Finite temperature phase diagram of the extended Bose-Hubbard model in the presence of disorder

This paper presents a mean-field study of the finite-temperature phase diagram of the disordered Extended Bose-Hubbard model, revealing how thermal fluctuations compete with quantum effects to melt Mott insulator and charge-density-wave phases into normal fluids or Bose glasses, with disorder further suppressing the stability of these insulating states.

The Gist

Imagine a giant, perfectly organized dance floor where thousands of tiny, energetic dancers (atoms) are trying to move around. This is the world of ultracold atoms in a laboratory. Usually, these atoms are so cold they behave like a single, synchronized wave (a Superfluid), gliding effortlessly across the floor. But if you make the floor "sticky" or force the dancers to stay in specific spots, they get stuck in a rigid grid, refusing to move (a Mott Insulator).

This paper is a study of what happens to this dance floor when you turn up the heat and throw some chaos into the mix. The authors are looking at a specific setup called the Extended Bose-Hubbard Model, which is like a rulebook for how these atoms interact.

Here is the story of their findings, broken down into simple concepts:

1. The Setup: The Dance Floor and the Rules

The scientists are using Rydberg atoms (atoms that have been excited to a huge size) trapped in optical lattices (a grid made of laser light).

• The Neighbors: In this dance, atoms care about their neighbors.
  • Nearest-Neighbor (NN): They care about the person standing right next to them.
  • Next-Nearest-Neighbor (NNN): By adjusting the lasers, they can make the atoms care about the person standing two spots away, too.
• The Disorder: In the real world, nothing is perfect. Sometimes the floor is bumpy, or the music is slightly off-key. This is called disorder. In the lab, they create this by adding random "noise" to the laser grid.

2. The Temperature Factor: Heating Up the Dance Floor

The main question of the paper is: What happens when we stop being at absolute zero and start warming things up?

• At Absolute Zero (The Frozen State): The atoms are perfectly ordered. They form rigid "lobes" of stability. You have the Mott Insulator (everyone stands in their own spot) and the Charge Density Wave (CDW) (everyone stands in a checkerboard pattern, alternating empty and full spots).
• Turning up the Heat (Thermal Fluctuations): As you add heat, the atoms start to jiggle. This is like the dancers getting sweaty and restless.
  • The Melting: The rigid checkerboard patterns (CDW) are the first to break. They "melt" into a Normal Fluid. The atoms are no longer stuck in a perfect pattern; they are just a chaotic, flowing soup.
  • The Mott Insulator: The rigid "one-per-spot" pattern is tougher. It survives a bit longer, but eventually, even it melts into the chaotic soup if the heat gets high enough.

The Analogy: Think of a block of ice (the Insulator). As you heat it, it turns to water (the Normal Fluid). The paper calculates exactly when the ice turns to water for different patterns of ice.

3. The Chaos Factor: Adding Disorder (The "Bose Glass")

Now, imagine the dance floor isn't just hot; it's also uneven. Some spots are sticky, some are slippery, and some are bumpy. This is disorder.

• The Bose Glass: When disorder is introduced, a new phase appears called the Bose Glass.
  • What is it? Imagine a crowd of people trying to move through a maze. They aren't stuck in a rigid grid (like the Insulator), but they can't flow freely either (like the Superfluid). They are "stuck" in random pockets created by the bumps in the floor. They are compressible (squishy) but not flowing.
• The Competition: The paper finds that disorder fights against the order.
  • If the disorder is weak, the atoms can still form their rigid patterns (Insulators), but the patterns get squeezed.
  • If the disorder is strong, it destroys the patterns entirely, leaving only the "Bose Glass" and the "Normal Fluid."

4. The Big Discovery: Who Melts First?

The authors mapped out a "Phase Diagram," which is like a weather map for the atoms. Here is what they found:

1. The Weak Links: The Charge Density Wave (CDW) patterns are the most fragile. They melt into the chaotic fluid at relatively low temperatures.
2. The Strong Links: The Mott Insulator patterns are more robust. They can handle more heat before melting.
3. The Disorder Effect: Adding disorder makes everything melt faster. It's like adding a strong wind to a melting ice sculpture; it breaks apart sooner.
4. The Survivors: Even at high temperatures, if there is disorder, the Bose Glass survives. It's the "zombie" phase that refuses to become a superfluid or a perfect crystal, existing in a messy, disordered state.

5. Why Does This Matter?

This isn't just about abstract math.

• Real-World Labs: Scientists are building these systems right now using lasers and atoms. They need to know exactly how hot they can get before their experiments fail.
• Future Tech: Understanding how matter behaves when it's both hot and messy helps us design better materials, perhaps for superconductors (materials that conduct electricity with zero resistance) or quantum computers.

Summary in a Nutshell

The paper is a guidebook for a chaotic dance floor. It tells us:

• If you heat up a perfectly ordered crowd, they eventually lose their formation and become a chaotic mob.
• If you add bumps and noise to the floor, the crowd gets stuck in random pockets (Bose Glass).
• The "checkerboard" formation breaks first, while the "one-per-spot" formation lasts a bit longer.
• But no matter how hot it gets, if the floor is bumpy enough, the crowd will never become a smooth, flowing river; they will remain stuck in a messy, disordered state.

The authors created a mathematical toolkit that can predict exactly when these transitions happen, helping scientists engineer the perfect conditions for future quantum technologies.

Technical Summary

Here is a detailed technical summary of the paper "Finite temperature phase diagram of the extended Bose-Hubbard model in the presence of disorder" by Madhumita Kabiraj and Raka Dasgupta.

1. Problem Statement

The Bose-Hubbard Model (BHM) is a cornerstone of condensed matter physics, describing the quantum phase transition between Mott Insulator (MI) and Superfluid (SF) phases. While extensively studied at zero temperature, realistic ultracold atom experiments operate at finite, non-zero temperatures where thermal fluctuations compete with quantum fluctuations.

Furthermore, the Extended Bose-Hubbard Model (EBHM) introduces long-range interactions (nearest-neighbor and beyond), leading to additional phases like Charge Density Waves (CDW) and Supersolids. The presence of disorder (e.g., impurities or speckle potentials) introduces the Bose Glass (BG) phase.

The Gap: There is a lack of comprehensive theoretical frameworks that simultaneously treat:

1. Finite temperature effects.
2. Long-range interactions (up to next-nearest neighbors).
3. Disorder (specifically uniform disorder).
4. The competition between these factors in determining the stability of insulating lobes (MI, CDW, BG).

2. Methodology

The authors employ a Mean-Field Perturbative Approach to construct phase diagrams.

• Hamiltonian: They utilize the Extended Bose-Hubbard Hamiltonian including:
  • Hopping term ($t$) between nearest neighbors.
  • On-site interaction ($U$).
  • Nearest-neighbor interaction ($V$).
  • Next-nearest-neighbor interaction ($V'$) in extended cases.
  • Disorder potential ($\epsilon_i$) modeled as a uniform distribution in the interval $[-\Delta/2, \Delta/2]$.
• Experimental Realization: The model is motivated by Rydberg-dressed atoms in optical lattices. By tuning the Rydberg excitation level and lattice spacing, the system can be engineered to exhibit either only nearest-neighbor (NN) interactions or both NN and next-nearest-neighbor (NNN) interactions.
• Theoretical Framework:
  • Decoupling: The hopping term is decoupled using a mean-field approximation ($b^\dagger_i b_j \to \langle b^\dagger_i \rangle b_j + \dots$), introducing a superfluid order parameter $\psi$.
  • Sublattice Decomposition: The lattice is divided into sublattices (A, B for NN; A, B, C, D for NNN) to handle the symmetry breaking of CDW phases.
  • Thermal and Disorder Averaging:
    • Thermal: The partition function is calculated for the unperturbed system ($t=0$), and thermal averages are taken over the energy states.
    • Disorder: A disorder average is performed over the uniform distribution of the chemical potential shift ($\tilde{\mu} = \mu + \epsilon$).
  • Phase Boundaries: The phase boundaries are determined by solving coupled equations for the order parameters. The transition from insulating to fluid phases is identified by the vanishing of the order parameter ($\psi \to 0$) and the behavior of compressibility ($\kappa$) and Inverse Participation Ratio (IPR).
  • Validation: The melting of phases is cross-verified using the IPR (calculated via exact diagonalization for small systems) to distinguish between localized (insulating) and delocalized (fluid) states.

3. Key Contributions

1. Unified Framework: The paper presents a versatile mathematical framework capable of treating long-range interactions, disorder, and finite temperature simultaneously.
2. Finite-Temperature Phase Diagrams: It maps out the phase diagrams for both pure and disordered systems, explicitly showing how insulating lobes "melt" into a Normal Fluid (NF) as temperature increases.
3. Disorder vs. Interaction Competition: It quantifies how disorder strength ($\Delta$) relative to interaction strengths ($V, V'$) dictates the existence of specific phases (e.g., CDW phases vanish if $\Delta > V$).
4. Classification of Insulators: It provides a clear classification of insulating phases (MI, CDW, BG) based on three criteria:
   • Superfluid order parameter ($\psi$).
   • Compressibility ($\kappa$).
   • Inverse Participation Ratio (IPR).

4. Key Results

A. Pure Systems (No Disorder)

• Melting of Lobes: At non-zero temperatures, the sharp boundaries of MI and CDW lobes soften. A "Normal Fluid" (NF) region with finite compressibility emerges between the insulating lobes and the superfluid phase.
• Transition Temperatures:
  • CDW Lobes: Melt at lower temperatures ($K_B T^* \approx 0.08V$).
  • Mott Lobes: Melt at higher temperatures ($K_B T^* \approx 0.1U$).
  • The transition temperature depends on the specific interaction strength defining the lobe width.
• NNN Interactions: When interactions extend to next-nearest neighbors, a new type of CDW phase (CDW 2) emerges with fractional densities (e.g., $1/4, 3/4$). These phases have even lower melting temperatures proportional to the weaker NNN interaction strength ($V'$).

B. Disordered Systems

• Emergence of Bose Glass (BG): Disorder creates a compressible insulating phase (BG) between the incompressible MI and CDW lobes.
• Disorder Thresholds:
  • If disorder $\Delta$ is less than the interaction strength ($V$ or $V'$), distinct CDW lobes survive, separated by BG regions.
  • If $\Delta \geq V$, the CDW lobes are completely suppressed, leaving only MI and BG.
  • If $\Delta \geq U$, even the MI lobes vanish, leaving only BG.
• High-Temperature Behavior:
  • As temperature increases, MI and CDW lobes melt into the Normal Fluid.
  • Crucial Finding: The Bose Glass phase persists at high temperatures even when MI and CDW have melted. This is a signature of disorder-induced localization that thermal fluctuations cannot easily destroy.
• Transition Temperature Dependence: The melting temperature ($T^*$) for all insulating phases decreases as the disorder strength ($\Delta$) increases.

C. Phase Classification Summary

Phase	Order Parameter ($\psi$)	Compressibility ($\kappa$)	IPR	Localization
MI / CDW	Zero	Zero	High	Localized
Bose Glass	Zero	Finite	High	Localized
Normal Fluid	Zero	Finite	Low/Medium	Delocalized
Superfluid	Non-Zero	Finite	Low	Delocalized

5. Significance

• Experimental Relevance: The results are directly applicable to current experiments with ultracold Rydberg atoms in optical lattices, where finite temperatures and tunable disorder are inevitable. The paper provides guidelines on how to tune lattice spacing and interaction strengths to observe specific phases before thermal melting occurs.
• Theoretical Insight: It clarifies the hierarchy of stability among different insulating phases. The finding that CDW phases are more fragile to thermal fluctuations than MI phases, and that disorder can stabilize a glassy phase at high temperatures, offers new insights into the interplay of quantum and thermal fluctuations.
• Scalability: The proposed framework is generalizable. It can be extended to study longer-range interactions (beyond NNN) and different disorder distributions (e.g., Gaussian, speckle), making it a robust tool for future studies of complex quantum many-body systems.

In conclusion, the paper establishes that while thermal fluctuations eventually destroy all ordered insulating phases (MI, CDW) turning them into a normal fluid, disorder can sustain a localized, compressible Bose Glass phase even at elevated temperatures, fundamentally altering the high-temperature phase diagram of the extended Bose-Hubbard model.