When you break a whole number into smaller pieces—say, writing 5 as 3+2 or 2+2+1 or 1+1+1+1+1—you're exploring what mathematicians call partitions. Seems simple enough. But partitions hide structures so rich they connect number theory, quantum physics, and unsolved conjectures dating back centuries.

A new study reveals explicit formulas for coefficients of polynomials linked to these partitions. The work addresses a problem woven through multiple branches of mathematics: how to untangle the separate influences of two different arithmetic functions operating simultaneously within recursively defined polynomials.

At the heart of the research lies the Dedekind eta-function, a mathematical object introduced in the 1800s. Powers of this function generate sequences whose coefficients carry profound arithmetic information. Euler studied these in 1748. Jacobi followed in 1828. Ramanujan pushed the boundaries in 1916, discovering connections to what's now called the tau function—a sequence so mysterious that whether it ever hits zero remains unproven.

The Lehmer conjecture, proposed in 1947, states that Ramanujan's tau function never vanishes. Despite decades of computational verification and theoretical effort, no proof exists. The new formulas provide fresh tools for investigating such questions.

The researchers work with polynomials built from normalized arithmetic functions. Think of an arithmetic function as a rule assigning a number to each positive integer. The rule might count divisors, return the number itself, or follow some other pattern. Two such functions, labeled g and h, generate polynomials through a recursive definition: start with 1, then build each new polynomial from all previous ones using a weighted sum controlled by g and h.

For special choices of g and h, these polynomials become famous objects. D'Arcais polynomials emerge from one combination and connect to hook length formulas discovered by physicists studying random partitions. Pochhammer polynomials appear from another choice. So do Chebyshev polynomials and Laguerre polynomials—each a cornerstone in different mathematical contexts.

The central achievement here is a master formula. It expresses any coefficient as a sum over partitions, where each term factors cleanly into two parts: one depending only on g, the other only on h. Previous approaches mixed these contributions inseparably.

Why does separation matter?

Consider an experiment. Fix g and vary h, or vice versa. The factored formula lets you track exactly how changes propagate. You can study how polynomial roots shift as parameters move. You can investigate whether coefficients stay positive, whether sequences stay log-concave—properties crucial in combinatorics and representation theory.

The formula applies the principle of separating variables to a discrete, recursive setting. It's less common than it sounds. Most recursions tangle their inputs irreversibly.

One application involves the conversion formula connecting two different normalizations. When h assigns each integer to itself versus assigning 1 to everything, coefficients transform in a precise, previously known way. The new master formula proves this transformation while revealing its structural origin.

Another concerns log-concavity. A sequence is log-concave if each term squared exceeds the product of its neighbors. This property appears throughout mathematics—in probability, in combinatorics, in geometry. The researchers show that under specific conditions, their polynomials inherit log-concavity. They also demonstrate when it fails, providing explicit bounds.

The Nekrasov-Okounkov polynomials, which arose from supersymmetric gauge theory and connect to random partition models in statistical mechanics, fit the framework. These polynomials express products over hook lengths—numbers associated with partition diagrams. Recent work proved they're log-concave for almost all degrees. The new formulas show immediately that specific initial coefficients satisfy the log-concavity condition.

Connections to modular forms run throughout. The Dedekind eta-function is a modular form, meaning it transforms predictably under certain symmetries. Powers of eta generate spaces of modular forms studied intensively for their connections to elliptic curves, Galois representations, and L-functions. The coefficients of these powers—encoded in the polynomials under study—carry arithmetic data about these deeper objects.

The reciprocals of Eisenstein series, another class of modular forms, also yield to the new methods. These reciprocals have coefficients expressible through the polynomials, linking modular form theory to partition combinatorics.

Technically, the proof proceeds by induction, leveraging the recursive structure. Define H as a function on partitions depending on h and n. Define G as a product over partition parts depending on g. The main theorem states that each coefficient equals the sum of G(μ)H(μ,n) over all partitions μ of the appropriate size.

For h equal to the identity function, H has a closed form involving factorials and multinomial coefficients. For h constant at 1, another closed form appears involving binomial coefficients. These special cases yield explicit coefficient formulas—Stirling numbers when g is trivial, Lah numbers when g is the identity.

The researchers note connections to symmetric polynomials. Their polynomials, viewed through roots, exhibit symmetries captured by elementary symmetric functions. This opens another perspective: studying root distributions as g and h vary.

One tantalizing question: can you develop a deformation theory? As h transitions smoothly from 1 to the identity function, roots should migrate continuously. For h=1, roots cluster—one at zero, the rest piled at -1. For h=identity, roots spread evenly: 0, -1, -2, and so on. What happens in between?

Applications extend beyond pure mathematics. The polynomials connect to representation theory of Lie algebras, where partition combinatorics parametrizes fundamental objects. They touch statistical mechanics through random partition models. They influence number theory through modular forms and L-functions.

The formulas also enable computational advances. Calculating high-degree coefficients previously required computing all lower degrees sequentially. The master formula allows direct computation via partition enumeration—potentially more efficient depending on the coefficient index.

The work mentions, almost in passing, that certain polynomials satisfy second-order linear difference equations. This places them in the family of orthogonal polynomials, objects with applications ranging from numerical analysis to quantum mechanics to signal processing.

Open questions remain. The general conversion formula holds only for functions of moderate growth—those whose generating series converge near zero. Can it extend further? How do the formulas behave asymptotically as degree grows? What finer structural properties do the coefficients inherit from g and h?

The mathematics here traces lineage through three centuries. Euler's pentagonal number theorem. Jacobi's triple product. Ramanujan's congruences. Macdonald's identities. Each generation uncovered new layers.

This study adds precision instruments to the toolkit. When you can separate variables in a recursion, you gain leverage. You can isolate effects. You can optimize one parameter while holding another fixed. You can ask sharper questions.

The language of the paper—arithmetic functions, normalized sequences, recursive definitions—might seem abstract. But these encode concrete counting problems. How many ways can you partition a number? How do divisor sums behave? What patterns emerge in prime distributions?

Mathematics proceeds by finding unexpected connections. Partitions link to modular forms. Modular forms link to elliptic curves. Elliptic curves link to rational points on algebraic varieties. Those link to Galois representations. Those link to L-functions and the Riemann hypothesis.

The formulas proved here sit at one junction in this web. They don't resolve the Lehmer conjecture or prove the Riemann hypothesis. They do provide new ways to probe both.

Science advances through better measurement tools as much as through bold conjectures. These formulas measure precisely how arithmetic functions contribute to polynomial coefficients. That precision will enable discoveries not yet imagined.

Credit & Disclaimer: This article is a popular science summary written to make peer-reviewed research accessible to a broad audience. All scientific facts, findings, and conclusions presented here are drawn directly and accurately from the original research paper. Readers are strongly encouraged to consult the full research article for complete data, methodologies, and scientific detail. The article can be accessed through https://doi.org/10.1007/s10958-025-07689-1