Home page for accesible maths 4 Number theory

Style control - access keys in brackets

Font (2 3) - + Letter spacing (4 5) - + Word spacing (6 7) - + Line spacing (8 9) - +

4.4 Applications of Bézout’s Theorem

In this section we shall explore some of the theoretical consequences of Bézout’s Theorem.

Our first proposition combines several statements from Section 4.2 into a single result.

Proposition 4.4.1

Let a,b{0}a,b\in\mathbb{Z}\setminus\{0\}, and let dd\in\mathbb{N} be a common factor of aa and bb. Then the following three conditions are equivalent:

  1. (a)

    dd is the highest common factor of aa and b;b;

  2. (b)

    dd is an integral linear combination of aa and b;b;

  3. (c)

    every common factor of aa and bb is a factor of dd.

Proof. Theorem 4.2.17 shows that (a) implies (b), while Lemma 4.2.13 ensures that (b) implies (c), and finally (c) implies (a) by Lemma 4.2.8. \Box

We shall next give an explicit description of all the integral linear combinations of a pair of non-zero integers.

Proposition 4.4.2

Let a,b{0}a,b\in\mathbb{Z}\setminus\{0\}. Then the integral linear combinations of aa and bb are precisely the multiples of hcf(a,b)\mathrm{hcf}(a,b).

Example 4.4.3

If we take a=18a=18 and b=30b=30, we know that hcf(a,b)=6=2a-b;\mathrm{hcf}(a,b)=6=2a-b; since for instance 24=4624=4\cdot 6 we have 24=4(2a-b)=8a-4b=818-430.24=4(2a-b)=8a-4b=8\cdot 18-4\cdot 30.

Proof. Let d=hcf(a,b)d=\mathrm{hcf}(a,b) and nn\in\mathbb{Z}. We can then rewrite the statement as follows:

(r,s)(n=ra+sb)(d|n).(\exists\,r,\,s\in\mathbb{Z})(n=ra+sb)\ \Leftrightarrow\ (d|n).

‘‘\Rightarrow’’. Suppose that n=ra+sbn=ra+sb for some r, sr,\,s\in\mathbb{Z}. Then, as d|ad|a and d|b,d|b, Lemma 4.2.13 implies that d|nd|n.

‘‘\Leftarrow’’. Suppose that d|nd|n, say n=mdn=md for some mm\in\mathbb{Z}. By Bézout’s Theorem, we can write d=ua+vbd=ua+vb for some u, vu,\,v\in\mathbb{Z}, and therefore we have

n=md=(mu)a+(mv)b=ra+sbn=md=(mu)a+(mv)b=ra+sb

provided that we define r=mur=mu and s=mvs=mv. \Box

Proposition 4.4.4

Let a,b{0}a,b\in\mathbb{Z}\setminus\{0\} and cc\in\mathbb{N}. Then hcf(ca,cb)=chcf(a,b)\mathrm{hcf}(ca,cb)=c\cdot\mathrm{hcf}(a,b).

Example 4.4.5

Since hcf(12,20)=4\mathrm{hcf}(12,20)=4, we have hcf(120,200)=104=40.\mathrm{hcf}(120,200)=10\cdot 4=40.

Proof. Let d=hcf(a,b)d=\mathrm{hcf}(a,b); then d|ad|a and d|bd|b so cd|cacd|ca and cd|cbcd|cb, so that cdcd is a common factor of caca and cbcb. By Theorem 4.2.17 d=ra+sbd=ra+sb for some r,sr,s\in\mathbb{Z}, so cd=r(ca)+s(cb)cd=r(ca)+s(cb); thus Corollary 4.2.15 implies that hcf(ca,cb)=cd.\mathrm{hcf}(ca,cb)=cd. \Box

Corollary 4.4.6

Let a,b{0}a,b\in\mathbb{Z}\setminus\{0\} with d=hcf(a,b)d=\mathrm{hcf}(a,b), and write a=dαa=d\alpha and b=dβb=d\beta, where α,β\alpha,\beta\in\mathbb{Z}. Then hcf(α,β)=1\mathrm{hcf}(\alpha,\beta)=1.

Example 4.4.7

hcf(27,45)=9\mathrm{hcf}(27,45)=9; we have 27=9327=9\cdot 3 and 45=9545=9\cdot 5, and 33 and 55 are coprime.

Proof. By Proposition 4.4.4 we have

d=hcf(a,b)=hcf(dα,dβ)=dhcf(α,β),d=\mathrm{hcf}(a,b)=\mathrm{hcf}(d\alpha,d\beta)=d\cdot\mathrm{hcf}(\alpha,% \beta),

so as d0d\neq 0, hcf(α,β)=1\mathrm{hcf}(\alpha,\beta)=1 as required. \Box

Our final result is called Euclid’s Lemma; it will be important in our study of prime numbers in Section 4.6.

Theorem 4.4.8

Let a,b,p{0}a,b,p\in\mathbb{Z}\setminus\{0\}, and suppose that p|abp|ab and that aa and pp are coprime. Then p|bp|b.

Example 4.4.9

60=51260=5\cdot 12; as 6|606|60 and hcf(6,5)=1\mathrm{hcf}(6,5)=1, we have 6|12.6|12.

Proof. Since p|abp|ab, we can take qq\in\mathbb{Z} with pq=abpq=ab. As hcf(p,a)=1\mathrm{hcf}(p,a)=1, Theorem 4.2.17 implies that there are r,sr,s\in\mathbb{Z} such that rp+sa=1rp+sa=1. Then

b\displaystyle b =b1=b(rp+sa)=brp+sab\displaystyle=b\cdot 1=b(rp+sa)=brp+sab
=brp+spq=p(br+sq),\displaystyle=brp+spq=p(br+sq),

and so p|bp|b as required. \Box