Scribd
Upload
522 views

Some Solutions For Burton's Elementary Number Theory: 2 Divisibility Theory in The Integers

The document summarizes key results from Burton's Elementary Number Theory textbook. It provides proofs and solutions to exercises involving divisibility, the Euclidean algorithm, Diophantine equations, prime numbers, congruences, representations of integers in different bases, number-theoretic functions like sums of divisors, and properties of quadratic reciprocity. The document establishes theoretical foundations for working with integers and their relationships.

Uploaded by

DYAH AYU SEKAR KINASIH PURWANINGRUM
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
522 views

Some Solutions For Burton's Elementary Number Theory: 2 Divisibility Theory in The Integers

The document summarizes key results from Burton's Elementary Number Theory textbook. It provides proofs and solutions to exercises involving divisibility, the Euclidean algorithm, Diophantine equations, prime numbers, congruences, representations of integers in different bases, number-theoretic functions like sums of divisors, and properties of quadratic reciprocity. The document establishes theoretical foundations for working with integers and their relationships.

Uploaded by

DYAH AYU SEKAR KINASIH PURWANINGRUM
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Some Solutions for Burton’s Elementary Number Theory

2 Divisibility Theory in the Integers


2.4 The Euclidean Algorithm
10 Given nonzero integers a and b, establish the following facts concerning lcm(a, b) :
(a) gcd(a,b)=lcm(a,b) if and only if a = ±b.
(b) If k > 0, then lcm(ka, kb) = klcm(a, b).
(c) If m is any common multiple of a and b, then lcm(a, b) | m. [Hint: Put t = lcm(a, b) and use the Division
Algorithm to write m = qt + r, where 0 ≤ r < t. Show that r is a common multiple of a and b.]
Solution: (a) If a = ±b then gcd(a, b) = |a| = lcm(a, b).
Conversely if gcd(a, b) = lcm(a, b) then a | lcm(a, b) and lcm(a, b) = gcd(a, b) | a tells us that lcm(a, b) = ±a.
Similarly, lcm(a, b) = ±b so a = ±b.
(b) By Theorem 2.7, since k > 0 we know gcd(ka, kb) = kgcd(a, b). So lcm(ka, kb)kgcd(a, b) = k 2 ab = k 2 lcm(a, b)gcd(a, b)
so lcm(ka, kb) = klcm(a, b).
(c) Let t = lcm(a, b). By the Division Algorithm, there exist integers q and 0 ≤ r < t such that m = qt + r. Then
r = m − qt so since a and b both divide m and t, they also divide r. So r is a common multiple of a and b strictly
less than the lowest positive common multiple, so r = 0. So m = qt which means t | m.

2.5 The Diophantine equation ax + by = c


4 If a and b are relatively prime positive integers, prove that the Diophantine equation ax − by = c has infinitely
many solutions in the positive integers. [Hint: There exist integers x0 and y0 such that ax0 +by0 = c. For any integer
t, which is larger than both |x0 |/b and |y0 |/a, a positive solution of the given equation is x = x0 + bt, y = −(yo − at).]
Solution using hint: Since gcd(a, b) = 1 | c we know there exist integers x0 , y0 such that ax0 + by0 = c by Theorem
2.9. Consider x = x0 + bt, y = −(y0 − at). Need to show they are solutions and they are positive.

ax − by = ax0 + abt + by0 − abt = ax0 + by0 = c.


So they are solutions.
Now we show they are positive if t > |xb0 | , |ya0 | . If t > |xb0 | then x = x0 + bt > x0 + |x0 | ≥ 0 and if t > |ya0 | then
y = −(y0 − at) > −(y0 − |y0 |) ≥ 0. (You may show x0 + |x0 | ≥ 0 by considering cases where x0 ≥ 0 and x0 < 0.)
So we have positive solutions for infinitely many t.
Alternative solution (basically the same): By Corollary to Theorem 2.9, there is a solution x0 , y0 and all
other solutions are of the form x = x0 − bt, y = y0 − at. If t < xb0 , ya0 so −bt > −x0 and −at > −y0 . So
x = x0 − bt > x0 − x0 = 0 and y = y0 − at > y0 − y0 = 0. So x, y are positive solutions for all t small enough, and
there are infinitely many such t.
Note: Remember to write that if t < xb0 then x > 0 and not the other way round! (Actually it’s an if and only if,
but the direction we care about is t < xb0 =⇒ x > 0).

3 Primes and their distribution


3.1 The fundamental theorem of arithmetic
5(b) If gcd(a, b) = p, a prime, what are the possible values of gcd(a2 , b2 ), gcd(a2 , b), gcd(a3 , b2 )?
Solution: a = pm, b = pn where gcd(m, n) = 1 so gcd(a2 , b2 ) = p2 gcd(m, n) = p2 .
Note that p cannot divide both ( m and n.
p if p - n
gcd(a2 , b) = p gcd(pm2 , n) = 2
.
p if p | n

1
(
p2 if p - n
gcd(a3 , b2 ) = p2 gcd(pm3 , n2 ) = .
p3 if p | n

3.2 The sieve of Eratosthenes


12(a) Assuming that pn is the nth prime number, establish that pn > 2n − 1 for n ≥ 5.
Solution: Proceed by induction on n. If n = 5 then pn = 11 > 9 = 2n − 1. For the inductive step, assume
pn−1 > 2(n − 1) − 1. Note that since n ≥ 5, pn ≥ pn−1 + 2 > 2n − 1.

4 The theory of congruences


4.3 Binary and decimal representations of integers
5(b) Give criteria for the divisibility of N by 3 and 8 that depend on the digits of N when written in the base 9.
Solution: Let N = am 9m + am−1 9m−1 + · · · + a1 9 + a0 where 0 ≤ ai < 9 for all i be the representation of N in base
9. Since 9 ≡ 0 (mod 3) we have N ≡ a0 (mod 3). So 3 | N ⇐⇒ 3 | a0 .
Since 9 ≡ 1 (mod 8) we have N ≡ am + · · · + a0 (mod 8). Then 8 | N ⇐⇒ 8 | am + · · · + a0 . (Or just use part (a)
with b = 9.)

6 Number-theoretic functions
6.1 The sum and number of divisors
7(b) σ(n) is an odd integer if and only if n is a perfect square or twice a perfect square. [Hint: If p is an odd prime,
then 1 + p + · · · + pk is odd only when k is even.]
Solution:

Lemma. If p is an odd prime then 1 + p + · · · + pk is odd iff k is even. If p = 2 then 1 + p + · · · + pk is odd.

Proof of Lemma. 1 + 2 + · · · + 2k = 1 + 2(1 + 2 + · · · + 2k−1 ) is odd.


Let p be odd. 1 + p + · · · pk is a sum of k + 1 odd numbers, so when k is even it is odd, and when k is odd it is
even.

Now we can prove both directions of the problem using this lemma and the fact that when n = 2k pk11 · · · pkr r where pi
are distinct odd primes (note that k may be 0), we have σ(n) = (1+2+· · ·+2k )(1+p1 +· · ·+pk11 ) · · · (1+pr +· · ·+pkr r ).
Assume σ(n) is odd. Then each (1 + pi + · · · + pki i ) is odd, and so all ki = 2li are even by the lemma. If k = 2l
is even then n = (2l pl11 · · · plrr )2 is a perfect square. If k = 2l + 1 is odd then n = 2(2l pl11 · · · plrr )2 is twice a perfect
square.
Assume n is a perfect square or twice a perfect square. Then n = 2k pk11 · · · pkr r where the pi are distinct odd primes
and the ki are even, so by the lemma σ(n) = (1 + 2 + · · · + 2k )(1 + p1 + · · · + pk11 ) · · · (1 + pr + · · · + pkr r ) is odd.

6.3 The greatest integer function


2(a) Let x be a real number and show [x + n] = [x] + n for any integer n.
Solution: Since x − 1 < [x] ≤ x we have x + n − 1 < [x] + n ≤ x + n and [x] + n is an integer, so since [x + n] is
the unique integer in (x + n − 1, x + n] we get [x] + n = [x + n].

2
8 Primitive roots and indices
8.1 The order of an integer modulo n
4 Assume that the order of a modulo n is h and the order of b modulo n is k. Show that the order of ab modulo n
divides hk; in particular if gcd(h, k) = 1, then ab has order hk.
Solution: Let m be the order of ab modulo n. We have(ab)hk ≡ ahk bhk ≡ 1(mod n) and so by Theorem 8.1, m | hk.
Now assume gcd(h, k) = 1. Then 1 ≡ (ab)mh ≡ amh bmh ≡ bmh and so k | mh. Since gcd(h, k) = 1 we know k | m.
Similarly, h | m. But gcd(h, k) = 1 implies that hk | m. We already knew that m | hk and so we get m = hk.

9 The quadratic reciprocity law


9.3 Quadratic reciprocity
10a Let p be an odd prime. Show (5/p) = 1 if and only if p ≡ 1, 9, 11, 19 (mod 20).
Solution: Since 5 ≡ 1 (mod 4) we know (5/p) = (p/5) ≡ p2 (mod 5). So (5/p) = 1 if and only if p2 ≡ 1 (mod 5) if
and only if p ≡ 1 or 4 (mod 5) if and only if p ≡ 1, 4, 6, 9, 11, 14, 16, 19 (mod 20) if and only if p ≡ 1, 9, 11, 19 (mod 20)
(since p is an odd prime).
Note: because each step is “if and only if” we have done both directions together.

You might also like