[Paper Review] Jacob's ladders and the tangent law for short parts of the Hardy-Littlewood integral

Jan Moser|arXiv (Cornell University)|Jun 3, 2009

Analytic Number Theory Research7 references18 citations

TL;DR

This paper introduces a novel geometric approach using Jacob's ladders—nonlinear integral solutions—to derive almost-exact asymptotic formulas for short and microscopic parts of the Hardy-Littlewood integral ∫_T^{T+U} Z²(t) dt. By exploiting the chord angle α(T,U) of the curve y = ½φ(T), the method yields a tangent law expression: ∫_T^{T+U} Z²(t) dt = (U ln(φ(T)/2) − aU) tan[α(T,U)] + O(1/T^{1/3−4ε}), which cannot be derived via traditional trigonometric sum methods and features an error term vanishing as T → ∞.

ABSTRACT

The elementary geometric properties of the Jacob's ladders \cite{7} lead to a class of new formulae for short parts of the Hardy-Littlewood integral. This class of formulae cannot be obtained by methods of Balasubramanian, Heath-Brown and Ivic.

Motivation & Objective

• To develop a new method for estimating short and microscopic intervals ∫_T^{T+U} Z²(t) dt where 0 < U ≤ T^{1/3+ε}, beyond the reach of Balasubramanian, Heath-Brown, and Ivić’s methods.
• To establish an almost-exact formula for the Hardy-Littlewood integral using Jacob’s ladders, with an error term that vanishes as T → ∞, contrasting with the unbounded error in existing asymptotic expansions.
• To demonstrate that geometric properties of Jacob’s ladders—specifically chord angles—yield a new tangent law for the integral, providing a novel structural insight into the distribution of Z(t) values.
• To challenge the limitations of classical trigonometric sum techniques by showing that the new method produces results inaccessible through those approaches, particularly for U < 1.
• To provide a geometric interpretation of the Hardy-Littlewood integral’s behavior on short intervals, using the angle α(T,U) of the chord connecting points on the Jacob’s ladder curve.

Proposed method

• The method relies on the nonlinear integral equation (1.2): ∫₀^{μ[x(T)]} Z²(t)e^{−2t/x(T)} dt = ∫₀^T Z²(t) dt, whose solutions are Jacob’s ladders φ(T).
• The core formula (1.3) expresses the full Hardy-Littlewood integral as ∫₀^T Z²(t) dt = (φ(T)/2) ln(φ(T)/2) + (c − ln(2π))φ(T)/2 + c₀ + O(ln T / T), with φ(T) as the Jacob’s ladder.
• For short intervals, the paper derives (2.1): ∫_T^{T+U} Z²(t) dt = (U ln(φ(T)/2) − aU) tan[α(T,U)] + O(1/T^{1/3−4ε}), where α(T,U) is the angle of the chord joining (T, ½φ(T)) and (T+U, ½φ(T+U)).
• The derivation uses the difference φ(T+U) − φ(T) = 2U tan[α(T,U)], linking geometric chord properties to the integral’s asymptotic behavior.
• The method avoids trigonometric sums and instead uses elementary geometric properties of the Jacob’s ladder curve, enabling analysis of intervals with U < 1.
• Asymptotic estimates like tan[α(T,U₀)] = 1 − (1−c)/ln T + O(1/ln²T) are derived from the Balasubramanian formula, linking geometric and analytic structures.

Experimental results

Research questions

• RQ1 Can short and microscopic parts of the Hardy-Littlewood integral ∫_T^{T+U} Z²(t) dt be estimated beyond the reach of Balasubramanian, Heath-Brown, and Ivić’s methods?
• RQ2 Does the geometric structure of Jacob’s ladders—specifically chord angles—yield a new asymptotic law for short integrals?
• RQ3 Can an almost-exact formula for ∫₀^T Z²(t) dt be derived independently of trigonometric sum methods, with an error term vanishing as T → ∞?
• RQ4 What is the geometric meaning of the angle α(T,U) in the context of the Hardy-Littlewood integral, and how does it relate to the distribution of Z(t)?
• RQ5 To what extent does the tangent law (2.1) represent a fundamental structural feature of the Z-function on short intervals, and how does it compare to classical estimates?

Key findings

• The paper establishes a new tangent law: ∫_T^{T+U} Z²(t) dt = (U ln(φ(T)/2) − aU) tan[α(T,U)] + O(1/T^{1/3−4ε}), where α(T,U) is the chord angle of the Jacob’s ladder curve, providing a geometric asymptotic formula.
• The error term in the almost-exact formula (1.3) is O(ln T / T), which tends to zero as T → ∞, unlike the unbounded O(T^{1/3+ε}) error in Balasubramanian’s formula.
• For U = U₀ = T^{1/3+2ε}, the tangent of the chord angle satisfies tan[α(T,U₀)] = 1 − (1−c)/ln T + O(1/ln²T), linking geometric and analytic structures.
• The method produces results unattainable by trigonometric sum techniques, particularly for intervals with U < 1, where existing methods fail.
• The chord-based formula (2.8) holds for a continuum of intervals [N,M] ⊂ [T, T+T^{1/3+2ε}], with error O((M−N)/ln T) + O(1/γ^{1/3−4ε}) for γ ≤ N < M ≤ ρ̄.
• For a fixed angle α = π/6, the formula ∫_γ^{γ+U(π/6)} Z²(t) dt = (1/√3)(U ln γ − aU) + O(U/ln γ) + O(1/γ^{1/3−4ε}) is derived, demonstrating explicit geometric control.