MATH 3005 Homework Han-Bom Moon

Homework 1 Solution
Chapter 1 and Section 2.1.
deadline: Sep. 16, 1:00 pm

• Do not abbreviate your answer. Write everything in full sentences.

• Write your answer neatly. If I couldn’t understand it, you’ll get 0 point.
• You may discuss with your classmates. But do not copy directly.

1. Write a = 189720 and b = 36192 as products of positive prime numbers. By using

them, find (a, b).

a = 23 · 32 · 5 · 17 · 31, b = 25 · 3 · 13 · 29.

Therefore (a, b) = 23 · 3 = 24.

2. Let a, b be two positive integers.

(a) Show that if a|b and b|a, then a = b.

Because a|b, there is k ∈ Z such that b = ak. Since b|a, there is ` ∈ Z such that
a = b`. So a = b` = ak`. Because a is nonzero, by dividing both side by a,
we obtain k` = 1. So k = ±1 and b = ±a. Because both a and b are positive,
k = 1 and a = b.
(b) Show that (a) may not be true for arbitrary integers.
If a = 2 and b = −2, then a|b and b|a but a 6= b.
(c) Show that (a, b) = a if and only if a|b.
Suppose that (a, b) = a. Then by definition, a is a divisor of b. Conversely,
suppose that a|b. Then a is a common divisor of a and b, because a|a. Since
a is positive, every divisor of a is less than or equal to a. Therefore a is the
greatest common divisor of a and b.

3. Let a, b and c be three nonzero integers such that (a, c) = (b, c) = 1. Show that
(ab, c) = 1.
(Here I would like to propose a proof without using a prime factorization.)
Because (a, c) = 1, there are s, t ∈ Z such that sa + tc = 1. Because (b, c) = 1, there
are k, ` ∈ Z such that kb + `c = 1. Now

1 = (sa + tc)(kb + `c) = skab + sa`c + tckb + tc`c = sk · ab + (sa` + tkb + tc`)c.

So 1 can be written as a linear combination of ab and c. Therefore the smallest

positive linear combination of ab and c is 1. Thus (ab, c) = 1.

1
MATH 3005 Homework Han-Bom Moon

Solution by Ella Chapman. Suppose that (ab, c) = k. k ≥ 1 since it is a positive

integer. Then there are `, m ∈ Z such that ab = k` and c = km. Since (a, c) = 1,
there are s, t ∈ Z such that sa + tc = 1. So we have

b = bsa + btc = sab + btc = sk` + btkm = k(s` + btm).

Therefore k|b. So k is a positive common divisor of b and c, which has to be 1,

since (b, c) = 1.

4. Let a, b be two positive integers. Suppose that q and r are two integers obtained
from Division Algorithm: a = bq + r and 0 ≤ r < b. Show that (a, b) = (b, r).
Let d = (a, b). Then a = kd and b = `d for some k, ` ∈ Z. Now r = a − bq =
kd − `dq = (k − `q)d. So d|r. So d is a common divisor of b and r. This implies
(a, b) = d ≤ (b, r).
On the other hand, let e = (b, r). Then b = xe and r = ye for some x, y ∈ Z. We
have a = bq + r = xeq + ye = (xq + y)e. So e|a, and e is a common divisor of a
and b. Therefore (b, r) = e ≤ (a, b). So we obtain (a, b) = (b, r).
Note that the existence of s, t ∈ Z such that sa + tb = d does NOT imply that
d = (a, b).

5. The previous problem implies that we may reduce the computation of the great-
est common divisor of two integers into the same problem for smaller integers:
By using this idea, find (a, b) for a, b in Problem 1. This method is called Euclid’s
algorithm. Compare this computation with Problem 1 - which one will be faster if
a and b are very large? Why do you think so?
By applying division algorithm, we obtain 189720 = 36192 · 180960 + 8760. Then
(189720, 36192) = (36192, 8760).
From 36192 = 8760 · 4 + 1152, (36192, 8760) = (8760, 1152).
From 8760 = 1152 · 7 + 696, (8760, 1152) = (1152, 696).
From 1152 = 696 · 1 + 456, (1152, 696) = (696, 456).
From 696 = 456 · 1 + 240, (696, 456) = (456, 240).
From 456 = 240 · 1 + 216, (456, 240) = (240, 216).
From 240 = 216 · 1 + 24, (240, 216) = (216, 24).
Finally, because 24|216, (216, 24) = 24 by Problem 2.(c). Therefore (a, b) = 24.
When a and b are large, Euclid’s algorithm will provide a faster solution, because
in the computation of the prime factorization, finding a prime factor is very diffi-
cult. (The only way to find a prime factor is to divide the given integer by many
prime numbers.)

2
MATH 3005 Homework Han-Bom Moon

6. Let R be the set of real numbers. Define a relation ≡ as

a ≡ b ⇔ a − b ∈ Z.

(a) Show that this is an equivalence relation.

For any a ∈ R, a − a = 0 ∈ Z. So a ≡ a. If a ≡ b, then a − b ∈ Z. Then
b − a = −(a − b) ∈ Z, too. Thus b ≡ a. Finally, if a ≡ b and b ≡ c, a − b ∈ Z
and b − c ∈ Z. Then a − c = (a − b) + (b − c) ∈ Z. Therefore a ≡ c. This
proves the fact that ≡ is indeed an equivalence relation.
√
(b) List five elements of [ 3].
√ √ √ √ √
For instance, 3, 3 + 1, 3 + 2, 3 − 1, 3 − 2, · · · .
(c) (Optional) Find an interval which is naturally bijective to R/ ≡.
If we have any real number a whose decimal expression is a = b1 b2 · · · bk .c1 c2 c3 · · · ,
then a ≡ a0 where a0 = 0.c1 c2 c3 · · · because a − a0 = b1 b2 · · · bk ∈ Z. There-
fore every real number is equivalent to a unique real number on [0, 1). Thus
R/ ≡ = [0, 1).

7. Let X be a set and ≡ be an equivalence relation on X. Prove that for any two
elements x and y of X, if [x] ∩ [y] 6= ∅, then [x] = [y].
Suppose that [x] ∩ [y] 6= ∅. Then there is z ∈ [x] ∩ [y]. Because z ∈ [x], z ≡ x. Also
since z ∈ [y], z ≡ y. Note that by symmetry, x ≡ z. So by transitivity, x ≡ y.
Take any w ∈ [x]. Then w ≡ x. From x ≡ y and transitivity, w ≡ y. So w ∈ [y].
This implies [x] ⊂ [y]. If v ∈ [y], then v ≡ y and by symmetry, y ≡ v. From x ≡ y
and y ≡ v, x ≡ v. By symmetry, v ≡ x. So v ∈ [x]. Therefore [y] ⊂ [x]. Thus we
obtain [x] = [y].

8. Let n be a positive integer.

(a) For two integers a and b, is it true that ab ≡ 0 (mod n) implies a ≡ 0 (mod n)
or b ≡ 0 (mod n)? If it is true, prove it. If it is not, give a counterexample.
Counterexample: n = 0, a = 2, b = 3. Then 2 · 3 ≡ 6 ≡ 0 (mod 6), but
2 6≡ 0 (mod 6) and 3 6≡ 0 (mod 6).
(b) Now let n be a positive prime number. Do (a).
(This is indeed a part of one of theorems we proved in class.)
Suppose that n is a prime number. If ab ≡ 0 (mod n), then n|ab. Since n is
prime, n|a or n|b by Theorem 1.5. So a ≡ 0 (mod n) or b ≡ 0 (mod n).