A conjecture of B. Lin on cubic partition pairs (Q1743628): Difference between revisions

From MaRDI portal
Importer (talk | contribs)
Created a new Item
 
Added link to MaRDI item.
links / mardi / namelinks / mardi / name
 

Revision as of 06:44, 1 February 2024

scientific article
Language Label Description Also known as
English
A conjecture of B. Lin on cubic partition pairs
scientific article

    Statements

    A conjecture of B. Lin on cubic partition pairs (English)
    0 references
    13 April 2018
    0 references
    The article concerns ``cubic partition pairs,'' the combinatorial object generated by the function \[\sum_{n\geq 0} b(n) q^n = \frac{1}{(q;q)^2_\infty (q^2;q^2)^2_\infty}.\] \textit{B. L. S. Lin} [J. Number Theory 171, 31--42 (2017; Zbl 1357.11105)] had previously shown that \(b(81n+61) \equiv 0 \pmod{27}\) and conjectured that the modulus could be improved to \(3^5 = 243\); here Hirschhorn gives an infinite family of congruences for \(b(n)\) modulo indefinitely large powers of 3, and as a consequence of the method proves that Lin's congruence can be improved to \(3^6 = 729\), best possible for this progression. To do so, he gives a complete description of any progression \[\sum_{n \geq 0} b \left( 3^k n + \frac{3^{k+\varepsilon}+1}{4}\right) q^n,\] where \(\varepsilon = 0\) if \(k\) is odd and 1 if \(k\) is even, in terms of integral linear combinations of certain shifts of \(\eta\)-quotients. The coefficients in the linear combination yield the progressions desired: for instance, when \(k=4\), the minimum power of 3 dividing any coefficient is 6. The methodology will be familiar in its basic techniques to readers conversant with Hirschhorn's earlier work, although the strength of this result means the argument is somewhat more involved than the typical result in the area. Beginning with classical dissections, he constructs the relevant modular equation and employs the huffing operator. A final induction establishes the general case, describing the coefficients in the linear combinations in terms of a recursively constructed matrix. Analysis of the powers of 3 appearing in the coefficients yields the general result of an indefinitely increasing lower bound on the power in a family of congruences: \[b \left( 3^k n + \frac{3^{k+\varepsilon}+1}{4}\right) \equiv 0 \pmod{3^{k-1+\varepsilon}}.\] (This is not best possible, as several more particular results including the congruence mentioned earlier show; these are established by directly calculating the coefficients involved.)
    0 references
    cubic partition pairs
    0 references
    recurrence
    0 references
    congruences
    0 references

    Identifiers