How close can one come to proving that there are infinitely many primes, $p$ and $q$, such that $q = 4p + 1$?

The idea for this question came from reading the question and answers posed by user39898, Quadratic reciprocity and proving a number is a primitive root, URL (version: 2012-12-12): Quadratic reciprocity and proving a number is a primitive root

Based on that result if there are infinitely many pairs of such primes then $2$ is a primitive root modulo infinity many primes.

This would relate to Artin's conjecture on primitive roots: https://en.wikipedia.org/wiki/Artin%27s_conjecture_on_primitive_roots

Because that conjecture has not been proven, all I am looking for is how close we are to proving that such a set of primes is infinite or suggestions on how to proceed. I don't know how to begin answering it.

  • 1
    $\begingroup$ No-one knows. Probably, this will never be known. $\endgroup$
    – Will Jagy
    Commented Feb 21, 2016 at 5:24
  • $\begingroup$ I am aware of that. What do we know? What is the largest pair found so far? $\endgroup$ Commented Feb 21, 2016 at 5:26
  • 2
    $\begingroup$ oeis.org/A023212 $\endgroup$
    – Will Jagy
    Commented Feb 21, 2016 at 5:32
  • $\begingroup$ @Dietrich Burde Yes, that is true, but that doesn't mean nothing is known about the question. For example, a Sophie Germain prime cannot be one of the $p$ in this pair for $p > 3$ as mentioned in the the OEIS list A023212 that Will Jagy linked to. $\endgroup$ Commented Feb 21, 2016 at 13:54

1 Answer 1


The tuple $(x, 4x+1)$ is an admissible $2$-tuple in the language of the $k$-tuples conjecture of the conjectures of Hardy and Littlewood. It is believed that there are infinitely many $x$ such that $x$ and $4x+1$ are simultaneously prime. More generally, given an admissible $k$-tuple, it is beleived that there are infinitely many $x$ such that all $k$ elements of the $k$-tuple are simultaneously prime.

We do not know how to prove this. And before a few years ago, there was essentially no major progress towards the $k$-tuple conjecture.

The recent works of Zhang-Maynard-Tao-Polymath8 have made major progress towards a weakened version of the $k$-tuple conjecture. Namely, for any $k$, it has been shown that there exists some $n = n(k)$ (meaning that it depends on $k$) such that for any admissible $n$-tuple, there are infinitely many $x$ making $k$ elements of the $n$-tuple simultaneously prime.

For $2$-tuples, the number is on the order of $250$. What this means is that if one considers a $250$-tuple (like $(x, 4x+1, \ldots, \text{lots of other pieces})$) then there are infinitely many $x$ such that one of the pairs are simultaneously prime. Under the most ideal circumstances, one might hope to reduce the number $250$ to something smaller, but the current methods of proof are not capable of reducing this number all the way down to $2$. Similarly, one might hope to isolate a particular pair, but there is no known way towards this either.

So this is the current state of the art, and there is no current widely-known method or idea of significantly improving this. It should take a genuine new idea (or more likely several new ideas) for progress to be made. I would not be surprised if it took many decades for further significant progress.


You must log in to answer this question.

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