Abstract:In this paper, we build on recent results of Frank Garvan and Rishabh Sarma as well as classical results of Bruce Berndt in order to establish the 11-dissection of the deviations of the rank and crank modulo 11. Using our new dissections we re-derive results of Garvan, Atkin, Swinnerton-Dyer, Hussain and Ekin. By developing and exploiting positivity conditions for quotients of theta functions, we will also prove new rank-crank inequalities and make several conjectures. For other applications of our methods, we will prove new congruences for rank moments as well as the Andrews' smallest parts function and Eisenstein series.
What problem does this paper attempt to address?
### What problem does this paper attempt to solve?
The main objective of this paper is to explore and establish the dissections of rank and crank biases under modulo 11. Specifically, based on the recent and classical research results of Frank Garvan, Rishabh Sarma, Bruce Berndt, etc., the author re - derives the results of Garvan, Atkin, Swinnerton - Dyer, Hussain, and Ekin, etc. regarding rank and crank biases modulo 11. In addition, by developing and using the positivity conditions of theta function quotients, the author also proves new rank - crank inequalities and proposes several conjectures.
### Background of the paper
In number theory, especially in partition theory, Ramanujan discovered the well - known congruence relations:
\[
p(5n + 4)\equiv0\pmod{5},
\]
\[
p(7n + 5)\equiv0\pmod{7},
\]
\[
p(11n + 6)\equiv0\pmod{11}.
\]
Dyson introduced the concept of "rank" to explain the first two congruence relations and conjectured that there exists another function, called "crank", which can explain the third congruence relation. Andrews and Garvan later discovered the specific definition of crank:
For a partition \(\pi\), let \(\lambda(\pi)\) denote the largest part, \(\vartheta(\pi)\) denote the number of 1s, and \(\mu(\pi)\) denote the number of parts greater than \(\vartheta(\pi)\). The crank \(c(\pi)\) is defined as:
\[
c(\pi):=\begin{cases}\lambda(\pi),&\text{when }\vartheta(\pi) = 0\\\mu(\pi)-\vartheta(\pi),&\text{otherwise}\end{cases}
\]
### Main results
The author defines the rank and crank biases under modulo \(r\) as:
\[
D(a,r;q):=\sum_{n = 0}^{\infty}\left(N(a,r,n)-\frac{p(n)}{r}\right)q^{n},
\]
\[
DC(a,r;q):=\sum_{n = 0}^{\infty}\left(M(a,r,n)-\frac{p(n)}{r}\right)q^{n}.
\]
Then, the author gives the 11 - dissection formulas of rank and crank biases under modulo 11. For example:
\[
DC(0,11)=v_{11}(10,- 12,-2,8,6,4,-4,-6,-8,2),
\]
\[
DC(1,11)=v_{11}(-1,10,-2,-3,-5,4,7,5,3,2),
\]
\[
\vdots
\]
These formulas not only help to re - derive the existing results but also are used to prove new congruence relations and inequalities. For example, the author proves new congruence relations regarding rank moments, Andrews' smallest parts function \(spt(n)\) and Eisenstein series.
### Applications and prospects
In addition, the author also proposes some new conjectures, such as conjectures about rank - crank inequalities, and explores how to use the numerical computing environment to generate higher - order inequalities. These results not only deepen the understanding of partition theory but also provide new directions and tools for future research.