Definition
A least common multiple (LCM) is \(m \in \mathbf{Z_{\geq 0}}\) such that:
  1. \(a \ | \ m\) and \(b \ | \ m\)
  2. If \(n \in \mathbf{Z}\), \(a \ | \ n\) and \(b \ | \ n\), then \(m \ | \ n\).



In fact the LCM is unique if it exists. To prove it’s existence we’ll need the following definition

Definition (Well Ordering Principle)
A non-empty subset of \(\mathbf{N}\) has a smallest element



Next, we’ll prove the existence of the LCM.

Proposition
The LCM always exists


Proof
Let \(a, b \in \mathbf{Z}\). Let \(J = \mathbf{Z}a \cap \mathbf{Z}b\). We want to show that \(J\) itself is a set of multiples so we can write \(J = \mathbf{Z}m\) for some \(m \in \mathbf{Z}_{\geq 0}\). This means that \(m\) must be the LCM of \(a\) and \(b\) by definition. [Why? If \(J = \mathbf{Z}m\), then \(m\) is the least element in \(J\), all the other elements are multiples of \(m\). We have \(a \ | \ m\) and \(b \ | \ m\) because \(J\) also contains common multiples of \(a\) and \(b\). And for any \(n \in \mathbf{Z}\), if \(a \ | \ n\) and \(b \ | \ n\), then \(m \ | \ n\). This is true because \(m\) is the smallest of all these common multiples.] We will use division by remainder to prove this!

Observe that \(J\) is closed under addition and subtraction. So if \(c, d \in J\), then any integer combination of \(c\) and \(d\) will be in \(J\). So for any \(m, n \in \mathbf{Z}\), \(cm + dn \in J\).

We have two cases:
Case 1: If either \(a = 0\) or \(b = 0\), then \(J = \{0\} = \mathbf{Z}0\).

Case 2: If \(a \neq 0\) and \(b \neq 0\), then \(J\) is the set of common multiples of \(\mathbf{Z}a\) and \(\mathbf{Z}b\). Consider \(J \cap \mathbf{N}\). This is the set of positive common multiples. This set is not empty because it will at least include \(|ab|\). So \(J \cap \mathbf{N}\) is a non-empty subset of \(\mathbf{N}\). By the well-ordering principle, there is a smallest number in \(J \cap \mathbf{N}\). Let \(m \in J \cap \mathbf{N}\) be the smallest common multiple. (so we’re setting \(m\) to be the smallest positive common multiple of \(\mathbf{Z}a\) and \(\mathbf{Z}b\))

We claim that \(J = \mathbf{Z}m\) (the set of multiples of \(m\)). We will show this by proving that \(J \subseteq \mathbf{Z}m\) and that \(\mathbf{Z}m \subseteq J\).

\(\mathbf{Z}m \subseteq J\): This is true since \(m \in J\) and we showed earlier that \(m\) is closed under addition and subtraction so any multiple of \(m\) is also in \(J\) and therefore \(\mathbf{Z}m \subseteq J\).

\(J \subseteq \mathbf{Z}m\): Suppose \(x \in J = \mathbf{Z}a \cap \mathbf{Z}b\). Consider what happens when we divide \(x\) by \(m\). We can write \(x = qm + r\). with \(q, r \in \mathbf{Z}\) and \(0 \leq r < m\). Re-write the equation as

$$ \begin{align*} r = x - qm. \end{align*} $$

Observe now that \(x \in J\) by assumption. \(qm\) is a multiple of \(m\) so it’s in \(J\). So their difference is also in \(J\) which means that \(r \in J\). If \(r = 0\), \(x = qm \in \mathbf{Z}m\) and we’re done. If \(r \neq 0\), then \(r \in J\), then \(r \in J \cap \mathbf{N}\) but also \(r < m\). But that’s a contradiction because \(m\) is the smallest element in \(J \cap \mathbf{N}\) and so this case doesn’t happen. So \(J = \mathbf{Z}m\) as desired. \(\ \blacksquare\).



Consequences of the LCM

The next few propositions are some consequences of the LCM.

Proposition
If \(d = gcd(a,b) = 1\), then \(m = lcm(a,b) = ab\)


Proof
We are given that \(d = gcd(a,b) = 1\). This means that \(a\) and \(b\) are relatively prime. Condition 1 is true since \(a \ | \ ab\) and \(b \ | \ ab\). For condition two of the LCM definition, let \(n \in \mathbf{Z}\) such that \(a \ | \ n\) and \(b \ | \ n\). By the proposition from lecture 5, \(ab \ | \ n\). (Proposition: If \(a\) and \(b\) are relatively prime, then if \(a \ | \ n\) and \(b \ | \ n\), then \(ab \ | \ n\)). \(\ \blacksquare\)

Proposition
Let \(e\) be a common divisor of \(a\) and \(b\) so \(e \ | \ a\) and \(e \ | \ b\). Then \(lcm(\frac{a}{e}, \frac{b}{e}) = \frac{1}{e} lcm(a,b)\).


Proof
[TODO] \(x\) is a common divisor of \(\frac{a}{e}\) and \(\frac{b}{e}\) if and only if \(xe\) is a common divisor of \(a\) and \(b\).

Proposition
If \(a, b \in \mathbf{N}\), \(d = gcd(a,b)\) and \(m = lcm(a,b)\), then \(m = \frac{ab}{d}\)


Proof
We need the previous two propositions

  1. If \(d = 1\), then \(m = lcm(a,b) = ab\)
  2. If \(e \ | \ a\) and \(e \ | \ b\), then \(lcm(\frac{a}{e}, \frac{b}{e}) = \frac{1}{e} lcm(a,b)\)

If we divide both \(a\) and \(b\) by \(d\), their gcd will become 1 so \(gcd(\frac{a}{d}, \frac{b}{d}) = 1\). Since the gcd is 1, then \(lcm(\frac{a}{d}, \frac{b}{d}) = \frac{b}{d}\frac{a}{d}\) by part \((1)\). So now let \(e = d\), so by \((2)\),

$$ \begin{align*} \frac{1}{d} lcm(a, b) &= lcm(\frac{a}{d}, \frac{b}{d}) \\ \frac{m}{d} &= \frac{b}{d}\frac{a}{d} \\ m &= \frac{ab}{d} \\ \end{align*} $$





Pairwise Relatively Prime


Definition
\(a_1,...,a_k \in \mathbf{Z}\) are pairwise relatively prime if \(gcd(a_i,a_j) = 1\) for all \(1 \leq i,j \leq k, i \neq j\)



Some facts based this:

Lemma
If \(a_1,...,a_k \in \mathbf{Z}\) are pairwise relatively prime, then \(a_k,b = a_1,...,a_{k-1}\) are relatively prime so \(gcd(a_k,b) = 1\).


Proof
Suppose for the sake of contradiction that \(a_k\) and \(b\) are not relatively prime so they have a common divisor. Let \(p\) be a prime that is a common divisor of \(a_k\) and \(b\). So \(p \ | \ b\) which means that \(p \ | \ a_1a_2...a_{k-1}\). But since \(p\) is prime, then by the proposition from the previous lecture, \(p\) must divide one of the factors in \(a_1a_2...a_{k-1}\). Let \(p\) divide \(a_i\) for some \(1 \leq i \leq k - 1\). However \(p \ | \ p_k\). This means that \(gcd(a_i,a_k) \neq 1\). This is a contradiction since we assumed that \(a_1,...,a_k\) are pairwise relatively prime. Therefore, we must have \(a_k\) and \(b\) be relatively prime. \(\ \blacksquare\)

The consequence of this lemma is the following

Proposition
If \(a_1, a_2,...,a_k\) are pairwise relatively prime and if \(a_i \ | \ n\) for all \(i = 1,...,k\), then \(a_1a_2,...,a_k \ | \ n\)


Proof
By Induction on \(k\)
Base Case: If \(k=1\), then there is nothing to prove.
If \(k = 2\), then \(gcd(a_1,a_2)=1\). If \(a_1 \ | \ n\) and \(a_1 \ | \ n\), then by the proposition from lecture 5, we must have \(ab \ | \ n\).

Inductive Case \(k \geq 3\):
Let \(b = a_1a_2...a_{k-1}\). By the previous lemma we know that \(gcd(a_k,b) = 1\). Also by the inductive hypothesis if \(a_1,...,a_{k-1} \ | \ n\), then \(a_1...a_{k-1} = b \ | \ n\). but \(gcd(a_k,b) = 1\) so \(a_k\) and \(b\) are relatively prime. Since they are relatively prime and they both divide \(n\), then \(ba_k = a_1...a_{k-1}a_k\) also divides \(n\) as we wanted to show. \(\ \blacksquare\)





References