1
$\begingroup$

If the irrationality criterion below is wrong feel free to edit this post.

The irrationality criterion, as far I understood is that if exists sequences $p_n,q_n (n=1,2,3,...)$ where $p_n,q_n$ are integers then the number $\alpha \neq p_n/q_n$ is irrational if \begin{align} \left| \alpha -\frac{p_n}{q_n}\right|<\frac{1}{q_n^{1+\delta}} \end{align} where $\delta>0$.

If we multiply everything by $q_n$ we have $|q_n\alpha-p_n|<1/q_n^\delta$ so assuming that $\alpha=r/s$ with both $r,s$ integers, we find a integer in the interval $(0,1)$ because $\delta>0$, which is a contradiction.

So we must find sequences that $p_n,q_n$ such that $p_n/q_n$ converges very quickly to $\alpha$. How does one create such sequences and how to find $\delta$ ? I guess they don't fall from the sky.

For example, how we prove that $e$ is irrational using this criterion?

$$\left| e -\frac{p_n}{q_n}\right|<\frac{1}{q_n^{1+\delta}}$$

$\endgroup$
0
$\begingroup$

$e$ is rather exceptional in that it has a closed-form continued fraction:

$$e = [2; 1, 2, 1, 1, 4, 1, 1, 6, 1, 1, 8,\ldots]$$

The corresponding convergents $p_n/q_n$ are $$ 2, 3, 8/3, 11/4, 19/7, 87/32, 106/39, 193/71, 1264/465, 1457/536, \ldots $$ which satisfy your inequality with $\delta = 1$.

Of course, if you have the continued fraction you really don't need your criterion: any non-terminating simple continued fraction represents an irrational number.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.