On a conjecture on shifted primes with large prime factors, II
Yuchen Ding
TL;DR
This paper addresses the distribution of shifted primes with large prime factors by studying the counting function $T_c(x)$, the number of primes $p\le x$ with $P^+(p-1)\ge p^c$. Employing a sieve-based approach, including a reduction to a two-variable sieve bound and Banks–Shparlinski-type estimates, the author proves a quantitative upper bound on the limsup: for $8/9\le c<1$, $\limsup_{x\to\infty} T_c(x)/\pi(x) \le 8\left(\frac{1}{c}-1\right)$. This yields an unconditional Erdős-type bound and clarifies the relationship to Chen–Chen conjectures, while also offering conditional refinements under the Elliott–Halberstam conjecture. The methods combine a sieve framework with Brun–Titchmarsh-type inequalities and the Dickman-related asymptotics to bound shifted-prime counts in terms of $x/\log x$, signaling where potential improvements in constants may appear under stronger hypotheses.
Abstract
Let $\mathcal{P}$ be the set of primes and $π(x)$ the number of primes not exceeding $x$. Let also $P^+(n)$ be the largest prime factor of $n$ with convention $P^+(1)=1$ and $$ T_c(x)=\#\left\{p\le x:p\in \mathcal{P},P^+(p-1)\ge p^c\right\}. $$ Motivated by a 2017 conjecture of Chen and Chen, we show that for any $8/9\le c<1$ $$ \limsup_{x\rightarrow\infty}T_c(x)/π(x)\le 8(1/c-1), $$ which clearly means that $$ \limsup_{x\rightarrow\infty}T_c(x)/π(x)\rightarrow 0, \quad \text{as}~c\rightarrow1. $$
