On 3-divisibility of 9- and 27-regular partitions (Q2115280)

The author contributes to the body of literature on congruences for \(m\)-regular partitions. Here they study the 9-regular and 27-regular partitions; denote the number of 9-regular or 27-regular partitions of \(n\) by \(b_9(n)\) and \(b_{27}(n)\) respectively. Then the author gives complete classifications for their divisibilities by 3 in terms of the factorizations of \(3n+1\) or \(12n+13\) respectively. Denote by \(\mathrm{ord}_p(n)\) the exponent of the prime \(p\) in the factorization of \(n\). The first main theorem of the paper is, for \(n\) a nonnegative integer: Theorem. We have that \(b_9(n) \equiv 0 \pmod{3}\) if and only if at least one of the following holds: there is a prime \(p \equiv 2 \pmod{3}\) such that \(\mathrm{ord}_p(3n+1)\) is odd, or there is a prime \(p \equiv 1 \pmod{3}\) such that \(\mathrm{ord}_p(3n+1) \equiv 2 \pmod{3}\). The second theorem gives conditions for divisibility of \(b_{27}(n)\) by 3. The conditions, while still definable in terms of the prime factorization of \(12n+13\), are more numerous and more complicated to state. They require (if the easier conditions fail) determining the squares in the unique representation \(p = x_p^2 + y_p^2\) for \(p \equiv 1 \pmod{4}\) prime, which is algorithmically efficient if notably more complicated than the previous theorem. The second main theorem is: Theorem. For \(n\) a nonnegative integer, \(b_{27}(n) \equiv 0 \pmod{3}\) if and only if at least one of the following conditions holds: \begin{itemize} \item[1.] \(12n+13\) is a square all of whose prime divisors are congruent to 11 modulo 12; \item[2.] there is a prine \(p \equiv 3 \pmod{4}\) such that \(\mathrm{ord}_p(12n+13)\) is odd; \item[3.] the cardinality of the set \(T\) of primes \(p \equiv 5 \pmod{12}\) such that \(\mathrm{ord}_p(12n+13)\) is odd, is neither 0 nor 2; \item[4.] \(\#T=2\) and, for all \(p \in T\), \(9 \vert (x_p + y_p)\); \item[5.] \(\#T=2\), \(3 \vert \vert (x_p + y_p)\) for all \(p \in T\), and there is some prime \(p \equiv 1 \pmod{12}\) for which \(\mathrm{ord}_p(12n+13) \equiv 2 \pmod{3}\); or \item[6.] \(\#T=0\) and a condition holds requiring several quantities to be further defined. \end{itemize} As one might expect from the statements, the bulk of the paper by far is the proof of the latter theorem. The proofs are via the theory of modular forms of \(CM\)-type, and are considerably more algebraic than the usual run of papers in this area, relying on investigation of primes in the number field \(\mathbb{Q}(\sqrt{-3})\). The reader will need to be familiar with Hecke eigenforms and the arithmetic of these quadratic number fields. An investigator who finds the paper interesting might usefully compare it with \textit{X. Xiong}'s paper [Int. J. Number Theory 12, No. 5, 1195--1208 (2016; Zbl 1372.11099)] characterizing the overpartition function modulo 16, and Merca's followup, in which theorems of similar flavor were given, relying on counting representations of \(n\) by sums of squares. There might be a larger family of results related to these ideas.