Weakly interacting Bose gases in the canonical ensemble

This paper derives a first-order perturbative recursion formula for the canonical partition function of weakly interacting Bose gases, demonstrating that while it shares the same Feynman diagrams as the grand-canonical approach, it employs distinct rules to accurately characterize ground-state occupancy statistics and thermodynamic properties in box traps with Dirichlet boundary conditions.

The Gist

Imagine you have a room full of identical, invisible dancers. In the world of quantum physics, these are bosons (like atoms in a gas). When the room gets cold enough, something magical happens: all the dancers suddenly stop dancing individually and start moving in perfect unison, forming a single, giant "super-dancer." This is called a Bose-Einstein Condensate.

The paper you provided is a mathematical guidebook for predicting exactly how these dancers behave when they are in a fixed room with a fixed number of people, and when they occasionally bump into each other.

Here is the breakdown of their work using simple analogies:

1. The Problem: Counting in a Crowded Room

Physicists usually study these gases using a method called the "Grand-Canonical Ensemble." Imagine this as a room with an open door where people can wander in and out freely. It's mathematically easy to calculate things this way, but it's not how real experiments work. In real labs, you have a sealed box with a specific number of atoms (say, 500). You can't add or remove atoms; the number is fixed. This is the Canonical Ensemble.

The authors wanted to figure out how to do the math for this "sealed box" scenario, especially when the atoms start to interact (bump into each other) slightly.

2. The Old Way: The "Cycle" Trick

For atoms that don't bump into each other (ideal gas), physicists already had a clever trick. They realized that because the atoms are identical, you can think of them as forming loops or cycles.

• Imagine one atom dancing in a circle, or two atoms swapping places and dancing in a figure-eight.
• The math involves counting all the possible ways these loops can form to fill the room.
• The authors used a recursive formula (a step-by-step recipe) to count these loops. You calculate the answer for 1 atom, then use that to find the answer for 2, then 3, and so on, up to your total number of atoms.

3. The New Challenge: Adding "Bumps" (Interactions)

The tricky part of this paper is adding weak interactions. Imagine the dancers are no longer just floating; they are wearing slightly sticky shoes. They don't crash hard, but they occasionally brush against each other.

The authors tried to add this "stickiness" to their loop-counting recipe.

• The Diagrams: They found that the pictures (called Feynman diagrams) used to describe these interactions look exactly the same as the ones used for the "open door" (Grand-Canonical) method.
• The Twist: However, the rules for how to calculate the numbers on those pictures are different because the room is sealed. It's like using the same map for two different cities; the streets look similar, but the traffic laws are different.

4. The Glitch and the Fix

When they first applied their new rules to the "sticky" dancers, they hit a snag. At very low temperatures (when the dancers are very cold and slow), their math predicted a negative number of ways to arrange the room.

• Analogy: It's like trying to calculate the number of ways to arrange chairs in a room and getting an answer of "-5." That's impossible and unphysical.

To fix this, the authors performed a resummation.

• Analogy: Imagine you are adding up a long list of numbers, but the numbers keep flipping signs and getting huge, making the total swing wildly. Instead of adding them one by one, you group them together in a smarter way to see the true, stable pattern underneath.
• By "resumming" their recipe, they created a new, stable formula that never gives negative results, even at very low temperatures.

5. What They Found: The "Box Trap"

They tested their new theory on a specific scenario: a gas in a box with hard walls (Dirichlet boundary conditions). This is important because real experiments often use "digital mirrors" to create box-shaped traps for atoms.

They calculated two main things:

1. The "Condensate Fraction" (How many dancers are in sync?): They tracked how many atoms joined the "super-dancer" group as the temperature dropped.
2. The "Fluctuations" (How wobbly is the group?): They measured how much the number of dancers in the group jitters.

Key Results:

• Small vs. Large Groups: For small numbers of atoms, the "wobble" (fluctuations) and the "heat capacity" (how much energy it takes to warm them up) gave slightly different answers for when the phase change happens.
• The Big Picture: As the number of atoms gets huge (approaching the thermodynamic limit), these two different measurements converged to the same answer.
• The Interaction Effect: When the atoms were slightly sticky (interacting), the temperature at which they all synchronized shifted. Interestingly, the shift calculated by looking at the "wobble" was slightly different from the shift calculated by looking at the "heat," and they settled on two different final values in the limit of infinite atoms.

Summary

In short, this paper provides a new, corrected mathematical recipe for predicting how a fixed number of slightly sticky atoms behave in a sealed box. They fixed a mathematical error that caused "negative numbers" at low temperatures and showed that while small groups of atoms behave a bit differently than huge groups, the theory holds up and matches what we expect from the "open door" method when the group gets large enough.

What they did NOT do:

• They did not apply this to medical treatments or clinical uses.
• They did not claim this solves the problem of quantum computing directly.
• They did not extend the results to systems with strong, violent collisions (only "weak" interactions).
• They did not claim to explain the behavior of atoms at absolute zero where quantum effects dominate completely (they noted their method works best for "larger temperatures" where thermal effects matter).

Technical Summary

Technical Summary: Weakly Interacting Bose Gases in the Canonical Ensemble

Problem Statement
Theoretical descriptions of Bose-Einstein condensates (BECs) often rely on the grand-canonical ensemble due to its mathematical convenience in the thermodynamic limit. However, this ensemble yields unphysical predictions for ground-state particle number fluctuations at absolute zero, which grow linearly with particle number, whereas the canonical ensemble correctly predicts vanishing fluctuations. While the canonical treatment for ideal (non-interacting) Bose gases is well-established via cycle decomposition of the permutation group, extending this framework to include weak two-particle interactions has proven difficult due to combinatorial complexity. Existing perturbative approaches for the canonical partition function often fail at low temperatures, producing unphysical negative partition functions. This paper addresses the need for a systematic canonical description of weakly interacting Bose gases, particularly relevant for current experiments with small numbers of atoms in box traps where finite-size effects and particle number conservation are critical.

Methodology
The authors develop a canonical perturbation theory for a weakly interacting Bose gas, treating the interaction as a first-order perturbation. The methodology proceeds through the following steps:

1. Cycle Decomposition and Recursion: Building on the path integral representation of the partition function, the authors utilize the cycle decomposition of the symmetric permutation group. For non-interacting gases, this leads to a recursive formula for the $N$-particle partition function $Z_N^{(0)}(\beta)$ based on the one-particle partition function $Z_1(\beta)$.
2. Perturbative Expansion: The interaction is introduced via a two-particle potential $V^{(int)}$. The partition function is expanded to first order in the interaction strength. This expansion introduces "interacting cycles" where the product of single-particle propagators connects the coordinates involved in the interaction.
3. Diagrammatic Representation: The authors derive a diagrammatic representation for the canonical ensemble. While the topology of the Feynman diagrams matches those in the grand-canonical ensemble (specifically Hartree and Fock channels), the Feynman rules differ. Each line represents an imaginary-time propagator with a specific number of windings around a Feynman cylinder, and the summation constraints reflect the fixed particle number $N$.
4. Resummation: A direct perturbative recursion formula is derived but is found to yield negative partition functions at low temperatures. To resolve this, the authors perform a resummation of the recursion relation. This involves substituting the standard one-particle partition function with a renormalized version, $Z_1(k, \beta)$, which incorporates interaction effects into an effective energy spectrum $E_n(k)$.
5. Statistical and Thermodynamic Extraction: Using the resummed recursion formula, the authors derive expressions for the probability distribution of ground-state occupancy, its moments, and cumulants. Thermodynamic quantities, specifically entropy and heat capacity, are calculated via the Helmholtz free energy derived from the renormalized partition function.
6. Application to Box Traps: The formalism is applied to a dilute Bose gas in a three-dimensional box trap with Dirichlet boundary conditions, using a contact interaction potential. This choice reflects current experimental setups using digital mirror devices.

Key Contributions

• Canonical Perturbation Theory: The paper establishes a systematic first-order perturbative framework for weakly interacting Bose gases within the canonical ensemble, overcoming previous difficulties in handling the combinatorial complexity of interactions.
• Resummed Recursion Formula: The authors derive a resummed recursion formula (Eq. 62) that avoids the unphysical negative partition function problem at low temperatures. This formula allows for the calculation of thermodynamic and statistical properties up to first order in the interaction.
• Diagrammatic Rules: The work clarifies the diagrammatic rules for the canonical ensemble, demonstrating that while the diagrams resemble the grand-canonical case, the rules for calculating weights (involving winding numbers and fixed $N$) are distinct.
• Ground-State Statistics: The method provides a direct way to calculate the cumulants of the ground-state occupancy (condensate fraction and fluctuations) without resorting to the grand-canonical ensemble.
• Finite-Size Analysis: The approach explicitly incorporates finite-size corrections by using Dirichlet boundary conditions, making it directly applicable to finite experimental systems.

Results
The theory was applied to a system of 500 particles in a box trap with a gas parameter $an^{1/3} = 0.1$.

• Cumulants: The first four cumulants of the ground-state occupancy were calculated. The results show that interactions modify the skewness and kurtosis of the distribution. The third cumulant (skewness) changes sign relative to the critical temperature, and the fourth cumulant (kurtosis) indicates deviations from Gaussianity.
• Thermodynamics: Entropy and heat capacity were computed. The heat capacity peak was used to estimate the critical temperature shift.
• Critical Temperature Shift: The study analyzed the shift in critical temperature $\Delta T_c$ as a function of particle number $N$.
  • For the ideal gas, the critical temperatures derived from heat capacity and ground-state fluctuations converge to the same thermodynamic limit value as $N \to \infty$.
  • For the interacting gas, the critical temperatures derived from fluctuations and heat capacity converge to different values in the thermodynamic limit. The fluctuation-based shift ($\Delta T_c/T_c^{(0)} \approx 0.133$) coincides with literature values obtained from Quantum Monte Carlo (QMC) and variational perturbation theory in the grand-canonical ensemble. The heat capacity-based shift is larger ($\approx 0.171$).
• Limitations: The authors note that the first-order perturbative approach does not reproduce the quantum depletion at zero temperature, as this is a non-perturbative effect captured by Bogoliubov theory. However, the method provides reliable results for larger temperatures and includes finite-size corrections.

Significance
The paper argues that the canonical ensemble is necessary for accurately describing experiments with small to intermediate numbers of atoms, particularly in box traps, where particle number fluctuations in the grand-canonical ensemble are unphysical. By providing a resummed recursive formula, the authors offer a practical tool to calculate thermodynamic and statistical properties of weakly interacting Bose gases while strictly conserving particle number. The convergence of the fluctuation-based critical temperature shift to established grand-canonical results in the thermodynamic limit validates the approach, while the divergence of the heat capacity-based shift highlights the subtle differences between ensembles even in the limit of large $N$. This work bridges the gap between ideal canonical treatments and interacting systems, offering a theoretical framework tailored to modern finite-size quantum gas experiments.