A Couple of Other Connections Between Number Theory and Topology

Following the video I made (link and notes above) about how RH can be restated via topology, I decided to follow it up with a little addendum- two other connections (not as deep) between number theory and topology.

Infinitude of primes

Firstly, one can show the infinitude of primes using topology. More precisely, we will put a topology on \mathbb{Z} and show that there can’t be a finite number of primes. Firstly, let N(a,b)=\{a+nb: n\in\mathbb{Z}\}. Then we define open sets as follows:

Definition:

A set U is open if for every a\in U, there exists b>0 such that N(a,b)\subset U.

Note the following important facts:

  • Any non-empty, open set is infinite.
  • All of the N(a,b)s are clopen.

Now here’s the punchline: since every number is a product of primes, we have that:

\mathbb{Z}\setminus\{\pm 1\}=\bigcup N(0,p)

If there were a finite number of primes, then the union on the right would be closed, which would directly imply that \{\pm 1\} is open in this topology, but this would contradict that all open sets in this topology are open. Therefore, there are infinitely many primes.

A weird way to do a normal sum

Theorem:

\sum_{j=0}^{n}(-1)^j\binom{n}{j}=0.

Proof:

Let X=\left(S_{1}\right)^{n}. By the Künneth formula, H^{j}(X, \mathbb{Q}) has dimension \binom{n}{j}. The Euler characteristic of X is

\sum_{j=0}^{n}(-1)^j dim_{\mathbb{Q}}H^{j}(X; \mathbb{Q})= \sum_{j=0}^{n}(-1)^j\binom{n}{j}.

On the other hand, the Lefschetz fixed-point theorem shows that \chi(X)=0. Therefore,

\sum_{j=0}^{n}(-1)^j\binom{n}{j}=0.

Published by gregoriousmaths

I am an aspiring mathematician and am using this blog to post my mathematical notes on random topics in maths- hopefully you find them helpful!

Leave a comment