See a previous answer given here.
Basically, by the density of rationals, for $x, y \in \mathbb{R}$, there exists an $r \in \mathbb{Q} \setminus \{ 0 \}$, $$\dfrac{x}{\sqrt{2}} < r < \dfrac{y}{\sqrt{2}} \Longleftrightarrow x < r\sqrt{2} < y\text{.}$$
The product of a non-zero rational ($r$) and an irrational ($\sqrt{2}$) is irrational (which is easy to show), so we've proven the density of irrationals.
The statement of the density of rationals I have is:
If $x, y \in \mathbb{R}$, there exists an $r \in \mathbb{Q}$ such that $x < r < y$.
The problem I see with this is that $r$ could very well be $0$. How are we allowed to assume that there exists an $r\in \mathbb{Q} \setminus \{ 0 \}$? I would guess intuitively that we could work with a sequence of nested intervals of some sort (perhaps?) to find a non-zero rational, but sequences have not been covered in this course yet.