Solution of the Ising model with Brascamp-Kunz boundary conditions by the transfer matrix method

This paper presents an exact solution of the square lattice Ising model under Brascamp-Kunz boundary conditions using the Schultz-Mattis-Lieb transfer matrix method by transforming the system into one with toroidal boundary conditions, thereby deriving the partition function in a fermionic representation and analytically identifying the Fisher zeros and physical critical point.

The Gist

Imagine a giant, endless checkerboard where every square holds a tiny magnet. These magnets can point either Up or Down. This is the Ising Model, a famous puzzle in physics that helps us understand how materials like iron become magnetic or how water turns into steam.

For decades, physicists have tried to solve this puzzle for different shapes and rules. One specific version, called the Brascamp-Kunz (B-K) model, is special because it has very strict rules about what happens at the very top and bottom edges of the board.

Here is the story of what Li and Wang did in this paper, explained without the heavy math.

1. The Problem: A Tricky Edge Case

Usually, when physicists study these magnetic grids, they imagine the board is wrapped around like a donut (a torus). This makes the math easier because the top edge connects to the bottom, and the left connects to the right. It's like a video game world where if you walk off the right side, you appear on the left.

However, the Brascamp-Kunz version is different.

• The Top Edge: All magnets are forced to point Up.
• The Bottom Edge: The magnets must alternate: Up, Down, Up, Down.
• The Sides: They still wrap around like a donut.

This creates a "cylinder" shape with very specific, rigid rules at the ends. While smart mathematicians had already solved this using a method called "Pfaffians" (think of it as a complex accounting trick), no one had solved it using the Transfer Matrix method, which is the standard "gold tool" for these problems.

2. The Solution: The "Magic Limit" Trick

The authors, Li and Wang, wanted to use the standard tool (the Transfer Matrix) to solve this tricky edge case. But the tool wasn't built for these weird edges.

So, they invented a clever workaround, like a magician's sleight of hand:

1. The Setup: They imagined a slightly different board. Instead of the strict B-K rules, they put a "super-strong glue" (a huge magnetic force) on the top and bottom rows.
   • On the top row, they glued the magnets so they really wanted to be Up.
   • On the bottom row, they glued them so they really wanted to be Down (or Up, depending on the spot).
2. The Magic Trick: They then asked, "What happens if we make this glue infinitely strong?"
3. The Result: When the glue is infinitely strong, the magnets have to follow the rules. The "glued" system naturally turns into the "Brascamp-Kunz" system we wanted to study.

By taking this "limit" (making the glue infinite), they transformed a messy, hard-to-solve problem into a clean, standard problem that their math tools could handle.

3. The Translation: From Magnets to Ghost Particles

Once they had the problem set up, they used a famous technique called the Schultz-Mattis-Lieb (SML) method.

• The Analogy: Imagine you have a room full of people (the magnets) shouting at each other. It's chaotic and hard to track.
• The Translation: The SML method is like a translator that turns the shouting people into a line of ghostly particles (fermions) that don't talk to each other but move in a very orderly, predictable way.
• The Benefit: Instead of solving a giant, tangled knot of equations, they just had to solve a simple list of independent steps. It's like turning a tangled ball of yarn into a neat row of separate threads.

4. The Discovery: Finding the "Fisher Zeros"

The ultimate goal of solving this model is to find the Fisher Zeros.

• What are they? Imagine the temperature of the system is a dial. As you turn the dial, the "ghost particles" start to dance. At a very specific temperature, the dance changes completely. This is the Critical Point (where a magnet becomes magnetic, or ice melts).
• The Map: The Fisher Zeros are like coordinates on a map that tell you exactly where this change happens.
• The Result: Because the B-K model has such a neat "double product" structure (thanks to their method), the authors could write down the exact coordinates of these zeros. They found that for a finite board, the zeros float in the complex number world, but as the board gets infinitely large, they line up perfectly on a circle, revealing the exact temperature where the phase transition happens.

5. Why Does This Matter?

You might ask, "Why bother solving a specific grid with weird edges?"

• It's a New Tool: This paper adds a new member to the family of "Transfer Matrix" solutions. It proves that even for tricky boundary conditions, you can use this powerful method if you are clever enough to set up the "glue" trick.
• It Connects the Dots: It shows that the weird B-K model and the standard donut-shaped model are actually cousins. By understanding how to switch between them, physicists can solve other difficult lattice models (like honeycomb or triangular grids) that were previously too hard to crack.
• It's Beautiful: In physics, finding an "exact solution" is like finding a perfect, unbroken diamond in a pile of dirt. It gives us a 100% clear view of how nature works in that specific scenario, without needing to guess or approximate.

In Summary:
Li and Wang took a difficult magnetic puzzle with rigid edges, used a "magic glue" trick to turn it into a standard puzzle, translated the magnets into orderly ghost particles, and successfully mapped out exactly when the system changes its state. They didn't just solve a puzzle; they built a new bridge for other scientists to cross.

Technical Summary

1. Problem Statement

The paper addresses the exact solution of the square lattice Ising model under Brascamp-Kunz (B-K) boundary conditions in the absence of a magnetic field.

• Context: While the Ising model under standard toroidal (periodic) boundary conditions was famously solved by Onsager (using transfer matrices) and Kaufman (using spinor analysis), the B-K boundary conditions present a unique challenge.
• Specifics of B-K BCs: The system consists of an $M \times 2N$ lattice. The boundaries are defined as:
  • Periodic in the horizontal ($N$) direction.
  • Fixed "all-plus" ($++++\dots$) spins on the upper edge (row 0).
  • Fixed alternating ($+-+\dots$) spins on the lower edge (row $M+1$).
• Motivation: Previous solutions for B-K BCs relied on the Pfaffian method (combinatorial formulation). A derivation using the transfer matrix method, specifically the Schultz-Mattis-Lieb (SML) fermionic representation, was missing. Furthermore, B-K BCs are crucial for analytically determining Fisher zeros (partition function zeros in the complex temperature plane), as they allow the partition function to be expressed in a double product form, unlike the sum-of-four-products form found in toroidal BCs.

2. Methodology

The authors employ the Schultz-Mattis-Lieb (SML) method within the transfer matrix formalism. The core strategy involves mapping the B-K system to a toroidal system with specific boundary interactions and taking a limit.

A. Mapping to Toroidal Boundary Conditions

Instead of solving the B-K system directly, the authors construct an auxiliary system on an $(M+2) \times 2N$ lattice with toroidal boundary conditions (periodic in both directions).

• Special Interactions: They introduce interaction constants $J_3$ in the zeroth row and $-J_3$ in the $(M+1)$-th row.
• The Limit: They take the limit $J_3 \to +\infty$.
  • In this limit, the Boltzmann factors enforce specific spin configurations on the boundary rows.
  • The zeroth row is forced to be either all $+$ or all $-$.
  • The $(M+1)$-th row is forced to be alternating $+-$ or $-+$.
  • This results in four distinct boundary condition sets. The authors prove that the partition functions for these four sets are identical due to spin-flip symmetry and periodicity.
  • Key Relation: The B-K partition function ($Z_{B-K}$) is shown to be exactly one-quarter of the toroidal partition function ($Z(J_3)$) in the limit $J_3 \to \infty$:
    $$Z_{B-K} = \frac{1}{4} \lim_{J_3 \to +\infty} \frac{1}{e^{4N\beta J_3}} Z(J_3)$$

B. Transfer Matrix and Fermionic Representation

1. Transfer Matrix Construction: The partition function $Z(J_3)$ is expressed as the trace of a product of transfer matrices ($V_1, V_2, V'_1$) acting on a Hilbert space of dimension $2^{2N}$.
2. Pauli to Fermion Mapping:
   • The transfer matrices are first written in terms of Pauli matrices ($\sigma_x, \sigma_z$).
   • A canonical transformation ($\sigma_x \to \sigma_z, \sigma_z \to -\sigma_x$) is applied to symmetrize the matrices.
   • Jordan-Wigner Transformation: Spin operators are mapped to fermionic creation ($C^\dagger$) and annihilation ($C$) operators.
3. Handling Boundary Terms: The operator $U = e^{i\pi \sum C^\dagger C}$ (related to the parity of the total number of fermions) appears due to the periodic boundary conditions in the fermionic representation.
   • The authors decompose the transfer matrix into even and odd sectors based on the eigenvalues of $U$ ($+1$ and $-1$).
   • The trace is split into contributions from these two sectors.

C. Diagonalization and Direct Product Decomposition

• Bogoliubov Transformation: A linear canonical transformation is applied to the fermion operators to diagonalize the Hamiltonian. This involves defining new fermion modes $\eta_q$.
• Momentum Quantization:
  • The even sector (B-K BCs) corresponds to anti-cyclic boundary conditions for the fermions, leading to momenta $q = \pm \frac{(2l-1)\pi}{2N}$.
  • The odd sector corresponds to cyclic boundary conditions, leading to momenta $q = 0, \pi, \pm \frac{2l\pi}{2N}$.
• Trace Calculation: The partition function is reduced to the trace of a direct product of $2 \times 2$ matrices for each momentum mode $q$.
• The Limit $K_3 \to \infty$: The authors explicitly calculate the limit of the trace terms. They demonstrate that the odd sector contribution vanishes in this limit, while the two sub-parts of the even sector contribute equally.

3. Key Results

A. Exact Partition Function

The authors derive the exact partition function for a finite $M \times 2N$ lattice under B-K BCs:
$$Z_{B-K} = 2^{2MN} \prod_{l=1}^{N} \prod_{j=1}^{M} \left( \cosh 2K_1 \cosh 2K_2 - \sinh 2K_1 \cos \frac{(2l-1)\pi}{2N} - \sinh 2K_2 \cos \frac{j\pi}{M+1} \right)$$
where $K_1 = \beta J_1$ and $K_2 = \beta J_2$.

• Significance: This result is a double product form, which is analytically tractable, unlike the sum-of-four-products form for toroidal BCs.

B. Fisher Zeros and Critical Points

Using the double product form, the Fisher zeros (roots of $Z$ in the complex $z = \sinh 2K_1$ plane) are explicitly solved:
$$z_{l,j} = \frac{\sinh 2K_2 \cos \frac{(2l-1)\pi}{2N} \pm i \cosh 2K_2 \sqrt{\dots}}{\cos \frac{j\pi}{M+1}}$$

• Thermodynamic Limit: As $M, N \to \infty$, the Fisher zeros form continuous loci that intersect the real axis at:
  $$\sinh 2K_1 \sinh 2K_2 = \pm 1$$
  This identifies the physical critical point for the phase transition.
• Free Energy: The free energy density in the thermodynamic limit is recovered, matching Onsager's famous solution:
  $$f = -k_B T \left[ \ln 2 + \frac{1}{8\pi^2} \int_0^{2\pi} d\theta \int_0^{2\pi} d\phi \ln(\cosh 2K_1 \cosh 2K_2 - \sinh 2K_1 \cos \theta - \sinh 2K_2 \cos \phi) \right]$$

C. Special Cases

• 1D Limit ($J_2=0$): The solution correctly reduces to the known 1D Ising model results.
• Isotropic Case ($J_1=J_2$): The Fisher zeros lie on the unit circle in the complex $\bar{z} = \sinh 2K$ plane, recovering Fisher's original findings regarding the critical point $\bar{z}=1$.

4. Significance and Contributions

1. Methodological Innovation: This is the first derivation of the B-K Ising model solution using the transfer matrix method and the SML fermionic representation. It bridges a gap between combinatorial (Pfaffian) solutions and operator-based solutions.
2. Technical Technique: The paper introduces a robust technique of mapping boundary conditions via interaction limits. By setting specific boundary interactions ($J_3$ and $-J_3$) and taking $J_3 \to \infty$, the complex B-K problem is transformed into a standard toroidal problem where powerful SML tools can be applied.
3. Analytical Clarity: The derivation clarifies why the B-K boundary conditions yield a double product form (due to the suppression of the odd fermion sector and the specific momentum quantization), whereas toroidal BCs yield a sum of four products.
4. Broader Applicability: The authors suggest this "limit of interactions" technique can be applied to other lattice models (e.g., Kagomé) and other boundary conditions (e.g., open BCs), expanding the toolkit for solving statistical mechanical models.

In summary, the paper provides a rigorous, alternative derivation of a classic exactly solvable model, offering new insights into the relationship between boundary conditions, fermionic representations, and the analytic structure of the partition function.