8
$\begingroup$

The question is if there exists a standard way of proving that $$ \cos{ π \over 9}$$ is an irrational number.

$\endgroup$
2
  • 4
    $\begingroup$ What do you mean with a "standard way" ? It is possible to determine the minimal polynomial of $\cos(\frac{\pi}{9})$, but this is not quite easy. $\endgroup$
    – Peter
    Nov 26, 2016 at 19:10
  • 1
    $\begingroup$ i know how to find the minimal polynomial of this. But he said i could do it much easier.(like the way you prove that sqrt(2)is irrational) $\endgroup$ Nov 26, 2016 at 19:13

4 Answers 4

16
$\begingroup$

Hint: Use $\frac{1}{2}=\cos\frac{\pi}{3}=4\cos^3\frac{\pi}{9}-3\cos\frac{\pi}{9}$, then prove that $4x^3-3x=\frac{1}{2}$ does not have rational roots

$\endgroup$
2
7
$\begingroup$

Observe that $\cos(\frac{\pi}{3}) = \dfrac{1}{2}$. Thus using the identity: $\cos (3x) = 4\cos^3 x - 3\cos x$, $\cos(\frac{\pi}{9})$ is a zero of the equation: $4x^3 - 3x = \dfrac{1}{2}$,and using the irrational root test, one concludes that this equation has no rational root, hence $\cos(\frac{\pi}{9})$ must be irrational.

$\endgroup$
5
$\begingroup$

Using either trig identities, or plug-and-simplify of $\frac{1}{2} (e^{\pi i/9}+e^{-\pi i /9})$, we verify that $\cos\frac{\pi}{9}$ is a root of $f(x)=8x^3-6x-1$. By the Rational Root Theorem, the only possible rational roots of this polynomial are $\pm 1, \pm \frac{1}{2}, \pm \frac{1}{4},\pm \frac{1}{8}$, and none of these eight are correct.

$\endgroup$
2
  • $\begingroup$ And how do we get the minimal polynomial ? You should work this out. $\endgroup$
    – Peter
    Nov 26, 2016 at 19:13
  • 2
    $\begingroup$ I got it through a lucky guess, i.e. wolfram alpha. $\endgroup$
    – vadim123
    Nov 26, 2016 at 19:13
1
$\begingroup$

Using Galois theory: Let $n\in\mathbb N$. If $2\cos(2\pi/n)=e^{2\pi i/n}+e^{-2\pi i/n}$ were rational, it would be fixed under the automorphism of $\mathbb Q[e^{2\pi i/n}]$ that sends $e^{2\pi i/n}$ to $e^{k2\pi i/n}$ for $\gcd(k,n)=1$, i.e., we would have $$\cos(2\pi/n)=\cos(2k\pi/n)$$ if $\gcd(k,n)=1$. If $n>4$ there are such angles with different cosine (to be seen e.g. by drawing a unit circle, or take $k=2$ or $n/2-1$ according to as $n$ is odd or even).
We conclude that $\cos(2\pi/n)$ is irrational for $n>4$.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .