5
$\begingroup$

Prove for some $n\in\mathbb{N}$ there are infinitely many primes $p$ , S.T the numbers $p-1,p+1,p+2$ have $n$ different prime factors .

Attempt:

by the fundamental theorem of arithmetic's

we know that $$p-1=p_1^{a_1}...p_k^{a_n}$$ also $$p+1=q_1^{a_1}...q_k^{a_n}$$ $$p+2=w_1^{a_1}...w_k^{a_n}$$

Noticing that all the factorizations consists of $n$ different prime factors.

  1. I tried proving a stronger case with adding $p$ and its factorization to achieve 4 consecutive numbers. However that led me no where.
  2. I tried to solve it with the Chinese remainder theorem using the abstract idea that there is only one unique $x_0$ solution. Not sure how to continue with this idea.
$\endgroup$
9
  • $\begingroup$ Couldn't you use the fact that if $gcd(a,b)=1$ there are infinitely many primes of the form $an+b$? $\endgroup$ Commented Dec 21, 2020 at 18:43
  • $\begingroup$ Do you mean "at least $n$"? If $n=1$, your post would appear to require $4$ consecutive primes. $\endgroup$
    – lulu
    Commented Dec 21, 2020 at 18:56
  • $\begingroup$ $(p-1,p+1)\ge2$ if p is odd. $\endgroup$
    – Piquito
    Commented Dec 21, 2020 at 19:04
  • 2
    $\begingroup$ If the claim is "exactly" $n$ then it is false, as the case $n=1$ shows. $\endgroup$
    – lulu
    Commented Dec 21, 2020 at 19:08
  • $\begingroup$ What exactly is the claim. That $p-1,p+1,p+2$ have exactly $n$ factors each? That the total number of factors for then $3$ are $n$? $\endgroup$
    – fleablood
    Commented Dec 21, 2020 at 19:31

1 Answer 1

4
$\begingroup$

As pointed out in the comments: If the question is asking for exactly $n$ factors, then the claim is false for every $n$, for the following simple reason:

If $p \neq 2,$ then $p \equiv 1 \pmod{2},$ thus $p-1$ and $p+1$ are both divisible by $2$, thus can not have different prime factorizations.

However, the statement is true for at least $n$ different prime factors:

Let $p_1,...,p_n, q_1,...,q_n, r_1,...,r_n$ be different odd prime numbers.

Let $P = p_1p_2 \cdots p_n, Q =q_1q_2\cdots q_n$ and $ R= r_1r_2\cdots r_n.$

(1) By the Chinese Remainder Theorem, \begin{align*} x-1 &\equiv 0 &\pmod{P} \\ x+1 &\equiv 0 &\pmod{Q}\\ x+2 &\equiv 0 &\pmod{R} \\ \end{align*} has a solution.

(2) Let $x\in \mathbb{N}$ be a solution of the congruence equations. Note that $x_k = x+ k \cdot PQR $ is also a solution for all $k\in \mathbb{N}$.

(3) Now $(x,PQR) =1$, so by Dirichlet's theorem of primes in arithemetic progression you get the result.

$\endgroup$
5
  • 1
    $\begingroup$ To stress: this argument makes sense if the question is asking for "at least $n$" distinct prime factors. The OP, however, has insisted on "exactly $n$" distinct prime factors (for which the claim is certainly false, at least for $n=1$). $\endgroup$
    – lulu
    Commented Dec 21, 2020 at 19:12
  • 2
    $\begingroup$ @lulu I am pretty sure the question asking for some $n \in \mathbb{N}$ $\endgroup$
    – ATB
    Commented Dec 21, 2020 at 19:18
  • $\begingroup$ Great answer. Much appreciated! $\endgroup$
    – Esty.R
    Commented Dec 21, 2020 at 20:40
  • $\begingroup$ @asemabdelraouf line (3) why $\gcd(x,PQR)=1$ ?? can you explain please ? $\endgroup$
    – ATB
    Commented Dec 22, 2020 at 16:28
  • 1
    $\begingroup$ Assume $p_1 | x, $ then $x = s p_1$ for some $s$. But using the congruence, $ x= 1 + t P$ for some $t,$ thus $ 1 = p_i ( s- t \frac{P}{p_i}),$ which is a contradiction. (Becuase $s- t \frac{P}{p_i}$ is an integer). You can do the same for all the primes dividing $PQR.$ $\endgroup$ Commented Dec 22, 2020 at 16:42

You must log in to answer this question.

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