1

I'm looking for theorems that say something about relations of factors between consecutive numbers. In this case relations between greatest primefactors of consecutive numbers. I've tested for $n<10,000$ but find no $n$ such that $7$ is the greatest prime factor of $2^n+1$.

I've found out that given two different primes $p,q$ there is always a natural number $n$ such that $p|n$ and $q|n+1$.

So I would like to find a number $n$ so that $7$ is the greatest prime factor of $2^n+1$ or a proof that there is no such $n$.

Jyrki Lahtonen
  • 140,891
Lehs
  • 14,252
  • 4
  • 28
  • 82

3 Answers3

3

As Jykri Lahtonen says, the most natural thing to do is to look at the congruence $2^n\equiv -1\pmod{7}$.

If you know a bit more about elementary number theory and know things like Fermat's little theorem, primitive generator of $\mathbb{F}_p^*$ or quadratic residues/nonresidues, you can probably interpret the following problem in a more enlightening perspective.

But assuming you don't, you can observe that $2^n$ is only ever $2, 4$ or $1\pmod{7}$ just by listing out the first few exponents. That should tell you what are the possible values of $2^n+1\pmod{7}$ and conclude no such $n$ exists.

daruma
  • 2,661
1

It can be proved using Zsigmondy's theorem that given any $n$, the number $2^n+1$ has a prime factor of the form $2nk+1$ for some $k\in \mathbb N$. So, for your question, you only need to check the cases $n=1,2,3$ by hand. You can find the proof here.

This completes the proof I guess.

Sayan Dutta
  • 10,345
  • Can anybody explain the reason for the downvote? Anything wrong with the answer? – Sayan Dutta Aug 24 '21 at 15:25
  • I think, the only reason for the downvotes (and the later deletion) is that there is a much easier solution, namely that $7\mid 2^n+1$ is not possible at all. The method is therfore an overkill, but the answer nevertheless absolutely correct. – Peter Aug 29 '21 at 10:00
  • @Peter Yes, later when I saw the other answers, I understood that mine was an overkill. So, I agree with you on that. But, can you tell me a different thing... I am quite new to this site and I often don't understand its activities properly. From your comment, I guess my answer was deleted, but how did it come into existance again? As far as I remember, I saw a few downvotes on the day I posted it, and today suddenly, I'm again recieving activities on this post. I don't understand how it happened :( – Sayan Dutta Aug 29 '21 at 12:19
  • The answer was undeleted and has now 2 upvotes. It seems this was not in your interest, but I thought an undeletion (and a later upvote) would make you happy. You can delete the answer , if you do not want its "revival". The theorem has its merit in similar situations, so I think the undeletion was justified. – Peter Aug 29 '21 at 14:09
  • @Peter no no, it's absolutely fine. In fact, I thank you for the undeleting it. I was just a little confused about how this site works. Now, I understand. So, thanks again :) – Sayan Dutta Aug 29 '21 at 15:05
1

To elaborate on daruma's answer,

First observe that any number is in one of the forms: $3n,3n+1$ or $3n+2$. Now $2^3 \equiv 1 \mod 7$ implies $$2^{3n} \equiv 1 \mod 7$$ so that we also have $$2^{3n+1} \equiv 2 \mod 7 $$ and $$ 2^{3n+2} \equiv 4\mod 7$$. So for any $k$ we have $2^k \equiv 1,2,4 \mod 7$

Finally observe that $2^n + 1 \equiv 0\mod 7 $ is same as $2^n \equiv 6\mod 7 $

Note : $2^n \equiv r \mod 7$ is same as the statement $7$ divides $2^n - r$

Infinity_hunter
  • 5,614