# Is $10^n+1$ composite for all $n\in \mathbb{N}$ greater then $2$?

Is $10^n+1$ composite for all $n\in \mathbb{N}$ greater then $2$?

I tried many values of $n$, and $10^n+1$ is composite each time (excpet $n=1,2$).

Is my conjecture correct? Thank you.

-
According to this it is an open question. – WimC Dec 15 '12 at 7:53
101 appears to always divide the result when $n \equiv 2(\mod 4)$ – orlandpm Dec 15 '12 at 7:57
@orlandpm: Yes, because $10^{4n+2}=100^{2n+1}\equiv (-1)^{2n+1}\pmod{101}\equiv -1\pmod{101}$. – Jonas Meyer Dec 15 '12 at 7:59
For odd values of $n$, $11|10^n+1$. – Mike Dec 15 '12 at 8:01

If $n = 2^l m$ where m is odd, then $$\displaystyle (10^{2^l})^m + 1 \equiv 0 \bmod (10^{2^l} + 1)$$.
So the interesting question is if $10^n + 1$ is composite when $n$ is a power of $2$.
Unfortunately I don't know what happens when $n$ is a power of $2$.
Heuristically they are probably all composite. WolframAlpha can verify them all up to $10^{2048}+1$. – Dan Brumleve Dec 15 '12 at 8:12
The proof given in the link can be adapted to $10^{2^n}$ case too. So you are right. However I cant seem to understand in what sense it is a good heuristic... Thanks for that link :) – Isomorphism Dec 15 '12 at 8:18
The heuristic is the prime number theorem which basically says that the probability that $n$ is prime is $\frac{1}{\log{n}}$. So when $n=10^{2^x}+1$ and we sum over all $x \gt 11$ this series converges. – Dan Brumleve Dec 15 '12 at 8:22
Note that this is the exact analogue of the question of when $2^n+1$ is prime: then also $n$ must itself be a power of two, leading to the "Fermat primes". – Greg Martin Dec 15 '12 at 10:23