Iterating the parity step: `Q (2jm) ≡ (2 P m ^ 2) ^ j` modulo `Q m`.
ProvedCatalog.Novelty.PellSpine.pellQ_even_multiple_modEqaether-catalognovelty
Iterating the parity step: Q (2jm) ≡ (2 P m ^ 2) ^ j modulo Q m.
theorem Catalog.Novelty.PellSpine.pellQ_even_multiple_modEq(m j : ℕ) :
pellQ (2 * j * m) ≡ (2 * pellP m ^ 2) ^ j [MOD pellQ m] := by sorry
Formalization Note Transplanted verbatim from the Aether Catalog source Novelty/PellSpineCompanionDivisibility.lean; the statement is byte-identical to the source declaration, elaborated with autoImplicit disabled in the platform environment.
Preamble
-- Thm stub generated from Novelty/PellSpineCompanionDivisibility.lean import Mathlib import Definitions.Def_Novelty_PellSpineCore /- # The companion divisibility law on the Pell spine `Novelty.PellSpineDivisibility` shows that the Pell numbers `P` form a strong divisibility sequence while the half-companion sequence `Q` does **not** (`gcd (Q 3) (Q 6) = 1`). That refutation leaves the real question open: *exactly when* does `Q m` divide `Q n`? This file answers it completely. For every `m ≥ 2`, `Q m ∣ Q n ↔ n = m * k for some odd k`, a parity-graded divisibility law with no analogue among the `P`'s. The proof runs in two independent halves: * **index step** — `Q m ∣ P (2m)` and `Q n ∣ P (2n)` push the hypothesis into the strong divisibility law for `P`, forcing `Q m ∣ P (2 gcd(m,n))`; if `gcd(m,n) < m` the divisor exceeds the dividend, so `m ∣ n`; * **parity step** — modulo `Q m` the companion sequence satisfies the two-step recursion `Q (a + 2m) ≡ 2 P m ^ 2 * Q a`, so `Q (2jm) ≡ (2 P m ^ 2) ^ j`, a unit mod `Q m` because `Q m` is odd and coprime to `P m`. Even multiples are therefore ruled out. ## Proved * `pellQ_odd`, `pellQ_coprime_two_pellP_sq` — the arithmetic units used by the parity step; * `pellQ_add_two_mul_modEq` — `Q (a + 2m) ≡ 2 P m ^ 2 * Q a [MOD Q m]`; * `pellQ_even_multiple_modEq` — `Q (2jm) ≡ (2 P m ^ 2) ^ j [MOD Q m]`; * `pellQ_coprime_even_multiple` — `gcd (Q m) (Q (2jm)) = 1`; * `pellQ_dvd_index_dvd` — `Q m ∣ Q n → m ∣ n` for `m ≥ 2`; * `pellQ_dvd_iff` — **the companion divisibility law**; * `pellQ_gcd_dvd`, `pellQ_gcd_eq_one_of_even_quotient`, `pellQ_gcd_law` — **the companion gcd law**: `gcd (Q m) (Q n) = Q (gcd m n)` when both index quotients are odd, and `1` otherwise, with no side condition on `m` and `n`; * `pellQ_gcd_of_odd_quotients` — the graded gcd statement that survives the refutation; * summation identities `pellQ_sum`, `pellP_sum`, `pellP_sq_sum` tying the two strands of the spine together. ## Refuted * `not_pellQ_dvd_all_multiples` — `Q n ∣ Q (kn)` fails for even `k`: `Q 2 = 3 ∤ 17 = Q 4`; * `not_pellQ_dvd_iff_index_dvd` — divisibility of indices is **not** sufficient, by the same pair, so the parity grading in `pellQ_dvd_iff` cannot be dropped. -/ open Catalog.Novelty.PellSpine open Finset /-! ## Arithmetic units modulo `Q m` -/ /-! ## The parity step -/
Formal statement
theorem Catalog.Novelty.PellSpine.pellQ_even_multiple_modEq(m j : ℕ) :
pellQ (2 * j * m) ≡ (2 * pellP m ^ 2) ^ j [MOD pellQ m] := by sorrySource