is squarefree. (Recall the
definition of perfect square from problem 4.

(a) Prove that a positive integer n ≥ 2 is squarefree if and only if n

can be written as the product of distinct primes.
Solution: Suppose that n is squarefree. Let n = pe11 pe22 · · · pess be
the prime factorization of n where the pi are distinct. Here we
have that the ei are positive integers. Suppose that e1 ≥ 2. Then
n = p21 (pe11 −2 pe22 · · · pess ). This would imply that n was divisible by
the perfect square p21 . This can’t happen since n is squarefree.
Hence e1 = 1. A similar argument shows that ei = 1 for all i.
Thus n = p1 p2 · · · ps is the product of distinct primes.
Conversely suppose that n is the product of distinct primes. By
way of contradiction, suppose that n was divisible by a perfect
square. Then n = m2 k where m ≥ 2 and k ≥ 1 are integers. Let
m = q1f1 q2f2 · · · qtft be the prime factorization of m where the qi are
primes and the fi are positive integers. Then

n = m2 k = q12f1 q22f2 · · · qt2ft k.

This contradicts the fact that n is the product of distinct primes

since, for example, q1 appears more than once in the factorization
for n. Therefore n is not divisible by any perfect squares.
(b) Express the number 32, 955, 000 = 23 · 3 · 54 · 133 as the product of
a squarefree number and a perfect square.
Solution:

32, 955, 000 = 23 · 3 · 54 · 133

= 22 · 54 · 132 · 2 · 3 · 13
= (2 · 52 · 13)2 · (2 · 3 · 13)
= 6502 · 78.

Hence 32, 955, 000 is the product of the perfect square 6502 and
the squarefree number 78 = 2 · 3 · 13.
(c) Let n ≥ 2 be a positive integer. Then either n is squarefree, or n
is a perfect square, or n is the product of a squarefree number and
a perfect square.
Solution: Let n ≥ 2 be a positive integer. We factor n into
primes using the fundamental theorem of arithmetic and break
the proof into cases.
case 1: Suppose that n’s prime factorization contains primes to
even powers and primes to odd powers. Then
2ea 2f1 +1 2f2 +1
n = p12e1 · p2e
2 · · · pa q 1
2
q2 · · · qb2fb +1

where the pi are the primes in the factorization of n that are raised
to an even power and the qi are the primes in the factorization of
n that are raised to an odd power. We then have that
 2
n = pe11 · pe22 · · · peaa q1f1 q2f2 · · · qbfb q1 · q2 · · · qb .

If all the ei and fi are zero then n is a squarefree number. Other-

wise, n is the product of a perfect square and a squarefree number.
case 2: Suppose that n’s prime factorization only contains primes
to odd powers. Then

n = q12f1 +1 q22f2 +1 · · · qb2fb +1

where the qi are primes. We then have that

 2
n = q1f1 q2f2 · · · qbfb q1 · q2 · · · qb .

If not all the fi are zero then n is the product of the perfect square
and the squarefree number. If all the fi are zero then

n = q1 · q2 · · · qb

and so n is a squarefree integer.

case 3: Suppose that n’s prime factorization only contains primes
to even powers. Then there are primes pi where

n = p2e 2e2 2ea

1 · p 2 · · · pa
1
= (pe11 · pe22 · · · peaa )2 .

Here n is a perfect square.

9. Suppose that x, y, z ∈ Z such that x > 0, y > 0, z > 0, gcd(x, y, z) = 1,
and x2 + y 2 = z 2 . Prove that gcd(x, z) = 1.
Solution: Suppose that x, y, z ∈ Z such that x > 0, y > 0, z > 0,
gcd(x, y, z) = 1, and x2 + y 2 = z 2 . We now show that gcd(x, z) = 1.
We do this by showing that the negation of this cannot happen.
Suppose that gcd(x, z) > 1. Then, by exercise 5a, there exists a prime
p such that p|x and p|z. Then x = pk and z = pm for some integers k
and m. Then (pk)2 + y 2 = (pm)2 . Hence p[pm2 − pk 2 ] = y 2 . Thus p|y 2 .
Recall that if a prime divides a product of two integers then the prime
must divide one of the integers. Therefore p|y. But then p|x, p|y, and
p|z, which implies that gcd(x, y, z) ≥ p. This contradicts the fact that
gcd(x, y, z) = 1. Therefore, cannot have that gcd(x, z) > 1.