Arithmetic Fluctuations and the Critical Strip: A New Look at the Mertens Function

Introduction

The distribution of prime numbers is fundamentally linked to the zeros of the Riemann zeta function, ζ(s). While traditional mathematical approaches focus on the growth and zeros of the zeta function itself, an alternative and equally deep perspective is offered by the study of its reciprocal. As investigated in the paper arXiv:hal-01667383, the properties of this reciprocal function are deeply encoded in the Möbius function μ(n) and its summatory counterpart, the Mertens function M(n). This article synthesizes the technical arguments connecting arithmetic sum bounds to the geometry of the critical strip.

The core motivation for this research is to establish a rigorous bridge between discrete arithmetic sums and continuous analytic continuation. The Riemann Hypothesis asserts that all non-trivial zeros of ζ(s) satisfy the condition that their real part is exactly 1/2. By examining the behavior of the Mertens function, researchers can translate this complex analytic statement into a concrete bound on a computable arithmetic function. This analysis explores how the convergence of the Dirichlet series for 1/ζ(s) in the region where the real part of s is greater than 1/2 would necessitate the truth of the hypothesis.

Mathematical Background

The Riemann zeta function is initially defined for complex numbers s with a real part greater than 1 by the absolutely convergent Dirichlet series where ζ(s) is the sum of 1/n^s for all positive integers n. In the same region, the function can be expressed as an infinite product over prime numbers, known as the Euler product. The reciprocal of the zeta function can be represented as a series involving the Möbius function: 1/ζ(s) = Σ μ(n)/n^s. The Möbius function μ(n) is defined as 1 if n is a square-free integer with an even number of prime factors, -1 if n is square-free with an odd number of prime factors, and 0 if n has a squared prime factor.

The Mertens function M(n) is the summatory function of the Möbius function, defined as the sum of μ(k) for k from 1 to n. A critical tool in the analytic continuation of these objects is the Dirichlet eta function, η(s), also known as the alternating zeta function. The relationship η(s) = (1 - 2^1-s)ζ(s) allows for the study of the zeta function through an entire function that lacks the pole at s = 1. The work in arXiv:hal-01667383 focuses on the proposition that if the absolute value of M(n) is bounded by the square root of n, then the non-trivial zeros must lie on the critical line.

Main Technical Analysis

Analytic Continuation of the Reciprocal Zeta Function

The technical argument relies on extending the domain of 1/ζ(s). For Re(s) > 1, the series converges absolutely. To move toward the critical line, we utilize the integral representation: 1/ζ(s) = s ∫ M(u)/u^s+1 du, where the integral is taken from 1 to infinity. This representation is valid initially for Re(s) > 1. However, if we can establish a growth bound on the Mertens function of the form |M(u)| < C u^1/2+ε for any positive ε, the integral becomes convergent for Re(s) > 1/2.

This convergence is significant because any zero of the zeta function in the region Re(s) > 1/2 would constitute a singularity for the reciprocal function. If the reciprocal function can be shown to be analytic in this region via the convergence of the Mertens integral, then no such zeros can exist. Because the functional equation of the zeta function relates values at s to values at 1-s, the absence of zeros to the right of the critical line implies their absence to the left, thereby confining all non-trivial zeros to the line Re(s) = 1/2.

The Role of Symmetry and Functional Equations

The symmetry properties of the complex plane play a vital role in this derivation. If ρ is a non-trivial zero, then 1 - ρ, the conjugate of ρ, and 1 minus the conjugate of ρ are also zeros. This four-way symmetry ensures that establishing analyticity for the reciprocal function in the right half of the critical strip is sufficient to prove the Riemann Hypothesis. The paper arXiv:hal-01667383 emphasizes that the discrete behavior of M(n), which only changes by increments of 0, 1, or -1, constrains the growth of the function in a way that is consistent with the square root boundary.

Novel Research Pathways

Pathway 1: Smoothed Mertens Transforms

A promising direction for future investigation involves the use of smoothed Mertens transforms. Pointwise estimates of the Mertens function are subject to significant oscillations, which famously led to the disproof of the original Mertens conjecture. However, by considering weighted sums of the form Σ μ(n) w(n/X), where w is a smooth kernel, researchers can obtain more stable analytic continuations. These smoothed transforms are less sensitive to local fluctuations and may provide a more robust path to establishing zero-free regions.

Pathway 2: Spectral Density of Möbius Fluctuations

The fluctuations of the Möbius function can be viewed as a pseudo-random walk. Following the logic in arXiv:hal-01667383, one could apply methods from signal processing and spectral density analysis to the sequence of μ(n). If the power spectrum of this sequence can be shown to be bounded in a manner corresponding to the square root law of random walks, it would provide probabilistic evidence for the growth rates required by the Riemann Hypothesis. This connects the arithmetic noise of prime distribution to the spectral properties of the zeta zeros.

Computational Implementation

(* Section: Mertens Function and Zeta Zero Analysis *)
(* Purpose: This code computes the Mertens function M(n) and explores its relationship with the zeros of the zeta function on the critical line. *)

Module[{nMax = 2000, muValues, mertensValues, sqrtBounds, zeros, tRange, absZetaReciprocal},
  muValues = Table[MoebiusMu[k], {k, 1, nMax}];
  mertensValues = Accumulate[muValues];
  sqrtBounds = Table[{k, Sqrt[k]}, {k, 1, nMax}];
  
  (* Compute first few zeros of the Zeta function *)
  zeros = Table[ZetaZero[j], {j, 1, 10}];
  
  (* Sample the reciprocal of the Zeta function along the critical line *)
  tRange = Table[t, {t, 0, 40, 0.1}];
  absZetaReciprocal = Table[{t, Abs[1/Zeta[1/2 + I*t]]}, {t, tRange}];
  
  (* Output plots *)
  {ListLinePlot[{mertensValues, Table[Sqrt[k], {k, 1, nMax}], Table[-Sqrt[k], {k, 1, nMax}]}, 
    PlotLabel -> "Mertens Function vs Sqrt(n)", 
    AxesLabel -> {"n", "M(n)"},
    PlotStyle -> {Blue, Red, Red}],
   ListLinePlot[absZetaReciprocal, 
    PlotLabel -> "Abs[1/Zeta(1/2 + it)]", 
    AxesLabel -> {"t", "Magnitude"}]}
]

In conclusion, the investigation into the Mertens function and its relationship to the reciprocal zeta function provides a powerful framework for addressing the Riemann Hypothesis. While the original Mertens conjecture has been disproven in its strongest pointwise form, the underlying logic connecting arithmetic growth to analytic continuation remains a cornerstone of modern number theory. Future research should focus on refining the bounds of Möbius sums and exploring smoothed variants of the Mertens function. By combining these arithmetic methods with the spectral properties of the zeta function, we may eventually uncover the definitive proof of the zero distribution on the critical line. The path forward requires a synthesis of discrete prime dynamics and continuous complex analysis.

References

• M. Sghiar, "The Mertens function and the proof of the Riemann's hypothesis," arXiv:hal-01667383
• A. M. Odlyzko and H. J. J. te Riele, "Disproof of the Mertens Conjecture," Journal für die reine und angewandte Mathematik, 1985.
• E. C. Titchmarsh, "The Theory of the Riemann Zeta-Function," Oxford University Press.