The Aesthetic Asymptotics of the Mayer Series Coefficients for a Dimer Gas on a Regular Lattice

This paper conjectures and provides strong numerical evidence that the Mayer series coefficients for dimer gases on various regular bipartite lattices follow a specific asymptotic exponential form, while also drawing surprising connections to Ising model susceptibility series and the partition function, and challenging combinatorialists to explain the latter's "magic" property.

The Gist

Imagine you are trying to predict the behavior of a massive crowd of tiny, dancing particles (called "dimers") on a grid, like a checkerboard or a 3D lattice. In the world of physics, these particles interact in complex ways, and scientists use a special mathematical recipe called a "Mayer series" to describe them. This recipe is a long list of numbers (coefficients) that get harder and harder to calculate the further you go down the list.

This paper, written by Paul Federbush, is like a detective story where the author tries to find a hidden pattern in the first 20 numbers of this list for various types of grids.

Here is the breakdown of the paper's journey, explained simply:

1. The Big Guess (The Conjecture)

The author has a hunch: even though these numbers look chaotic, they actually follow a very specific, elegant formula as they get larger. He proposes that if you look at the numbers far down the list, they grow in a way that can be described by a "magic formula" involving exponents (like $e^x$) and logarithms.

Think of it like this: If you were trying to predict the height of a growing plant every day, you might just guess it gets bigger by a random amount. But Federbush is saying, "No, there is a secret rhythm to the growth. If you know the rhythm, you can predict the future height with incredible accuracy, even if you only know the first few days of growth."

2. The Test Drive

To test this guess, the author looked at several different "grids" (lattices):

• Rectangular grids: Like a flat sheet (2D), a cube (3D), or even higher-dimensional shapes we can't visualize (up to 20 dimensions).
• Weird shapes: Tetrahedral (pyramid-like) and Body-Centered Cubic lattices.

He took the known first 20 numbers for these grids and tried to fit his "magic formula" to them. He adjusted the knobs (called $k$ values) on his formula until it matched the known data as closely as possible.

The Result: The match was stunningly good. The formula predicted the numbers almost perfectly, even for the smaller numbers in the list. The error was tiny—like measuring the distance from New York to London and being off by the width of a human hair.

3. The "Dual" Puzzle

The author realized that solving for these "magic knobs" directly was like trying to solve a giant, tangled knot of non-linear equations (very hard). So, he used a clever trick.

He turned the problem "inside out." Instead of looking at the growth directly, he looked at the ratio between one number and the one before it. He found that this ratio followed a much simpler, straight-line pattern (a linear equation).

• Analogy: Imagine trying to guess the next word in a sentence by analyzing the whole sentence (hard). Instead, he realized that if you just look at how the length of the sentence changes from one word to the next, the pattern becomes a simple, straight line. Once he solved the simple line, he could easily translate the answer back to the complex "magic formula."

4. The Surprising Discoveries

The paper ends with a few "odds and ends" that the author found while playing with the math:

• The "Magic" Dimension: The author defined a "dimension" ($d$) based on how many lines connect to a point. He found that his formula works regardless of what number you call the dimension, as long as you use the right math. It's like a universal key that fits many different locks.
• The Partition Function Challenge: He applied his method to a famous math problem called the "partition function" (which counts how many ways you can break a number into smaller parts). His formula worked perfectly here too. He issues a challenge to mathematicians: "Explain why this works! It's a magic trick we haven't figured out yet."
• Magnetic Connections: He also tested his method on the "Ising model" (a model for magnetism) and found that the numbers for magnetic materials behave very similarly to the numbers for the dancing particles, even though they seem like different worlds.

5. What This Paper Does Not Do

It is important to note what this paper is not about:

• It does not offer a new way to build computers or cure diseases.
• It does not claim to solve the phase transitions (like water turning to ice) in a practical, engineering sense.
• It does not provide a final proof that the formula is true for all numbers forever; it is a strong numerical observation based on the first 20 terms.

Summary

In short, this paper is a mathematical exploration. The author found a beautiful, hidden rhythm in the chaotic numbers describing particle interactions on grids. By using a clever "inside-out" trick, he showed that a simple formula can predict these complex numbers with amazing precision. He leaves the reader with a sense of wonder and a challenge: "We found the pattern, but now, can you explain the why?"

Technical Summary

Technical Summary: The Aesthetic Asymptotics of the Mayer Series Coefficients for a Dimer Gas on a Regular Lattice

Problem Statement
The paper investigates the asymptotic behavior of the Mayer series coefficients, denoted as $b(n)$, for a dimer gas on various regular lattices. While exact values for the first 20 coefficients are known for several lattices (including rectangular, body-centered cubic, and tetrahedral), the author seeks a precise asymptotic formula to describe these coefficients for large $n$. The central conjecture posits that for all regular lattices, the coefficients follow a specific asymptotic form:
$$(-1)^{n+1} b(n) \sim \exp\left(k_{-1}n + k_0 \ln(n) + \frac{k_1}{n} + \frac{k_2}{n^2} + \dots\right)$$
The primary challenge addressed is determining the optimal constants $k_{-1}, k_0, k_1, k_2$ using only the available finite data (specifically coefficients $b(13)$ through $b(20)$) and verifying the accuracy of this approximation for smaller values of $n$.

Methodology
The author employs a "linearized problem" approach to solve for the unknown constants in the exponential conjecture.

1. Data Source: The study utilizes Mayer series coefficients $b(n)$ derived from Virial coefficients $a(n)$, which were previously computed by Butera and Pernici for dimensions $d \le 10$ and $n \le 20$. The paper extends these results to infinite dimensions using a known relation between $a_d(n)$ and dimension $d$.
2. Dual Formulation: Instead of solving the highly non-linear system of equations required to fit the exponential form directly, the paper introduces a dual formulation. It approximates the ratio of consecutive coefficients:
   $$b(n) \approx (-1) \left(c_0 + \frac{c_1}{n} + \dots + \frac{c_r}{n^r}\right) b(n-1)$$
   By setting $r=6$ and imposing this relation for $14 \le n \le 20$, the author generates a system of seven linear equations to solve for the coefficients $c_0, \dots, c_6$.
3. Transformation: Once the linear coefficients $c_i$ are determined, they are analytically transformed into the exponential parameters $k_i$ (specifically $k_{-1}, k_0, k_1, k_2$) using derived asymptotic expansions (e.g., $k_{-1} \approx \ln(c_0)$, $k_0 \approx c_1/c_0$).
4. Validation: The accuracy of the approximation $B(n)$ (derived from the $k_i$) is tested against the known $b(n)$ values. The paper compares specific ratios ($b(n)/b(n-4)$) and calculates an $\ell_\infty$ error metric based on the relative difference between the true and approximated ratios.

Key Contributions and Results

• Asymptotic Formula: The paper provides explicit values for the constants $k_{-1}, k_0, k_1, k_2$ for rectangular lattices in dimensions $d=2, 3, 5, 11, 20$, as well as for body-centered cubic (bcc) lattices in dimensions 3, 4, 5, and the tetrahedral lattice.
• High Accuracy: The approximation is found to be strikingly accurate even for small $n$ (starting as early as $n=8$). The $\ell_\infty$ errors reported are generally on the order of $10^{-3}$ to $10^{-4}$, with some cases (like the partition function $p(n)$) reaching $10^{-7}$.
• Dimensional Independence of $k_{-1}$: In a digression, the author conjectures that for lattices with the same coordination number (defined as $2d$, where $d$ is half the number of edges per vertex), the leading coefficient $k_{-1}$ should be nearly identical. Data supports this, showing $k_{-1}$ values for rectangular and non-rectangular lattices with matching coordination numbers are very close (e.g., for coordination 8, rectangular $k_{-1} \approx 4.0893$ vs. bcc4 $k_{-1} \approx 4.0718$).
• Robustness to Parameter $d$: The author notes a "mysterious" discovery that the asymptotic approximation holds regardless of the specific value of $d$ used in the conversion formulas, provided the underlying lattice structure is regular.
• Extension to Other Models: The methodology was applied to the susceptibility series of the 2D Ising model (rectangular, triangular, honeycomb) and the partition function $p(n)$. In these cases, the coefficients were found to be positive, and the agreement between the approximation and the data was even more precise than in the dimer gas cases.

Significance and Claims
The author presents these findings as a "striking" empirical agreement between a four-term asymptotic expansion and finite series data. The paper frames the discovery as a "magic" property of the partition function and the dimer gas coefficients, suggesting deep, unexplained connections between statistical mechanics and combinatorics.

The paper explicitly states that the choice of the number of terms ($r=6$) and the specific method of optimization is "an art rather than a science," and the author does not claim to have proven the convergence or the theoretical necessity of the specific form. The work is presented as a detailed numerical study and a set of conjectures intended to motivate further theoretical research, particularly regarding the "positivity" of coefficients and the underlying reasons for the observed asymptotic behavior. The author challenges combinatorists to explain the "magic" property observed in the partition function $p(n)$.