Concluding a proof ($\pi$ is irrational)

109 Views Asked by At

I am making a proof for irrationality of $\pi$ and i proceeded as follows:

Let $\pi=\frac{u}{v}$ for some $u,v\in \Bbb{N}$, define family of integrals: $$I_n=\frac{v^{2n}}{n!}\int_0^\pi x^n(\pi-x)^n \sin x\,dx$$ By some elementary estimates we have $$0<I_n\leq\frac{v^{2n}}{n!}\pi^{2n+1}$$ thus by squeeze lemma we have $\lim_{n\to\infty}I_n=0$.

In the next part I prove that $I_0,I_1\in \Bbb{N}$ and applying some integration by parts i get recursive formula $$I_n=(4n-2)v^2I_{n-1}-u^2v^2I_{n-2}\tag{1}$$ Now I should conclude the proof and I have two ideas but I'm not really sure whether both of them are correct (if both, which one is better?):

a) Because of (1), we can say that $\lim_{n\to\infty}I_n=\infty$ because of the factor $4n-2$ thus implying there are two different limits for $I_n$ which is impossible thus contradiction.

b) Because $I_0,I_1\in \Bbb{N}$. Then because $u,v\in\Bbb{N}$ too, recursive formula shows that $\forall n\in\Bbb{N}:I_n\in\Bbb{N}$ but one cannot have infinite sequence of natural numbers tending to zero from above (obviously they can't approach from below).

1

There are 1 best solutions below

0
On BEST ANSWER

Good question! Given the fact that your estimate and (1) are correct, you should choose (b) rather than (a) because the latter is not correct.

As @InterstellarProbe has mentioned in the comment, there are some sequences $a_n$ tending to $0$, but with $(4n-2)a_n$ still tending to zero$.

The point is how fast $a_n$ tends to zero. If it tends to zero "faster than" $(4n-2)$ tends to $\infty$, then their product $(4n-2)a_n$ still tends to zero.