Bound for the prime-log series on the line 2
ProvedArgZetaTwo.norm_primeLog_learda-rh65k-transfernumber-theoryriemann-zeta
For every real , let be the defined complex prime-log series and the corresponding real prime sum. Then
The bound controls the logarithm of zeta on the right edge.
Preamble
import Definitions.Def_ArdaRH65K_Turing121_ArgZetaTwo
import Definitions.Def_ArdaRH65K_Turing121_DiffractionCore
import Definitions.Def_ArdaRH65K_Turing121_DlvpTheta
import Definitions.Def_ArdaRH65K_Turing121_DlvpZetaRateEffective
import Definitions.Def_ArdaRH65K_Turing121_RHInBoxAnalytic
import Definitions.Def_ArdaRH65K_Turing121_StripRepr
import Definitions.Def_ArdaRH65K_Turing121_StripReprR1
import Mathlib
import Theorems.Thm_ArgZetaTwo_norm_neg_log_one_sub_le
/- ArgZetaTwo.lean -- brick K6a of the ANDURIL Arb discharge
(telperion/docs/ANDURIL_ARB_DISCHARGE_2026-09-23.md, inventory row H2d, brick K6a).
THE BINDER IT DISCHARGES. Conjunct 5 of the `hArbT` binder of every emitted Turing band
(`TuringBand.BandStatement`, `BandGlue.BandData.EnclHyp`) is
hAV2 : DiffractionCore.argChangeVert riemannZeta 2 T0 T1 ∈ Set.Icc L1 H1,
the argument change of ζ up the line `Re s = 2` between the band edges. The emitted bands feed
it a 1e-12-wide Arb enclosure. This file proves a GENERIC enclosure, valid for every `T0`,
`T1` and hence for all 10,379 bands of the h280000 ladder at once:
argChangeVert riemannZeta 2 T0 T1 ∈ [-249/250, 249/250] (`hAV2_generic`)
because `|arg ζ(2 + i t)| ≤ log ζ(2) = log (π²/6) < 0.498` for every real `t`.
HOW.
* `argChangeVert_zeta_two_eq`: the argument change IS a difference of principal arguments,
argChangeVert ζ 2 T0 T1 = arg ζ(2 + i T1) - arg ζ(2 + i T0),
by the fundamental theorem of calculus for `y ↦ log ζ(2 + i y)` (principal `Complex.log`),
legitimate because `ζ(2 + i y)` stays in the right half-plane.
* `arg_zeta_eq_im_primeLog`, `abs_arg_zeta_two_le`: on `Re s = 2`, ζ(s) = exp L(s) with
L(s) = Σ_p -log(1 - p^{-s}) (Mathlib's `riemannZeta_eulerProduct_exp_log`), and
‖L(s)‖ ≤ Σ_p -log(1 - p^{-2}) =: G termwise (Taylor series of `-log(1 - z)` dominated
by that of `-log(1 - ‖z‖)`). Since G < π, `Complex.log (exp L) = L`, so
arg ζ(s) = Im L(s) ∈ [-G, G]; since G < π/2, Re ζ(s) > 0.
* `exp_primeSum`: exp G = ζ(2) = π²/6 (the same Euler product at the real point 2), so
G = log (π²/6); `log_pi_sq_div_six_lt`: log (π²/6) < 249/500 from π < 3.1416 and the
degree-6 Taylor lower bound of `exp`.
This is the "Λ(n) series at σ = 2" bound of the memo: mathematically G = Σ_n Λ(n)/(n² log n) =
log ζ(2); the formal route here is the prime Euler product, not the Λ(n) series itself.
(Not formalized, for orientation only: by Kronecker's theorem the supremum of |arg ζ(2+it)| is
Σ_p arcsin(p⁻²) ≈ 0.455, so the constant is within 10 percent of optimal.)
Trust: no hypotheses beyond those stated; axioms [propext, Classical.choice, Quot.sound]
(see AxiomGuardArgChange.lean). No `sorry`.
conjecture1_proved = False. A height-uniform bound on one edge of a finite zero count; nothing
here bears on the Riemann Hypothesis. -/
open Complex
section
namespace ArgZetaTwo
end ArgZetaTwo
open ArgZetaTwo
/-! ## 1. The logarithm of the Euler product -/
/-! ## 2. The real prime sum `G = Σ_p -log(1 - p^{-2})` -/
/-! ## 3. `|arg ζ(2 + i t)| ≤ G` -/
Formal statement
theorem ArgZetaTwo.norm_primeLog_le (t : ℝ) : ‖primeLog ((2 : ℂ) + (t : ℂ) * I)‖ ≤ primeSum := by sorry
Source