Skip to main content

Number Theory

Number theory is a part of mathematics that examines the abstract properties of numbers.

Division Theorem

Division is one of the four basic mathematical operations which allows us to split numbers into equal amounts. The theorem that defines how division works is the division theorem. This theorem states: Let aa, bb be integers, b>0b > 0. Then there are unique integers qq, rr such that a=qb+ra = q \cdot b + r and 0r<b0 \leq r < b.

Theorem: The Division Theorem

Proof: In order to prove this theorem, we will prove existence first, then uniqueness.

To prove existence, we need to look at all non-negative integers of the form akba - kb, where kk is an integer, and show that one of them is less than bb. Such integers do exist.

Take k=ak = -|a|. Then, since b1b \geq 1, akb=a+aba+a0a - kb = a + |a|b \geq a + |a| \geq 0.

Let rr be the smallest such integer. Let qq be the value of kk for which it occurs, i.e. r=aqbr = a - qb. To complete the proof, we show that r<br < b.

Suppose, on the contrary, that rbr \geq b then a(q+1)b=aqbb=rb0a - (q + 1) b = a - qb - b = r - b \geq 0. Thus a(q+1)ba - (q + 1) b is a non-negative integer of the form akba - kb.

But rr is the smallest such, and yet a(q+1)b<aqb=ra - (q + 1) b < a - qb = r. This is a contradiction. Hence r<br < b. That proves existence.

To prove uniqueness, we need to show that if there are two representations of aa, a=qb+r=qb+ra = qb + r = q^{\prime}b + r^{\prime}, 0r0 \leq r, r<br^{\prime} < b, then r=rr = r^{\prime} and q=qq = q^{\prime}.

Rearranging the above equations: rr=b(qq)r^{\prime} - r = b (q - q^{\prime}).

Taking absolute values: rr=bq=qr|r^{\prime} - r| = b |q = qr|.

But b<r0-b < -r \leq 0 and 0r<b0 \leq r^{\prime} < b, so b<rr<b-b < r^{\prime} - r < b meaning rr<b|r^{\prime} - r| < b.

So due to rr=bq=qr|r^{\prime} - r| = b |q = qr|, bqq<bb |q - q^{\prime}| < b. Hence qq<1|q - q^{\prime}| < 1 which means q=qq = q^{\prime}. Then due to rr=b(qq)r^{\prime} - r = b (q - q^{\prime}), r=rr = r^{\prime}. That proves uniqueness.

General Theorem

The issue with the division theorem we stated above is that it only covers the cases where b>0b > 0. The General Division Theorem states: Let aa, bb be integers, b0b \neq 0. Then there are unique integers qq, rr such that a=qb+ra = qb + r and 0rb0 \leq r \leq |b|.

Theorem: The General Division Theorem

Proof: We have proved the result in the case b>0b > 0, so assume b<0b < 0.

Then, since b>0|b| > 0, the previous theorem tells us there are unique integers qq^{\prime}, rr^{\prime} such that a=qb+ra = q^{\prime} \cdot |b| + r^{\prime} and 0r<b0 \leq r^{\prime} < |b|.

Let q=qq = -q^{\prime}, r=rr = r^{\prime}. Then, since b=b|b| = -b, we got a=qb+ra = qb + r, 0rb0 \leq r \leq |b|.

note

In the division theorem, The variable qq refers to the quotient of aa by bb, and rr is the remainder.

Also note this theorem does not work for b=0b = 0 because we cannot divide by zero.

Divisibility

If division of aa by bb produces a remainder r=0r = 0, we say aa is divisible by bb. Hence, aa is divisible by bb iff there is an integer qq such that a=bqa = b \cdot q. For example, 4545 is divisible by 99 because 45=9545 = 9 \cdot 5 but 4444 is not divisible by 99 because 44=94+844 = 9 \cdot 4 + 8.

We denote divisibility by bab|a which means aa is divisible by bb. bab|a is not the same as b/ab/a because bab|a refers to the relationship between aa and bb which is either true or false. On the other hand b/ab/a denotes a rational number which is the result of dividing bb by aa.

note

We can redefine prime numbers in terms of divisibility. A prime number is an integer p>1p > 1 that is divisible by 11 and pp.

We can better understand divisibility and use it for proofs by understanding its basic properties:

  1. a0a|0, aaa|a.
  2. a1a|1 iff a=±1a = \pm 1.
  3. If aba|b and cdc|d then acbdac|bd.
  4. If aba|b and bcb|c, then aca|c (for b0b \neq 0).
  5. [aba|b and bab|a] iff a=±ba = \pm b.
  6. If aba|b and b0b \neq 0, then ab|a| \leq |b| iff a±ba \neq \pm b.
  7. If aba|b and aca|c, then a(bx+cy)a|(bx + cy) for any integers xx, yy.

Fundemental Theorem of Arithmetic

Finally, the Fundemental Theorem of Arithmetic states that every natural number greater than 11 is either prime or can be expressed as a product of primes in a way that is unique except for the order in which they are written. The expression of a number as a product of primes is called its prime decomposition.

note

In order to proof uniqueness for this theorem, we need Euclid's Lemma which states If a prime pp divides a product abab, then pp divides at least one of aa, bb.

Theorem: The Fundemental Theorem of Arithmetic

Proof: In order to prove this theorem, we will prove existence first, then uniqueness.

We will prove existence using proof by contradiction.

Suppose there were a composite number (i.e. non prime) that could not be written as a product of primes. Then there must be a smallest such number. Call it nn. Since nn is not prime, there are numbers aa, bb with 1<a,b<n1 < a, b < n such that n=abn = ab.

If aa, bb are primes, then n=abn = ab is a prime decomposition of nn and we have a contradiction.

If either of aa, bb is composite, then because it is less than nn, it must be a product of primes, so by replacing one or both aa, bb by its prime decomposition in n=abn = ab, we get a prime decomposition of nn, and again we have a contradiction. That proves existence.

Now we can prove uniqueness. The prime decomposition of any natural number n>1n > 1 is unique not including the order the primes. We can prove this by contradiction.

Assume there is a number n>1n > 1 that has two (or more) different prime decompositions. Let nn be the smallest such number.

Let n=p1,p2,...,pr=q1,q2,...,qsn = p_1, p_2, ..., p_r = q_1, q_2, ..., q_s be two different prime decompositions of nn. Since p1p_1 divides (q1)(q2...qs)(q_1)(q_2 ... q_s), by Euclid's Lemma, either p1q1p_1|q_1 or p1(q2...qs)p_1|(q_2 ... q_s).

Hence, either p1=q1p_1 = q_1 or else p1=qip_1 = q_i for some ii between 22 and ss. But then we can delete p1p_1 and qiq_i from n=p1,p2,...,pr=q1,q2,...,qsn = p_1, p_2, ..., p_r = q_1, q_2, ..., q_s, which gives us a number smaller than nn that has two different prime decompositions, contrary to the choice of nn as the smallest such. That proves uniqueness.