← All problems

Erdős Problem #203 — Composite shifts of 2–3-smooth multiples

46 years old · traced to

Does there exist an integer m≥1m\geq 1 with gcd⁡(m,6)=1\gcd(m,6)=1 such that, for every pair of nonnegative integers k,ℓk,\ell, the number 2k3ℓm+12^k3^\ell m+1 is not prime? Equivalently, does there exist such an mm for which

∀k,ℓ∈Z≥0,2k3ℓm+1 is composite?\forall k,\ell\in\mathbb Z_{\geq 0},\qquad 2^k3^\ell m+1\text{ is composite}?

Every term is greater than 11, so “not prime” is equivalent to “composite.” The pair (k,ℓ)=(0,0)(k,\ell)=(0,0) is included and requires m+1m+1 to be composite. Erdős Problems #203

Background

This question concerns a two-exponent extension of the Sierpiński-number problem. A standard way to construct coefficients for which a family of exponential expressions is always composite is to find finitely many primes such that each exponent is covered by a congruence class on which one of those primes divides the expression. For a single power, the divisibility classes arise from the multiplicative orders of the base modulo the covering primes. The mixed family here has two independently varying exponents, so a construction for one exponent alone does not provide the required coverage. Izotov (1995) Erdős Problems #203

The classical one-base result is due to Sierpiński, who proved in 1960 that infinitely many positive odd coefficients aa have a2n+1a2^n+1 composite for every n>0n>0. The problem page attributes the present question to an Erdős–Graham reference and notes broader variants with products of powers of distinct primes. Its citation is “ErGr80, p. 27”; the separate 1980 Erdős survey is not that reference. Izotov (1995) Erdős Problems #203

References

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.