Full Quantum Work Statistics for Non-Homogeneous Many-Body Systems

This paper establishes a first-principles framework using thermal time-dependent density functional theory to calculate full quantum work statistics and dissipated-work moments in interacting many-body systems, demonstrating its predictive power in analyzing the Mott-to-band-insulator crossover within the Hubbard model.

The Gist

The Big Picture: Measuring the "Cost" of Shaking a Quantum System

Imagine you have a very complex, crowded dance floor (a quantum system) where everyone is holding hands and moving in sync. This is a "many-body system." Now, imagine you suddenly push the wall of the room or change the music tempo (an external force). The dancers will stumble, bump into each other, and eventually settle into a new rhythm.

The energy lost during this stumble—the "friction" of the dance floor—is called dissipated work. In the quantum world, this isn't just a smooth slide; it's a chaotic, jittery event full of random fluctuations.

This paper presents a new, high-precision map (a mathematical framework) to predict exactly how much energy is lost and how chaotic that loss will be, without needing to simulate every single dancer individually.

The Problem: The "Black Box" of Quantum Chaos

For a long time, scientists had two ways to study these systems:

1. The Phenomenological Way: They guessed how the system would react based on general rules, like saying, "It usually gets hot when you push it." This is like guessing the weather by looking at the sky without a thermometer. It's useful but not very accurate.
2. The Exact Way: They tried to calculate the movement of every single particle. For a system with billions of particles, this is like trying to count every grain of sand on a beach while a hurricane is blowing. It's computationally impossible.

The authors wanted a "Goldilocks" solution: a method that is accurate enough to see the details but simple enough to actually run on a computer.

The Solution: A "Shadow Puppet" Trick

The authors used a technique called Thermal Time-Dependent Density Functional Theory (thTDDFT).

Think of the real, complex quantum system as a giant, intricate puppet show with thousands of puppets interacting. It's too hard to track every string and joint.

• The Trick: Instead of tracking the real puppets, they create a "shadow puppet" show. This shadow show is much simpler (it's a system of non-interacting particles), but it is mathematically designed to cast the exact same shadow (density) on the wall as the real, complex system.
• The Benefit: By studying the simple shadow, they can figure out exactly what the complex system is doing. They don't need to know the secrets of every single interaction; they just need to know how the "shadow" moves.

The Key Discovery: Splitting the "Friction"

The paper makes a clever distinction between two types of "friction" or energy loss:

1. The "Adiabatic" Part (The Slow Stretch): Imagine slowly stretching a rubber band. Even if you do it perfectly slowly, the band resists because its shape is changing. This is energy loss due to the shape of the system changing, not because of chaos.
2. The "Non-Adiabatic" Part (The Sudden Snap): Imagine snapping that rubber band. The energy loss here comes from the sudden, chaotic jolts and transitions.

The authors developed a way to separate these two. They showed that the "chaotic" part (non-adiabatic) is directly linked to how the system responds to a quick poke (a "relaxation function"). By using their "shadow puppet" method, they can calculate this response function from first principles (basic laws of physics) rather than guessing.

The Test: The "Hubbard Model" Dance Floor

To prove their map works, they tested it on a famous theoretical model called the Hubbard model.

• The Setup: Imagine a line of dancers (electrons) on a grid. They can hop to the next spot, but if two dancers try to stand on the same spot, they get a "shock" (repulsion).
• The Experiment: They applied a "staggered" push (pushing odd-numbered dancers one way and even-numbered dancers the other way).
• The Result: As they changed the strength of the push and the temperature, the system switched between different "states of matter":
  • Mott Insulator: Dancers are stuck in place because they are afraid of bumping into neighbors.
  • Band Insulator: Dancers are stuck because the floor itself is tilted.
  • Bond-Order Insulator: A weird middle ground where dancers pair up in a specific pattern.

The authors found that their method could clearly see the "signatures" of these different phases in the energy loss. For example, right at the boundary where the system switches from one phase to another, the "friction" (energy loss) spiked dramatically. This confirmed that their method can detect subtle changes in the quantum world just by measuring how much energy is wasted.

Why This Matters

This paper doesn't invent a new battery or a new computer chip. Instead, it provides a new tool for measurement.

• Before: Scientists had to guess how quantum systems would behave when pushed, or they had to wait for supercomputers to crash while trying to calculate it.
• Now: They have a reliable, "first-principles" recipe to calculate exactly how much energy is lost and how the system fluctuates, even in complex, crowded quantum systems.

It bridges the gap between the messy reality of interacting particles and the clean, solvable math of "shadow" systems, allowing scientists to predict the thermodynamic cost of quantum processes with high precision.

Technical Summary

Technical Summary: Full Quantum Work Statistics for Non-Homogeneous Many-Body Systems

Problem Statement
The paper addresses the challenge of characterizing the nonequilibrium thermodynamics of interacting quantum many-body systems, specifically focusing on the statistics of work dissipated during finite-time driving protocols. While generalized linear-response theory (GLRT) has established that the full distribution of dissipated work can be expressed via a relaxation function, the practical application of this framework has been limited. In the absence of microscopic data, the relaxation function is often modeled phenomenologically, sacrificing predictive accuracy. Furthermore, existing approaches struggle to provide a first-principles route to reconstruct this function for non-homogeneous, correlated systems where quantum coherence and many-body correlations play a critical role. The authors aim to bridge the gap between thermal density functional theory (thDFT) and nonequilibrium work statistics to provide a microscopic, predictive framework for these systems.

Methodology
The authors develop a formalism based on Linear-Response Thermal Time-Dependent Density Functional Theory (LR-thTDDFT). The core of the approach involves deriving the relaxation function, $\psi(t)$, directly from microscopic first principles rather than phenomenological modeling.

1. Theoretical Framework:

   • The work statistics are analyzed within the GLRT framework, where all cumulants of the dissipated-work distribution are bilinear functionals of the driving protocol's time derivative, weighted by the Fourier transform of the relaxation function, $\tilde{\psi}(\omega)$.
   • The relaxation function is decomposed into adiabatic ($\psi_{ad}$) and non-adiabatic ($\psi_{na}$) components. The adiabatic part accounts for irreversibility due to the time deformation of the energy spectrum (independent of the driving rate), while the non-adiabatic part captures genuine dissipative transitions between energy levels.
   • Using the Runge-Gross and van Leeuwen theorems extended to finite temperatures, the authors establish a one-to-one correspondence between the time-dependent external potential and the system's density.
   • The non-adiabatic component of the relaxation function is explicitly linked to the imaginary part of the density-density response function, $\chi_{ij}^\beta(\omega)$, of the interacting system.
   • The interacting response function is obtained from the non-interacting Kohn-Sham (KS) response function via the Gross-Kohn formalism, which incorporates the finite-temperature Hartree-exchange-correlation (Hxc) kernel, $f_{Hxc}^\beta(\omega)$.

2. Approximations:

   • To make the problem tractable, the authors employ the Thermal Adiabatic Local Density Approximation (thALDA). In this approximation, the Hxc kernel is instantaneous in time (frequency-independent) and local in space.
   • This choice neglects memory effects and excitonic features beyond single particle-hole excitations but retains the simplicity and computational efficiency of local functionals.

3. Model System:

   • The framework is applied to the one-dimensional Hubbard model subject to a staggered external potential. This system is chosen for its rich phase diagram, including Mott-insulating (MI), band-insulating (BI), and bond-order insulating (BOI) phases.
   • The study analyzes the system across the Mott-to-band-insulator crossover, varying the interaction strength ($U$) and the staggered potential amplitude ($v_0$).

Key Contributions

• First-Principles Derivation: The paper establishes a direct route to calculate the relaxation function from microscopic densities and KS potentials, eliminating the need for phenomenological modeling of the dissipation kernel.
• Adiabatic/Non-Adiabatic Decomposition: The authors provide a clear physical interpretation of the relaxation function by separating it into adiabatic and non-adiabatic contributions, linking the former to static susceptibility and the latter to dynamical response.
• Microscopic Work Statistics: The work demonstrates how to compute the full quantum work distribution (specifically its first few cumulants) for interacting many-body systems using thTDDFT, bridging the gap between equilibrium DFT and nonequilibrium thermodynamics.
• Benchmarking: The approach is benchmarked against exact diagonalization for the two-site Hubbard model (Hubbard dimer), showing good qualitative and quantitative agreement even in strongly correlated regimes.

Results

• Phase Signatures: In the Hubbard model, the first moment of the dissipated work (irreversible entropy production) exhibits a pronounced peak that traces the crossover between the MI and BI regimes, providing a thermodynamic signature of the BOI region. This enhancement is attributed to the increased sensitivity of thermal densities to control parameter variations near competing phases.
• Cumulant Behavior: The first three cumulants of the dissipated-work distribution were analyzed as functions of the quench duration ($\tau$).
  • The third cumulant vanishes in the adiabatic limit ($\tau \to \infty$), consistent with the distribution approaching a Gaussian form where only the first two moments remain.
  • The first and second cumulants retain signatures of the phase transition (BOI to BI) even in the adiabatic regime, driven by the adiabatic contribution to the irreversible work.
  • The system in the band-insulating regime attains adiabaticity at faster quench rates compared to the Mott or BOI phases.
• Thermodynamic Uncertainty Relation (TUR): The study confirms that the TUR lower bound ($F_W \geq 2/\beta$) is not saturated in the presence of quantum fluctuations, even in the linear-response regime, distinguishing quantum behavior from classical expectations.
• Accuracy: The LR-thTDDFT results for the Hubbard dimer show mean relative errors of approximately $1.36 \times 10^{-2}$, $2.76 \times 10^{-2}$, and $1.76 \times 10^{-2}$ for the first, second, and third cumulants, respectively, compared to exact diagonalization.

Significance and Claims
The paper claims to provide a microscopic and transferable framework for quantum thermodynamics in correlated systems. By connecting nonequilibrium work statistics directly to thermal densities and KS potentials, the approach offers a predictive tool that avoids phenomenological assumptions. The authors emphasize that while the current implementation uses the thALDA (which has known limitations in strongly correlated regimes dominated by nonlocal interactions), the formalism itself is exact within linear response and allows for systematic improvements through more refined, nonlocal, or frequency-dependent exchange-correlation kernels.

The work highlights the utility of DFT-based methods for investigating nonequilibrium thermodynamics in large systems (e.g., ultracold atomic gases, low-dimensional materials) where computationally demanding many-body techniques (like DMRG or QMC) may be prohibitive. The authors modestly note that while the current study captures phase signatures and general trends, it does not attempt to extract universal scaling laws (such as Kibble-Zurek scaling) near critical points, which would require more advanced kernels and dedicated analysis. The primary contribution is the establishment of the theoretical link and the demonstration of its feasibility for extracting thermodynamic irreversibility from microscopic models.