Gaussian fermionic embezzlement of entanglement

This paper demonstrates that Gaussian operations are sufficient to embezzlement arbitrary Gaussian entangled states from ground states of non-interacting critical fermions, establishing embezzlement as a generic property of fermionic Gaussian states and bridging finite-size systems with abstract von Neumann algebra classifications through novel distance bounds.

The Gist

Here is an explanation of the paper "Gaussian fermionic embezzlement of entanglement" using simple language and creative analogies.

The Big Idea: The Ultimate Quantum "Magic Trick"

Imagine you have a massive, magical bank vault (the embezzling state) that contains an infinite amount of gold. You want to steal a tiny, specific amount of gold (a specific entangled quantum state) to give to a friend.

The catch? You must do this without the vault noticing. If the vault's total weight changes even slightly, the alarm goes off.

In the quantum world, this is called Embezzlement of Entanglement. It's a process where two people (Alice and Bob) can extract any desired quantum connection from a shared resource state using only local tools, while leaving the original resource state looking almost exactly the same as before.

For a long time, scientists knew this was possible in theory, but only if you had a "perfect" system with infinite size. This paper asks: Can we do this with real, finite-sized systems, and can we do it using only "simple" tools?

The Cast of Characters

1. The Resource (The Vault): A system of fermions (a type of particle like an electron). Specifically, the authors look at "Gaussian" states. Think of these as the "standard model" of quantum systems—predictable, well-behaved, and described by simple statistics (like a bell curve).
2. The Tools (Gaussian Operations): Usually, to manipulate quantum particles, you need complex, messy tools. But the authors ask: Can we do the trick using only "Gaussian tools"? These are like using a standard screwdriver instead of a laser cutter. They are simpler and easier to build in a lab.
3. The Target (The Entangled State): The specific quantum connection Alice and Bob want to create.

The Main Discovery: The "Dense Spectrum" Key

The authors discovered a specific "key" that unlocks this magic trick. They call it having a "dense spectrum."

The Analogy: The Piano vs. The Slide

• Imagine a standard piano. It has distinct keys (notes). If you want to play a note that falls between two keys, you can't do it perfectly. This is like a system with a "sparse" spectrum.
• Now, imagine a violin or a slide whistle. You can slide your finger to produce any pitch, no matter how small the difference. This is a "dense" spectrum.

The paper proves that if your quantum "vault" (the fermionic system) has a spectrum that is dense enough (like the violin), you can use simple Gaussian tools to extract any Gaussian entangled state.

The Result:
If the system is large enough and "dense" enough (like the ground state of a critical fermionic chain), you can perform this embezzlement with an error so small it's practically zero.

Why This Matters: From Theory to Reality

1. Bridging the Gap
Previously, embezzlement was a concept reserved for abstract math and infinite systems (the "thermodynamic limit"). This paper bridges that gap. It shows that even in a finite system (like a chain of atoms you could actually build in a lab), this property exists. It's not just a mathematical fantasy; it's a physical reality for certain materials.

2. The "Generic" Nature
The authors show that this isn't a rare fluke. It's a generic property. If you take a random, highly entangled Gaussian state, it's very likely to be a good "embezzler." It's like saying that if you have a large enough library, you can almost certainly find a book that contains the exact sentence you need, without having to rearrange the whole library.

3. The "Free Lunch" (Almost)
The paper addresses a curious side effect: Number Conservation.
Usually, in these systems, the total number of particles is conserved. However, embezzlement seems to break this rule locally.

• The Analogy: Imagine you have a huge bucket of water (the vault). You scoop out a cup of water (the entangled state) and give it to a friend. The bucket looks exactly the same as before. But, to make the math work, the average level of water in the bucket might shift slightly, even though the total amount hasn't changed. The paper explains that because the vault is so huge and the distribution of particles is so broad, you can shift the "average" without anyone noticing the change in the total count.

The "Secret Sauce": A New Mathematical Tool

To prove this, the authors had to invent a new way to measure the distance between two quantum states.

• Old Way: They tried to measure the difference between the "covariance matrices" (the blueprints of the system). But this method failed; it was like trying to measure the distance between two cities by only looking at their zip codes. It wasn't precise enough.
• New Way: They developed a new "ruler" (a novel mathematical bound) that looks deeper into the structure of the states. This new ruler showed that even if the blueprints look slightly different, the actual quantum states are incredibly close if the spectrum is dense enough.

Summary in a Nutshell

This paper proves that nature is generous. If you have a large, critical system of fermions (like electrons in a special material), it naturally contains the ability to generate any desired quantum connection. You don't need complex, impossible machinery; simple, standard quantum tools are enough to "embezzle" this entanglement, leaving the original system looking untouched.

It turns a theoretical concept from the realm of infinite mathematics into a practical, achievable goal for finite quantum systems, bringing us one step closer to understanding how quantum entanglement works in the real world.

Technical Summary

Here is a detailed technical summary of the paper "Gaussian fermionic embezzlement of entanglement" by Kera et al.

1. Problem Statement

The paper addresses the phenomenon of entanglement embezzlement, a protocol where two agents (Alice and Bob) acting on disjoint subsystems can extract an arbitrary entangled state $|\Psi\rangle$ from a shared "embezzling" resource state $|\Omega\rangle$ using only local unitary operations, while perturbing the resource state arbitrarily little.

While the existence of such states is well-established in the thermodynamic limit (infinite degrees of freedom) via the classification of von Neumann algebras (specifically Type III factors), their behavior in finite-size systems remains less understood. Specifically, the authors aim to resolve three open questions for fermionic Gaussian states (quasi-free states described by quadratic Hamiltonians):

1. Can the embezzlement operations be restricted to Gaussian unitaries (generated by quadratic Hamiltonians) if the target state is also Gaussian?
2. How does the achievable error $\varepsilon$ scale with the system size $n$?
3. How generic is the embezzling property among Gaussian states?

2. Methodology and Formalism

The authors employ the Gaussian formalism for fermionic systems, where states are fully characterized by their covariance matrices (two-point correlation functions).

• State Representation: A passive Gaussian state $\rho_G$ is defined by a covariance matrix $G$ ($0 \le G \le 1$) such that expectation values of creation/annihilation operators are determined by determinants of submatrices of$G$.
• Distance Metrics: A central technical challenge is relating the trace distance between quantum states to the distance between their covariance matrices. The standard bound $\text{dist}(\rho_F, \rho_G) \le 2 \text{dist}(F, G)$ (where $F, G$ are covariances) is insufficient for embezzlement because it implies that embezzlement is impossible for certain states (e.g., Fock vacuum vs. fully occupied state).
• Novel Bound: To overcome this, the authors derive a new, tighter bound relating the trace distance of Gaussian states to a specific functional of their covariances, denoted as $\eta(A, B)$:
  $$\eta(A, B) := \| \sqrt{1-A}\sqrt{B} - \sqrt{A}\sqrt{1-B} \|_2$$
  They prove:
  $$1 - e^{-\eta(A,B)^2/2} \le \text{dist}(\rho_A, \rho_B) \le \sqrt{2} \eta(A, B)$$
  Crucially, this functional $\eta$ satisfies the additivity property $\eta(A \oplus C, B \oplus C) = \eta(A, B)$, which is essential for analyzing the embezzlement process where a resource state $K$ is combined with a target state.

3. Key Contributions and Main Results

A. The Main Theorem (Theorem 1)

The authors prove that a Gaussian state with a covariance matrix $K$ having an $\varepsilon$-dense spectrum can act as an approximate embezzling state for any $d$-dimensional Gaussian states $F$ and $G$ using only Gaussian local unitaries.

• Definition: A matrix $K$ has an $\varepsilon$-dense spectrum if for every $x \in [0, 1]$, there exists an eigenvalue $\lambda_j(K)$ such that $|\lambda_j(K) - x| < \varepsilon$.
• Result: The error $\kappa(F, G|K)$ in embezzling $G$ from $K$ (using $F$ as a catalyst) is bounded by:
  $$\kappa(F, G|K) \le 11 d \varepsilon^{1/4}$$
  This demonstrates that as the spectrum becomes denser (i.e., $\varepsilon \to 0$), the embezzlement error vanishes.

B. Scaling and Finite-Size Analysis

The paper analyzes how the error scales with system size $n$:

• Optimal Scaling: If a system has $n$ modes, the best possible spectral density is $\varepsilon \sim 1/n$. This yields an error scaling of $\sim n^{-1/4}$.
• Critical Systems: For critical, non-interacting fermionic systems (e.g., XX spin chain, transverse-field Ising model), the entanglement entropy scales logarithmically with system size ($S \sim \log n$). This implies the spectral density scales as $\varepsilon \sim 1/\log n$. Consequently, the embezzlement error decays very slowly: $\kappa \sim (\log n)^{-1/4}$.
• Generality: The authors show that states with $\varepsilon$-dense spectra are generic among pure entangled Gaussian states.

C. Extension to General States

• From Gaussian to General: The paper proves that if a state can embezzle arbitrary Gaussian states using Gaussian operations, it can also embezzle arbitrary pure entangled states (including non-Gaussian ones) if general (non-Gaussian) local unitaries are allowed. This relies on the fact that single-mode Gaussian states can encode arbitrary qubit states.
• Number Conservation: The authors address the apparent paradox that passive Gaussian unitaries conserve the total fermion number, yet embezzlement can alter local particle number distributions. They explain that the embezzling resource must have a broad particle number distribution (due to the dense spectrum), allowing the mean particle number to shift locally without significantly altering the global probability distribution (in total variation distance).

4. Proof Strategy

The proof of the main theorem involves a reductionist approach:

1. Reduction to Covariances: Using the novel $\eta$-bound, the problem of minimizing the trace distance between states is reduced to minimizing the distance between covariance matrices.
2. Reduction to Commuting Case: The problem is further reduced to a classical problem involving the eigenvalues of the covariance matrices.
3. Spectral Analysis: By utilizing the $\varepsilon$-dense property, the authors construct a unitary transformation that aligns the eigenvalues of the combined system $(K \oplus F)$ to match $(K \oplus G)$ as closely as possible. They utilize Wielandt's theorem regarding the distance between unitary orbits of Hermitian matrices to bound the required unitary rotation.
4. Optimization: The free parameter $\delta$ (used to truncate eigenvalues near 0 and 1) is optimized to balance the truncation error and the unitary rotation error, leading to the final $O(\varepsilon^{1/4})$ bound.

5. Significance and Impact

• Bridging Finite and Infinite Regimes: The work provides a rigorous finite-size explanation for the abstract operator-algebraic results stating that critical fermionic systems are universal embezzlers. It quantifies the "approximate" nature of this property in finite systems.
• Operational Constraints: It establishes that Gaussian operations are sufficient for Gaussian embezzlement, which is physically relevant for systems governed by quadratic Hamiltonians (e.g., superconductors, cold atoms).
• New Mathematical Tools: The derived bound relating the trace distance of Gaussian states to the $\eta$-functional of their covariances is a significant technical contribution, potentially applicable to other problems in quantum information theory involving free fermions.
• Physical Examples: The results apply to ground states of critical 1D fermionic models and chiral edge modes in 2D systems (like $p_x + i p_y$ superconductors), confirming their utility as resources for quantum information tasks in the thermodynamic limit.

In summary, the paper demonstrates that Gaussian fermionic embezzlement is a generic property of critical systems, provided the operations are allowed to be Gaussian, and provides precise quantitative bounds on the efficiency of this process in finite-size regimes.