Contour representation of Zudilin's concrete linear forms
ProvedZudilinZeta.params13_contour_formulacomplex-analysiscontour-integrationzeta-values
Let be Zudilin's linear form with , , , , and for . Let and be the reflected Gamma kernel and vertical integral defined in the accompanying definition. For every integer and every complex satisfying ,
Here uses the rational function in Zudilin's note. The identity connects these arithmetic linear forms to the explicit integral for which a saddle asymptotic is available. There is no restriction on ; translating it changes only the parameterization of the vertical line.
Preamble
import Definitions.Def_ZudilinZetaContourKernel set_option autoImplicit false
Formal statement
theorem ZudilinZeta.params13_contour_formula (τ : ℂ) (hτ : 87 ≤ τ.re ∧ τ.re ≤ 175 / 2) (n : ℕ) (hn : 2 ≤ n) :
|ZudilinZeta.F ZudilinZeta.params13 n| = (n : ℝ) / (2 * Real.pi) *
|(ZudilinZeta.params13KernelIntegral n τ).re| := by sorrySource
Derived specialization of W. Zudilin, One of the numbers zeta(5), zeta(7), zeta(9), zeta(11) is irrational, Russian Math. Surveys 56:4 (2001), pp.774-775, the displayed rational function and equation (2), and the parameter choice on p.775, https://www.math.ru.nl/~zudilin/PS/zeta5-11%24.pdf. The contour argument is adapted from W. Zudilin, Irrationality of values of the Riemann zeta function, Izvestiya Math.66:3 (2002), Lemmas 2.3-2.4, pp.497-499, especially (2.5), (2.8)-(2.10), https://www.math.ru.nl/~zudilin/PS/zete_main.pdf. The present eta-parameter kernel and absolute-value normalization are a derived specialization, not a verbatim statement of Lemma 2.4.