GCD Structure, Primitive Divisors, and Prime Densities in a Dyadic Quadratic Sequence
Author’s Original Manuscript (preprint) submitted to The Fibonacci Quarterly (2026). Abstract. Let a_n = 2^(2n+1) + 2^n + 1 = f(2^n), where f(X) = 2X² + X + 1, and let K = Q(√−7). We give a unified account of three arithmetic layers for this sequence. First, quadratic scaling yields an exact same-root/conjugate-root decomposition of fixed-gap gcds; every odd prime-power divisor returns in exactly two exponent classes, and a graph form of the generalized Chinese remainder theorem counts the compatible classes for composite divisors. Second, multiplicative order gives an exact primitive-divisor criterion and, through paired-root occupancy, an unconditional lower bound liminf ω(∏_{n≤J} a_n)/J ≥ 1/4. A shifted-product argument further forces a primitive-bearing exponent after every N, with n_next(N) ≤ (21/10 + o(1))N²/log₂N. Third, writing η = (−1 + √−7)/2 and I_p = (p − 1)/ord_p(2) for split primes p, a relative Kummer-degree calculation yields the fixed-index quotient-coset law and, under the GRH hypothesis required by the Index Map theorem, an activity density equal to (6519/6700) times Stephens’ constant among primes splitting in K. The corresponding density among all rational primes is (6519/13400) times Stephens’ constant. The paper distinguishes unconditional arithmetic theorems from GRH-conditional density statements and leaves the universal termwise primitive-divisor conjecture open.
Authors
- Gerald Wayne Scanlon (ORCID: https://orcid.org/0009-0009-7644-8550)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-20
- DOI
- https://doi.org/10.5281/zenodo.24383342
- Primary Topic
- Analytic Number Theory Research
- Type
- preprint