Explicit zero-free regions for automorphic L L -functions

Steven Creech, Alia Hamieh, Simran Khunger, Kaneenika Sinha, Jakob Streipel, Kin Ming Tsang · 2025

Read Explicit zero-free regions for automorphic L L -functions by Steven Creech, Alia Hamieh, Simran Khunger, Kaneenika Sinha, Jakob Streipel, Kin Ming Tsang (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-20873.txt.

Introduction

L-functions encode deep arithmetic information, and knowing where they cannot vanish constrains prime numbers. Steven Creech and five co-authors prove a new explicit zero-free region in a 2025 number theory paper. Their L-function comes from a newform of even weight, squarefree level, and trivial character.

The authors give a precise region near the line of real part one where the L-function never vanishes. The boundary depends on the weight, the level, and the height up the critical strip. Their constants improve sharply on an earlier admissible value from Hoey and co-authors. Explicit regions, rather than existence statements, are what applications need. The result feeds into effective bounds in analytic number theory, where every constant counts.

Catalog record

CREATORSteven Creech, Alia Hamieh, Simran Khunger, Kaneenika Sinha, Jakob Streipel, Kin Ming Tsang
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE74 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-20873

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Explicit zero-free regions for automorphic L L -functions Steven Creech, Alia Hamieh, Simran Khunger, Kaneenika Sinha, Jakob Streipel, Kin Ming Tsang Abstract . Let L ⁡ ( s , f ) L(s,f) be the L L -function associated with a newform f f of even weight k k , squarefree level N N and trivial nebentypus. In this paper, we establish a new explicit zero-free region for L ⁡ ( s , f ) L(s,f) . More precisely, we prove that L ⁡ ( s , f ) L(s,f) does not vanish in the region Re ⁡ ( s ) ≥ 1 − 1 C ​ log ⁡ ( k ​ N ​ max ⁡ ( 1 , | Im ⁡ ( s ) | ) ) \operatorname{Re}(s)\geq 1-\frac{1}{C\log(kN\max(1,\lvert\operatorname{Im}(s)\rvert))} with C = 16.7053 C=16.7053 if | Im ⁡ ( s ) | ≥ 1 \lvert\operatorname{Im}(s)\rvert\geq 1 or | Im ⁡ ( s ) | ≤ 0.30992 log ⁡ ( k ​ N ) \lvert\operatorname{Im}(s)\rvert\leq\frac{0.30992}{\log(kN)} and C = 16.9309 C=16.9309 if 0.30992 log ⁡ ( k ​ N ) < | Im ⁡ ( s ) | ≤ 1 \frac{0.30992}{\log(kN)}<\lvert\operatorname{Im}(s)\rvert\leq 1 . This improves a result of Hoey et al. where 445.994 445.994 was shown to be an admissible value for C C . 2010 Mathematics Subject Classification 11M41, 11M26 (primary), 11F11, 11F66,11F67 (secondary) 1. Introduction Let k and N be positive integers with k even and N squarefree. Let f denote a Hecke newform of weight k with respect to the congruence subgroup \Gamma_{0}(N)=\left\{\begin{bmatrix}a&b\\ c&d\end{bmatrix}\in\mathrm{SL}_{2}(\mathbb{Z}):\;N\mid c\right\} . It is known that f admits a Fourier expansion of the form f(z)=\sum_{n=1}^{\infty}{n^{\frac{k-1}{2}}}\lambda_{f}(n)q^{n},\qquad q=e^{2\pi iz}. Here, \lambda_{f}(1)=1 , and for each n\in\mathbb{N} , we have \frac{T_{n}(f(z))}{n^{\frac{k-1}{2}}}=\lambda_{f}(n)f(z), where T_{n} denotes the n -th Hecke operator acting on the space S_{k}(N) of cusp forms of weight k with respect to \Gamma_{0}(N) . The standard L -function L(s,f) corresponding to f is defined as L(s,f)=\sum_{n=1}^{\infty}\frac{\lambda_{f}(n)}{n^{s}}, which converges absolutely for \operatorname{Re}(s)>1 . L(s,f) has an Euler product representation of the form L(s,f)=\prod_{p}\left(1-\frac{\lambda_{f}(p)}{p^{s}}+\frac{\chi_{0}(p)}{p^{2s}}\right)^{-1},\qquad\operatorname{Re}(s)>1, where \chi_{0}(p)=1 if p\nmid N and \chi_{0}(p)=0 if p\mid N . It is well-known that L(s,f) can be analytically continued to an entire function on the complex plane, and L(s,f)\neq 0 when \operatorname{Re}(s)>1 . This leads to the study of zero-free regions of L(s,f) in the vertical strip 0\leq\operatorname{Re}(s)\leq 1 . In the case of the Riemann zeta function and the Dirichlet L -functions, a delineation of explicit zero-free regions combined with explicit zero density estimates leads to vital applications such as effective error terms in the prime number theorem, its analogue for primes in arithmetic progressions and the average orders of several arithmetic functions. The last two decades have witnessed a surge of activity in this direction, most of which aimed at sharpening the results of the 60-year old paper of Rosser and Schoenfeld [ RS62 ] . Among the many recent contributions to these efforts, we mention [ For02 ] , [ Kad05 ] , [ Kad12 ] , [ Tru14 ] , [ Ram16 ] , [ Pla16 ] , [ Kad18 ] , [ KNW19 ] , [ BKL+21 ] , [ PT21 ] , [ BMOR21 ] , [ Kha24 ] , [ Bel24 ] , and [ DGL+25 ] . In general, explicit determination of zero-free regions of L -functions associated with an arithmetic or algebraic structure helps us better understand the nature and properties of this structure. For example, the zero-free regions for L(s,f) and the symmetric power L -functions L(s,\operatorname{Sym}^{m}f) for m\geq 2 (see Section 2.2 for the definition) yield effective estimates for the sums \sum_{p\leq x}\lambda_{f}(p^{m}) . Analogous to the fundamental ideas in the proof of the prime number theorem, wider zero-free regions for L(s,\operatorname{Sym}^{m}f) lead to sharper estimates for these sums. These estimates further lead to effective error terms in the Sato-Tate distribution law for the Hecke eigenvalues \{\lambda_{f}(p)\}_{p\to\infty}\subset[-2,2] of a fixed non-CM Hecke newform f (see, for example, [ Mur85 ] , [ RT17 ] , [ Tho21 ] and [ HIJTS22 ] ). The first zero-free region result for modular forms beyond the classical context of Dirichlet L -functions and their extensions to number fields is due to Moreno [ Mor72 ] who established a zero-free region for the L -function associated with the Ramanujan delta function \Delta(z)=\sum_{n=1}^{\infty}\tau(n)q^{n}\in S_{12}(1) . This zero-free region took the form (1) \operatorname{Re}(s)\geq 1-\frac{1}{c\log(\lvert\operatorname{Im}(s)\rvert+2)}, for some constant c>0 . Using ( 1 ), Moreno derived the estimate \sum_{p\leq x}\tau(p)p^{-\frac{11}{2}}\log p=\operatorname{O}\left(xe^{-A(\log x)^{1/2}}\right), where A is a suitably chosen positive constant, thereby improving Rankin’s estimates in [ Ran39 ] and [ Ran40 ] ). A generalization of Moreno’s zero-free region to all newforms of level N and weight k\geq 1 is given in [ IK04 , Theorem 5.39] as follows: there exists an absolute constant c>0 such that L(s,f) does not have any zeros in the region \operatorname{Re}(s)\geq 1-\frac{1}{c\log(N(\lvert\operatorname{Im}(s)\rvert+k+3))}, except possibly one simple real zero, called the exceptional zero, in which case f is self-dual. We know now that this exceptional zero can be provably eliminated thanks to the work of Hoffstein and Ramakrishnan [ HR95 ] on the non-existence of Siegel zeros for L -functions of cusp forms on \mathrm{GL}(2) . For a newform f of squarefree level N and weight k , Thorner [ Tho21 ] derived an unconditional estimate for \sum_{p\leq x}\lambda_{f}(p^{m})\log p . An essential ingredient in Thorner’s work is the formulation of zero-free regions for L(s,\operatorname{Sym}^{m}f) . Thorner shows that there exists an absolute constant c>0 such that L(s,\operatorname{Sym}^{m}f) does not have any zeros in the region \operatorname{Re}(s)\geq 1-\frac{1}{c\log(\mathfrak{q}(\operatorname{Sym}^{m}f)(3+\lvert\operatorname{Im}(s)\rvert))}, where \mathfrak{q}(\operatorname{Sym}^{m}f) denotes the analytic conductor of L(s,\operatorname{Sym}^{m}f) . The “state-of-the-art” result in this direction, which also provides explicit constants in the determination of zero-free regions is the following: Theorem ( [ HIJTS22 ] ) . Let N be a squarefree positive integer, and let k be an even positive integer. For a newform f of weight k with respect to \Gamma_{0}(N) , let \beta_{m}+it be a zero of L(s,\operatorname{Sym}^{m}f) such that \beta_{m}>1/2 . Then, \beta_{m}\leq 1-\frac{1}{C_{m}\log(1.214(m+7)N(k-1)\sqrt{1+t^{2}})}, where C_{m}=\frac{(m+7)^{2}}{2(2-\sqrt{3})^{2}}. In particular, if m=1 , let \beta+it be a zero of L(s,f) such that \beta>1/2 . Then, \beta\leq 1-\frac{1}{C\log(9.712N(k-1)\sqrt{1+t^{2}})}, where (2) C=\frac{64}{2(2-\sqrt{3})^{2}}<445.994. The general analytic framework applied to obtain zero-free regions for L(s,\operatorname{Sym}^{m}f) issues from the property that for each m\geq 1 , L(s,\operatorname{Sym}^{m}f) can be analytically continued to the entire complex plane and has a functional equation. This follows from the fact that L(s,\operatorname{Sym}^{m}f) is the L -function of a cuspidal automorphic representation on \textrm{GL}_{m+1}(\mathbb{A}_{\mathbb{Q}}) , proved by Newton and Thorne [ NT21a , NT21b ] . We note here that [ HIJTS22 ] provides explicit constants in the zero-free region of L(s,\operatorname{Sym}^{m}f) for any positive integer N . Let q(\operatorname{Sym}^{n}f) denote the arithmetic conductor of L(s,\operatorname{Sym}^{n}f) , and let Q be a positive constant such that q(\operatorname{Sym}^{n}f)\leq Q^{n+1} for all 1\leq n\leq 2m+1 . Then, L(s,\operatorname{Sym}^{m}f)\neq 0 in the region \operatorname{Re}(s)\geq 1-\frac{2(2-\sqrt{3})^{2}}{(m+7)^{2}\log\left(\frac{2}{\sqrt{e}}(k-1)Q(m+7)\sqrt{1+\lvert\operatorname{Im}(s)\rvert^{2}}\right)}. If N is squarefree, then q(\operatorname{Sym}^{m}f)=N^{m} for each m\geq 1 . However, such closed form expressions are not available for non-squarefree values of N (see [ HIJTS22 , Remark 2.1] ). The goal of the current article is to further improve the constant C given in ( 2 ) in the zero-free region

Read the complete text