Two-divisibility of the coefficients of certain weakly holomorphic modular forms

Darrin Doud,Paul Jenkins,John Lopez
DOI: https://doi.org/10.1007/s11139-011-9331-0
2011-07-30
Abstract:We study a canonical basis for spaces of weakly holomorphic modular forms of weights 12, 16, 18, 20, 22, and 26 on the full modular group. We prove a relation between the Fourier coefficients of modular forms in this canonical basis and a generalized Ramanujan tau-function, and use this to prove that these Fourier coefficients are often highly divisible by 2.
Number Theory
What problem does this paper attempt to address?
The problem that this paper attempts to solve is about the two - divisibility of the Fourier coefficients of weakly holomorphic modular forms of specific weights. Specifically, the author studied a canonical basis of the space of weakly holomorphic modular forms with weights 12, 16, 18, 20, 22, and 26 on the full modular group, and proved the relationship between the Fourier coefficients of these modular forms and the generalized Ramanujan τ - function. Using this relationship, the author further proved that these Fourier coefficients are usually highly divisible by 2. ### Main contributions of the paper: 1. **Construction of the canonical basis**: The author constructed a canonical basis of the space of weakly holomorphic modular forms of specific weights. 2. **Relationship between Fourier coefficients and τ - function**: The author proved the relationship between the Fourier coefficients of these modular forms and the generalized Ramanujan τ - function. 3. **Proof of two - divisibility**: Through the above relationship, the author proved that these Fourier coefficients can be divisible by high powers of 2 in many cases. ### Specific results: - For weights \(k\in\{12, 16, 18, 20, 22, 26\}\), the author defined some constants \(\gamma, \rho, \chi, \nu, \eta, \omega\) and proved the following theorems: - When \(m, n\) are odd positive integers, and \(a, b\) are non - negative integers, for \(k\in\{12, 16, 18, 20, 22, 26\}\), we have: \[ a_k(2^a m, 2^b n)\equiv0\pmod{2^\gamma b}\quad\text{when}\ a = 0 \] \[ a_k(2^a m, 2^b n)\equiv0\pmod{2^\nu}\quad\text{when}\ a>0, b = 0 \] \[ a_k(2^a m, 2^b n)\equiv0\pmod{2^\chi}\quad\text{when}\ a>b\geq1 \] \[ a_k(2^a m, 2^b n)\equiv0\pmod{2^{\rho+\gamma(b - a)}}\quad\text{when}\ b>a\geq1 \] - Additionally, the author also proved that: \[ a_k(- 1,2^b n)\equiv0\pmod{2^{\gamma b}} \] \[ a_k(0,2^b n)\equiv0\pmod{2^\eta}\quad\text{when}\ b = 0 \] \[ a_k(0,2^b n)\equiv0\pmod{2^\omega}\quad\text{when}\ b>0 \] ### Method overview: - **Application of Hecke operators**: The author used Hecke operators to establish the relationship between certain coefficients \(a_k(m, n)\) and the τ - function. - **Two - dissections**: The author used Kolberg's method to perform two - dissections on the q - series to separate the coefficients of even and odd powers. - **Special basis of modular forms**: The author constructed a special basis of modular forms of specific weights to more easily prove the congruence properties of the coefficients. Through these methods, the author successfully proved the high 2 - divisibility of the Fourier coefficients of weakly holomorphic modular forms of specific weights.