MacMahon-type q q -series Journal: Journal of Combinatorial Theory, Series A
Mircea Merca · 2025
Read MacMahon-type q q -series Journal: Journal of Combinatorial Theory, Series A by Mircea Merca (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-2512-00978.txt.
Introduction
Mircea Merca studies a family of q-series in a 2025 combinatorics paper. Q-series are series built from a base variable q and appear across number theory and partition counting. MacMahon pioneered many such series, and recent work by several authors extended his ideas to higher-order partition counts.
Merca studies a class of q-series that arise from nested divisor structures, where divisors inside divisors shape the terms. These are called MacMahon-type series. Results here connect divisor sums with partition generating functions, two classic themes in number theory. Such identities often reveal that two very different counting problems are secretly the same.
Catalog record
| CREATOR | Mircea Merca |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 30 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2512-00978
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
MacMahon-type q q -series Journal: Journal of Combinatorial Theory, Series A Mircea Merca Abstract Motivated by earlier work of P. A. MacMahon and recent contributions of T. Amdeberhan, G. E. Andrews, K. Ono, A. Singh, and R. Tauraso on higher-order partition enumerants, we study a class of q q -series arising from nested divisor structures. In particular, we consider the q q -series V k ( q ) = ∑ 1 ≤ n 1 ≤ n 2 ≤ ⋯ ≤ n k q n 1 + n 2 + ⋯ + n k ( 1 − q n 1 ) 2 ( 1 − q n 2 ) 2 ⋯ ( 1 − q n k ) 2 , V_{k}(q)=\sum_{1\leq n_{1}\leq n_{2}\leq\cdots\leq n_{k}}\frac{q^{\,n_{1}+n_{2}+\cdots+n_{k}}}{(1-q^{n_{1}})^{2}(1-q^{n_{2}})^{2}\cdots(1-q^{n_{k}})^{2}}, introduced recently as MacMahon-type generating functions. We further define a new MacMahon-type series W k ( q ) = ∑ 1 ≤ n 1 ≤ n 2 ≤ ⋯ ≤ n k q 2 ( n 1 + n 2 + ⋯ + n k ) − k ( 1 − q 2 n 1 − 1 ) 2 ( 1 − q 2 n 2 − 1 ) 2 ⋯ ( 1 − q 2 n k − 1 ) 2 , W_{k}(q)=\sum_{1\leq n_{1}\leq n_{2}\leq\cdots\leq n_{k}}\frac{q^{\,2(n_{1}+n_{2}+\cdots+n_{k})-k}}{(1-q^{2n_{1}-1})^{2}(1-q^{2n_{2}-1})^{2}\cdots(1-q^{2n_{k}-1})^{2}}, and establish families of identities, generating function relations, and hypergeometric representations for the truncated forms of V k ( q ) V_{k}(q) and W k ( q ) W_{k}(q) . Connections with overpartition pairs and bipartitions with distinct odd parts arise naturally in this context. 2010 MSC 11P81 , 11P82 , 05A19 , 05A20 1 Introduction The partitions of an integer represent a fascinating subject at the intersection of combinatorics and number theory. A partition of a positive integer n is a way of writing n as a sum of positive integers, where the order of the terms (called parts) does not matter [ 5 ] . The multiplicity of parts provides a useful alternative representation for the partitions of a positive integer n : n=t_{1}+2t_{2}+3t_{3}+\cdots+nt_{n}, where t_{i} is the number of times the part i appears. These counts t_{i} are non-negative integers (possibly zero), and this form is often encoded compactly as (1^{t_{1}}2^{t_{2}}\ldots n^{t_{n}})\qquad\text{or}\qquad(n^{t_{n}}\ldots 2^{t_{2}}1^{t_{1}}), which explicitly shows the multiplicity of each part. For instance, the integer 4 has five partitions, represented as: (4^{1}),\quad(3^{1}1^{1}),\quad(2^{2}),\quad(2^{1}1^{2}),\quad(1^{4}). Note that in this frequency-based description, it is possible that all t_{i}=0 ; this case corresponds to the empty partition of n=0 , conventionally denoted by the empty list () . Hence, the definitions are consistent when we regard () as the unique partition of zero. A partition of an integer n is said to involve parts of k distinct magnitudes if the set of values taken by its parts has a cardinality exactly equal to k . For example, if we consider partitions of 8 involving parts of 3 different magnitudes, we might have: \displaystyle(5^{1}2^{1}1^{1}),\ (4^{1}3^{1}1^{1}),\ (4^{1}2^{1}1^{2}),\ (3^{1}2^{2}1^{1}),\ (3^{1}2^{1}1^{3}). (1) The enumeration and structural analysis of these partitions has established deep connections between combinatorial counting problems and sophisticated analytical tools, particularly q -series and modular forms. Considering partitions involve parts of k distinct magnitudes, MacMahon introduced two fundamental families of partition functions: a^{\pm}_{k}(n):=\sum_{\begin{subarray}{c}\lambda_{1}t_{1}+\lambda_{2}t_{2}+\cdots+\lambda_{k}t_{k}=n\\ 1\leq\lambda_{1}<\lambda_{2}<\cdots<\lambda_{k}\\ (t_{1},t_{2},\ldots,t_{k})\in\mathbb{N}^{k}\end{subarray}}(\pm 1)^{t_{1}+t_{2}+\cdots+t_{k}+k}\,t_{1}t_{2}\cdots t_{k} and c^{\pm}_{k}(n):=\sum_{\begin{subarray}{c}(2\lambda_{1}-1)t_{1}+(2\lambda_{2}-1)t_{2}+\cdots+(2\lambda_{k}-1)t_{k}=n\\ 1\leq\lambda_{1}<\lambda_{2}<\cdots<\lambda_{k}\\ (t_{1},t_{2},\ldots,t_{k})\in\mathbb{N}^{k}\end{subarray}}(\pm 1)^{t_{1}+t_{2}+\cdots+t_{k}+k}\,t_{1}t_{2}\cdots t_{k}. These serve as the basis for two significant families of q -series, which MacMahon [ 15 ] established as their generating functions: A^{\pm}_{k}(q)=\sum_{n=0}^{\infty}a^{\pm}_{k}(n)\,q^{n}:=\sum_{0<n_{1}<n_{2}<\cdots<n_{k}}\prod_{i=1}^{k}\frac{q^{n_{i}}}{(1\mp q^{n_{i}})^{2}} and C^{\pm}_{k}(q)=\sum_{n=0}^{\infty}c^{\pm}_{k}(n)\,q^{n}:=\sum_{0<n_{1}<n_{2}<\cdots<n_{k}}\prod_{i=1}^{k}\frac{q^{2n_{i}-1}}{(1\mp q^{2n_{i}-1})^{2}}. We remark that A^{\pm}_{0}(q) and C^{\pm}_{0}(q) are defined to be 1 . These functions have recently experienced a resurgence of interest, resulting in a considerable number of new articles that highlight their continued mathematical importance [ 3 , 1 , 2 , 6 , 4 , 7 , 13 , 16 , 17 , 18 , 19 , 20 ] . Amdeberhan et al. [ 1 ] recently refined MacMahon’s original scope, suggesting that while the established approach is sound, the study could be expanded to include partitions with exactly k size of parts, possibly equal. Following this idea, we consider two families of q -series: V^{\pm}_{k}(q)=\sum_{n=0}^{\infty}v^{\pm}_{k}(n)\,q^{n}:=\sum_{1\leq n_{1}\leq n_{2}\leq\cdots\leq n_{k}}\prod_{i=1}^{k}\frac{q^{n_{i}}}{(1\mp q^{n_{i}})^{2}}, and W^{\pm}_{k}(q)=\sum_{n=0}^{\infty}w^{\pm}_{k}(n)\,q^{n}:=\sum_{1\leq n_{1}\leq n_{2}\leq\cdots\leq n_{k}}\prod_{i=1}^{k}\frac{q^{2n_{i}-1}}{(1\mp q^{2n_{i}-1})^{2}}, where the families of partition functions v^{\pm}_{k}(n) and w^{\pm}_{k}(n) are defined by v^{\pm}_{k}(n):=\sum_{\begin{subarray}{c}\lambda_{1}t_{1}+\lambda_{2}t_{2}+\cdots+\lambda_{k}t_{k}=n\\ 1\leq\lambda_{1}\leq\lambda_{2}\leq\cdots\leq\lambda_{k}\\ (t_{1},t_{2},\ldots,t_{k})\in\mathbb{N}^{k}\end{subarray}}(\pm 1)^{t_{1}+t_{2}+\cdots+t_{k}+k}\,t_{1}t_{2}\cdots t_{k} and w^{\pm}_{k}(n):=\sum_{\begin{subarray}{c}(2\lambda_{1}-1)t_{1}+(2\lambda_{2}-1)t_{2}+\cdots+(2\lambda_{k}-1)t_{k}=n\\ 1\leq\lambda_{1}\leq\lambda_{2}\leq\cdots\leq\lambda_{k}\\ (t_{1},t_{2},\ldots,t_{k})\in\mathbb{N}^{k}\end{subarray}}(\pm 1)^{t_{1}+t_{2}+\cdots+t_{k}+k}\,t_{1}t_{2}\cdots t_{k}. For example, if we consider partitions of 5 involving parts of 3 magnitudes (possibly equal), we might have: \displaystyle(3^{1}1^{1}1^{1}),\ (2^{1}2^{1}1^{1}),\ (2^{1}1^{1}1^{2}),\ (2^{1}1^{2}1^{1}),\ (1^{1}1^{1}1^{3}), \displaystyle(1^{1}1^{2}1^{2}),\ (1^{1}1^{3}1^{1}),\ (1^{2}1^{1}1^{2}),\ (1^{2}1^{2}1^{1}),\ (1^{3}1^{1}1^{1}). So we have: \displaystyle v^{+}_{3}(5)= \displaystyle 1\cdot 1\cdot 1+1\cdot 1\cdot 1+1\cdot 1\cdot 2+1\cdot 2\cdot 1+1\cdot 1\cdot 3 \displaystyle+1\cdot 2\cdot 2+1\cdot 3\cdot 1+2\cdot 1\cdot 2+2\cdot 2\cdot 1+3\cdot 1\cdot 1=27 \displaystyle v^{-}_{3}(5)= \displaystyle 1\cdot 1\cdot 1+1\cdot 1\cdot 1-1\cdot 1\cdot 2-1\cdot 2\cdot 1+1\cdot 1\cdot 3 \displaystyle+1\cdot 2\cdot 2+1\cdot 3\cdot 1+2\cdot 1\cdot 2+2\cdot 2\cdot 1+3\cdot 1\cdot 1=19. On the other hand, partitions of 5 involving odd parts of 3 magnitudes (possibly equal) are: \displaystyle(3^{1}1^{1}1^{1}),\ (1^{1}1^{1}1^{3}),\ (1^{1}1^{2}1^{2}),\ (1^{1}1^{3}1^{1}),\ (1^{2}1^{1}1^{2}),\ (1^{2}1^{2}1^{1}),\ (1^{3}1^{1}1^{1}). So we have: \displaystyle w^{+}_{3}(5)= \displaystyle 1\cdot 1\cdot 1+1\cdot 1\cdot 3+1\cdot 2\cdot 2 \displaystyle+1\cdot 3\cdot 1+2\cdot 1\cdot 2+2\cdot 2\cdot 1+3\cdot 1\cdot 1=22. We remark that w^{-}_{k}(n)=(-1)^{n+k}\,w^{+}_{k}(n). In this paper, in order to explore properties of the families of MacMahon-type q -series V_{k}^{\pm}(q) and W_{k}^{\pm}(q) , we introduce the following partitions functions. Definition 1 . For positive integers k , m and n , we define: (i) \displaystyle{v^{\pm}_{k,m}(n):=\sum_{\begin{subarray}{c}\lambda_{1}t_{1}+\lambda_{2}t_{2}+\cdots+\lambda_{k}t_{k}=n\\ 1\leq\lambda_{1}\leq\lambda_{2}\leq\cdots\leq\lambda_{k}\leq m\\ (t_{1},t_{2},\ldots,t_{k})\in\mathbb{N}^{k}\end{subarray}}(\pm 1)^{t_{1}+t_{2}+\cdots+t_{k}+k}\,t_{1}t_{2}\cdots t_{k};} (ii) \displaystyle{w^{\pm}_{k,m}(n):=\sum_{\begin{subarray}{c}(2\lambda_{1}-1)t_{1}+(2\lambda_{2}-1)t_{2}+\cdots+(2\lambda_{k}-1)t_{k}=n\\ 1\leq\lambda_{1}\leq\lambda_{2}\leq\cdots\leq\lambda_{k}\leq m\\ (t_{1},t_{2},\ldots,t_{k})\in\mathbb{N}^{k}\end{subarray}}(\pm 1)^{t_{1}+t_{2}+\cdots+t_{k}+k}\,t_{1}t_{2}\cdots t_{k}.} For positive integers k and m , we remark that the generating functions of v^{\pm}_{k,m}(n) and w^{\pm}_{k,m}(n) are truncated forms of the MacMahon-type q -series V^{\pm}_{k}(q) and W^{\pm}_{k}(q) , i.e., V^{\pm}_{k,m}(q)=\sum_{n=0}^{\infty}v^{\pm}_{k,m}(n)\,q^{n}:=\sum_{1\leq n_{1}\leq n_{2}\leq\cdots\leq n_{k}\leq m}\prod_{i=1}^{k}\frac{q^{n_{i}}}{(1\mp q^{n_{i}})^{2}}, and W^{\pm}_{k,m}(q)=\sum_{n=0}^{\infty}w^{\pm}_{k,m}(n)\,q^{n}:=\sum_{1\leq n_{1}\leq n_{2}\leq\cdots\leq n_{k}\leq m}\prod_{i=1}^{k}\frac{q^{2n_{i}-1}}{(1\mp q^{2n_{i}-1})^{2}}, with V^{\pm}_{0,m}(q) and W^{\pm}_{0,m}(q) defined to be 1 . In this context, we recall the definition of the q -binomial coefficient, also known as Gaussian polynomial. For integers m,k\in\mathbb{Z} , it is defined by {m\brack k}={m\brack k}_{q}:=\begin{cases}\dfrac{(q;q)_{m}}{(q;q)_{k}\,(q;q)_{m-k}},&\text{if }0\leq k\leq m,\\[6.0pt] 0,&\text{otherwise}.\end{cases} Here (a;q)_{n} denotes the q -shifted factorial (the q -Pochhammer symbol), defined for n\in\mathbb{Z}_{\geq 0} by (a;q)_{n}=\prod_{r=0}^{n-1}(1-aq^{\,r}),\qquad(a;q)_{0}=1. We also use the infinite q -product (a;q)_{\infty}=\lim_{n\to\infty}(a;q)_{n}, which exists and converges whenever |q|<1 . The first result of this paper exhibits two infinite families of linear combinations of the q -series V_{k,m}^{\pm}(q) and W_{k,m}^{\pm}(q) , expressed in terms of Gaussian polynomials or basic hypergeometric series [ 10 ] . To this end, we define \displaystyle H_{k,m}(q,z) \displaystyle:=\sum_{j=0}^{\infty}{m-1+j\brack j}_{q}{m-1+k+j\brack k+j}_{q}z^{j} \displaystyle=\frac{(q^{m};q)_{k}}{(q;q)_{k}}\,{{}_{2}}\phi_{1}\bigg(\begin{matrix}q^{m},q^{m+k}\\ q^{k+1}\end{matrix}\,;q,z\bigg), where {{}_{2}}\phi_{1}\bigg(\begin{matrix}a,b\\ c\end{matrix}\,;q,z\bigg)=\sum_{n=0}^{\infty}\frac{(a;q)_{n}(b;q)_{n}}{(q;q)_{n}(c;q)_{n}}z^{n}. Theorem 1 . Let k and m be non-negative integers. For |q|<1 , (i) \displaystyle{\sum_{j=k}^{\infty}(\mp 1)^{j-k}\binom{2j}{j-k}V^{\pm}_{j,m}(q)=(\pm q;q)^{2}_{m}\,q^{k}\,H_{k,m}(q,q^{2})} (ii) \displaystyle{\sum_{j=k}^{\infty}(\mp 1)^{j-k}\binom{2j}{j-k}W^{\pm}_{j,m}(q)=(\pm q;q^{2})^{2}_{m}\,q^{k}\,H_{k,m}(q^{2},q^{2})}. By Theorem 1 , we deduce that the q -series V^{\pm}_{j,m}(q) and W^{\pm}_{j,m}(q) admit explicit expressions in terms of q -binomial coefficients. To formulate these expressions, we introduce the inverse coefficients B_{k,j} , defined by B_{0,0}=1,\qquad B_{k,0}=2\quad(k\geq 1), and, for j\geq 1 and k\geq j , B_{k,j}=\frac{2k}{k+j}\binom{k+j}{2j}. With these coefficients, the q -series can be written in closed form, providing a direct combinatorial interpretation. Theorem 2 . Let j and m be non-negative integers. For |q|<1 , (i) \displaystyle{V^{\pm}_{j,m}(q)=(\pm q;q)_{m}^{2}\sum_{k=j}^{\infty}(\pm 1)^{k-j}B_{k,j}\,q^{k}\,H_{k,m}(q,q^{2})} , (ii) \displaystyle{W^{\pm}_{j,m}(q)=(\pm q;q^{2})^{2}_{m}\sum_{k=j}^{\infty}(\pm 1)^{k-j}B_{k,j}\,q^{k}\,H_{k,m}(q^{2},q^{2})} . In light of Theorems 1 and 2 , we are now able to deduce structural properties of the MacMahon-type q -series V^{\pm}_{k}(q) and W^{\pm}_{k}(q) . In particular, taking the